Logging


OpenShift Container Platform 4.15

Configuring and using logging in OpenShift Container Platform

Red Hat OpenShift Documentation Team

Abstract

Use logging to collect, visualize, forward, and store log data to troubleshoot issues, identify performance bottlenecks, and detect security threats in OpenShift Container Platform.

Chapter 1. Release notes

1.1. Logging 5.9

Logging is provided as an installable component, with a distinct release cycle from the core OpenShift Container Platform. The Red Hat OpenShift Container Platform Life Cycle Policy outlines release compatibility.

Note

The stable channel only provides updates to the most recent release of logging. To continue receiving updates for prior releases, you must change your subscription channel to stable-x.y, where x.y represents the major and minor version of logging you have installed. For example, stable-5.7.

1.1.1. Logging 5.9.10

This release includes RHSA-2024:10990.

1.1.1.1. Bug Fixes
  • Before this update, any namespace containing openshift or kube was treated as an infrastructure namespace. With this update, only the following namespaces are treated as infrastructure namespaces: default, kube, openshift, and namespaces that begin with openshift- or kube-. (LOG-6044)
  • Before this update, Loki attempted to detect the level of log messages, which caused confusion when the collector also detected log levels and produced different results. With this update, automatic log level detection in Loki is disabled. (LOG-6321)
  • Before this update, when the ClusterLogForwarder custom resource defined tls.insecureSkipVerify: true in combination with type: http and an HTTP URL, the certificate validation was not skipped. This misconfiguration caused the collector to fail because it attempted to validate certificates despite the setting. With this update, when tls.insecureSkipVerify: true is set, the URL is checked for the HTTPS. An HTTP URL will cause a misconfiguration error. (LOG-6376)
  • Before this update, when any infrastructure namespaces were specified in the application inputs in the ClusterLogForwarder custom resource, logs were generated with the incorrect log_type: application tags. With this update, when any infrastructure namespaces are specified in the application inputs, logs are generated with the correct log_type: infrastructure tags. (LOG-6377)

    Important

    When updating to Logging for Red Hat OpenShift 5.9.10, if you previously added any infrastructure namespaces in the application inputs in the ClusterLogForwarder custom resource, you must add the permissions for collecting logs from infrastructure namespaces. For more details, see "Setting up log collection".

1.1.1.2. CVEs

1.1.2. Logging 5.9.9

This release includes RHBA-2024:10049.

1.1.2.1. Bug fixes
  • Before this update, upgrades to version 6.0 failed with errors if a Log File Metric Exporter instance was present. This update fixes the issue, enabling upgrades to proceed smoothly without errors. (LOG-6201)
  • Before this update, Loki did not correctly load some configurations, which caused issues when using Alibaba Cloud or IBM Cloud object storage. This update fixes the configuration-loading code in Loki, resolving the issue. (LOG-6293)
1.1.2.2. CVEs

1.1.3. Logging 5.9.8

This release includes OpenShift Logging Bug Fix Release 5.9.8.

1.1.3.1. Bug fixes
  • Before this update, the Loki Operator failed to add the default namespace label to all AlertingRule resources, which caused the User-Workload-Monitoring Alertmanager to skip routing these alerts. This update adds the rule namespace as a label to all alerting and recording rules, resolving the issue and restoring proper alert routing in Alertmanager. (LOG-6181)
  • Before this update, the LokiStack ruler component view did not initialize properly, causing an invalid field error when the ruler component was disabled. This update ensures that the component view initializes with an empty value, resolving the issue. (LOG-6183)
  • Before this update, an LF character in the vector.toml file under the ES authentication configuration caused the collector pods to crash. This update removes the newline characters from the username and password fields, resolving the issue. (LOG-6206)
  • Before this update, it was possible to set the .containerLimit.maxRecordsPerSecond parameter in the ClusterLogForwarder custom resource to 0, which could lead to an exception during Vector’s startup. With this update, the configuration is validated before being applied, and any invalid values (less than or equal to zero) are rejected. (LOG-6214)
1.1.3.2. CVEs

1.1.4. Logging 5.9.7

This release includes OpenShift Logging Bug Fix Release 5.9.7.

1.1.4.1. Bug fixes
  • Before this update, the clusterlogforwarder.spec.outputs.http.timeout parameter was not applied to the Fluentd configuration when Fluentd was used as the collector type, causing HTTP timeouts to be misconfigured. With this update, the clusterlogforwarder.spec.outputs.http.timeout parameter is now correctly applied, ensuring Fluentd honors the specified timeout and handles HTTP connections according to the user’s configuration. (LOG-6125)
  • Before this update, the TLS section was added without verifying the broker URL schema, resulting in SSL connection errors if the URLs did not start with tls. With this update, the TLS section is now added only if the broker URLs start with tls, preventing SSL connection errors. (LOG-6041)
1.1.4.2. CVEs
Note

For detailed information on Red Hat security ratings, review Severity ratings.

1.1.5. Logging 5.9.6

This release includes OpenShift Logging Bug Fix Release 5.9.6.

1.1.5.1. Bug fixes
  • Before this update, the collector deployment ignored secret changes, causing receivers to reject logs. With this update, the system rolls out a new pod when there is a change in the secret value, ensuring that the collector reloads the updated secrets. (LOG-5525)
  • Before this update, the Vector could not correctly parse field values that included a single dollar sign ($). With this update, field values with a single dollar sign are automatically changed to two dollar signs ($$), ensuring proper parsing by the Vector. (LOG-5602)
  • Before this update, the drop filter could not handle non-string values (e.g., .responseStatus.code: 403). With this update, the drop filter now works properly with these values. (LOG-5815)
  • Before this update, the collector used the default settings to collect audit logs, without handling the backload from output receivers. With this update, the process for collecting audit logs has been improved to better manage file handling and log reading efficiency. (LOG-5866)
  • Before this update, the must-gather tool failed on clusters with non-AMD64 architectures such as Azure Resource Manager (ARM) or PowerPC. With this update, the tool now detects the cluster architecture at runtime and uses architecture-independent paths and dependencies. The detection allows must-gather to run smoothly on platforms like ARM and PowerPC. (LOG-5997)
  • Before this update, the log level was set using a mix of structured and unstructured keywords that were unclear. With this update, the log level follows a clear, documented order, starting with structured keywords. (LOG-6016)
  • Before this update, multiple unnamed pipelines writing to the default output in the ClusterLogForwarder caused a validation error due to duplicate auto-generated names. With this update, the pipeline names are now generated without duplicates. (LOG-6033)
  • Before this update, the collector pods did not have the PreferredScheduling annotation. With this update, the PreferredScheduling annotation is added to the collector daemonset. (LOG-6023)
1.1.5.2. CVEs

1.1.6. Logging 5.9.5

This release includes OpenShift Logging Bug Fix Release 5.9.5

1.1.6.1. Bug Fixes
  • Before this update, duplicate conditions in the LokiStack resource status led to invalid metrics from the Loki Operator. With this update, the Operator removes duplicate conditions from the status. (LOG-5855)
  • Before this update, the Loki Operator did not trigger alerts when it dropped log events due to validation failures. With this update, the Loki Operator includes a new alert definition that triggers an alert if Loki drops log events due to validation failures. (LOG-5895)
  • Before this update, the Loki Operator overwrote user annotations on the LokiStack Route resource, causing customizations to drop. With this update, the Loki Operator no longer overwrites Route annotations, fixing the issue. (LOG-5945)
1.1.6.2. CVEs

None.

1.1.7. Logging 5.9.4

This release includes OpenShift Logging Bug Fix Release 5.9.4

1.1.7.1. Bug Fixes
  • Before this update, an incorrectly formatted timeout configuration caused the OCP plugin to crash. With this update, a validation prevents the crash and informs the user about the incorrect configuration. (LOG-5373)
  • Before this update, workloads with labels containing - caused an error in the collector when normalizing log entries. With this update, the configuration change ensures the collector uses the correct syntax. (LOG-5524)
  • Before this update, an issue prevented selecting pods that no longer existed, even if they had generated logs. With this update, this issue has been fixed, allowing selection of such pods. (LOG-5697)
  • Before this update, the Loki Operator would crash if the CredentialRequest specification was registered in an environment without the cloud-credentials-operator. With this update, the CredentialRequest specification only registers in environments that are cloud-credentials-operator enabled. (LOG-5701)
  • Before this update, the Logging Operator watched and processed all config maps across the cluster. With this update, the dashboard controller only watches the config map for the logging dashboard. (LOG-5702)
  • Before this update, the ClusterLogForwarder introduced an extra space in the message payload which did not follow the RFC3164 specification. With this update, the extra space has been removed, fixing the issue. (LOG-5707)
  • Before this update, removing the seeding for grafana-dashboard-cluster-logging as a part of (LOG-5308) broke new greenfield deployments without dashboards. With this update, the Logging Operator seeds the dashboard at the beginning and continues to update it for changes. (LOG-5747)
  • Before this update, LokiStack was missing a route for the Volume API causing the following error: 404 not found. With this update, LokiStack exposes the Volume API, resolving the issue. (LOG-5749)
1.1.7.2. CVEs

CVE-2024-24790

1.1.8. Logging 5.9.3

This release includes OpenShift Logging Bug Fix Release 5.9.3

1.1.8.1. Bug Fixes
  • Before this update, there was a delay in restarting Ingesters when configuring LokiStack, because the Loki Operator sets the write-ahead log replay_memory_ceiling to zero bytes for the 1x.demo size. With this update, the minimum value used for the replay_memory_ceiling has been increased to avoid delays. (LOG-5614)
  • Before this update, monitoring the Vector collector output buffer state was not possible. With this update, monitoring and alerting the Vector collector output buffer size is possible that improves observability capabilities and helps keep the system running optimally. (LOG-5586)
1.1.8.2. CVEs

1.1.9. Logging 5.9.2

This release includes OpenShift Logging Bug Fix Release 5.9.2

1.1.9.1. Bug Fixes
  • Before this update, changes to the Logging Operator caused an error due to an incorrect configuration in the ClusterLogForwarder CR. As a result, upgrades to logging deleted the daemonset collector. With this update, the Logging Operator re-creates collector daemonsets except when a Not authorized to collect error occurs. (LOG-4910)
  • Before this update, the rotated infrastructure log files were sent to the application index in some scenarios due to an incorrect configuration in the Vector log collector. With this update, the Vector log collector configuration avoids collecting any rotated infrastructure log files. (LOG-5156)
  • Before this update, the Logging Operator did not monitor changes to the grafana-dashboard-cluster-logging config map. With this update, the Logging Operator monitors changes in the ConfigMap objects, ensuring the system stays synchronized and responds effectively to config map modifications. (LOG-5308)
  • Before this update, an issue in the metrics collection code of the Logging Operator caused it to report stale telemetry metrics. With this update, the Logging Operator does not report stale telemetry metrics. (LOG-5426)
  • Before this change, the Fluentd out_http plugin ignored the no_proxy environment variable. With this update, the Fluentd patches the HTTP#start method of ruby to honor the no_proxy environment variable. (LOG-5466)
1.1.9.2. CVEs

1.1.10. Logging 5.9.1

This release includes OpenShift Logging Bug Fix Release 5.9.1

1.1.10.1. Enhancements
  • Before this update, the Loki Operator configured Loki to use path-based style access for the Amazon Simple Storage Service (S3), which has been deprecated. With this update, the Loki Operator defaults to virtual-host style without users needing to change their configuration. (LOG-5401)
  • Before this update, the Loki Operator did not validate the Amazon Simple Storage Service (S3) endpoint used in the storage secret. With this update, the validation process ensures the S3 endpoint is a valid S3 URL, and the LokiStack status updates to indicate any invalid URLs. (LOG-5395)
1.1.10.2. Bug Fixes
  • Before this update, a bug in LogQL parsing left out some line filters from the query. With this update, the parsing now includes all the line filters while keeping the original query unchanged. (LOG-5268)
  • Before this update, a prune filter without a defined pruneFilterSpec would cause a segfault. With this update, there is a validation error if a prune filter is without a defined puneFilterSpec. (LOG-5322)
  • Before this update, a drop filter without a defined dropTestsSpec would cause a segfault. With this update, there is a validation error if a prune filter is without a defined puneFilterSpec. (LOG-5323)
  • Before this update, the Loki Operator did not validate the Amazon Simple Storage Service (S3) endpoint URL format used in the storage secret. With this update, the S3 endpoint URL goes through a validation step that reflects on the status of the LokiStack. (LOG-5397)
  • Before this update, poorly formatted timestamp fields in audit log records led to WARN messages in Red Hat OpenShift Logging Operator logs. With this update, a remap transformation ensures that the timestamp field is properly formatted. (LOG-4672)
  • Before this update, the error message thrown while validating a ClusterLogForwarder resource name and namespace did not correspond to the correct error. With this update, the system checks if a ClusterLogForwarder resource with the same name exists in the same namespace. If not, it corresponds to the correct error. (LOG-5062)
  • Before this update, the validation feature for output config required a TLS URL, even for services such as Amazon CloudWatch or Google Cloud Logging where a URL is not needed by design. With this update, the validation logic for services without URLs are improved, and the error message are more informative. (LOG-5307)
  • Before this update, defining an infrastructure input type did not exclude logging workloads from the collection. With this update, the collection excludes logging services to avoid feedback loops. (LOG-5309)
1.1.10.3. CVEs

No CVEs.

1.1.11. Logging 5.9.0

This release includes OpenShift Logging Bug Fix Release 5.9.0

1.1.11.1. Removal notice

The Logging 5.9 release does not contain an updated version of the OpenShift Elasticsearch Operator. Instances of OpenShift Elasticsearch Operator from prior logging releases, remain supported until the EOL of the logging release. As an alternative to using the OpenShift Elasticsearch Operator to manage the default log storage, you can use the Loki Operator. For more information on the Logging lifecycle dates, see Platform Agnostic Operators.

1.1.11.2. Deprecation notice
  • In Logging 5.9, Fluentd, and Kibana are deprecated and are planned to be removed in Logging 6.0, which is expected to be shipped alongside a future release of OpenShift Container Platform. Red Hat will provide critical and above CVE bug fixes and support for these components during the current release lifecycle, but these components will no longer receive feature enhancements. The Vector-based collector provided by the Red Hat OpenShift Logging Operator and LokiStack provided by the Loki Operator are the preferred Operators for log collection and storage. We encourage all users to adopt the Vector and Loki log stack, as this will be the stack that will be enhanced going forward.
  • In Logging 5.9, the Fields option for the Splunk output type was never implemented and is now deprecated. It will be removed in a future release.
1.1.11.3. Enhancements
1.1.11.3.1. Log Collection
  • This enhancement adds the ability to refine the process of log collection by using a workload’s metadata to drop or prune logs based on their content. Additionally, it allows the collection of infrastructure logs, such as journal or container logs, and audit logs, such as kube api or ovn logs, to only collect individual sources. (LOG-2155)
  • This enhancement introduces a new type of remote log receiver, the syslog receiver. You can configure it to expose a port over a network, allowing external systems to send syslog logs using compatible tools such as rsyslog. (LOG-3527)
  • With this update, the ClusterLogForwarder API now supports log forwarding to Azure Monitor Logs, giving users better monitoring abilities. This feature helps users to maintain optimal system performance and streamline the log analysis processes in Azure Monitor, which speeds up issue resolution and improves operational efficiency. (LOG-4605)
  • This enhancement improves collector resource utilization by deploying collectors as a deployment with two replicas. This occurs when the only input source defined in the ClusterLogForwarder custom resource (CR) is a receiver input instead of using a daemon set on all nodes. Additionally, collectors deployed in this manner do not mount the host file system. To use this enhancement, you need to annotate the ClusterLogForwarder CR with the logging.openshift.io/dev-preview-enable-collector-as-deployment annotation. (LOG-4779)
  • This enhancement introduces the capability for custom tenant configuration across all supported outputs, facilitating the organization of log records in a logical manner. However, it does not permit custom tenant configuration for logging managed storage. (LOG-4843)
  • With this update, the ClusterLogForwarder CR that specifies an application input with one or more infrastructure namespaces like default, openshift*, or kube*, now requires a service account with the collect-infrastructure-logs role. (LOG-4943)
  • This enhancement introduces the capability for tuning some output settings, such as compression, retry duration, and maximum payloads, to match the characteristics of the receiver. Additionally, this feature includes a delivery mode to allow administrators to choose between throughput and log durability. For example, the AtLeastOnce option configures minimal disk buffering of collected logs so that the collector can deliver those logs after a restart. (LOG-5026)
  • This enhancement adds three new Prometheus alerts, warning users about the deprecation of Elasticsearch, Fluentd, and Kibana. (LOG-5055)
1.1.11.3.2. Log Storage
  • This enhancement in LokiStack improves support for OTEL by using the new V13 object storage format and enabling automatic stream sharding by default. This also prepares the collector for future enhancements and configurations. (LOG-4538)
  • This enhancement introduces support for short-lived token workload identity federation with Azure and AWS log stores for STS enabled OpenShift Container Platform 4.14 and later clusters. Local storage requires the addition of a CredentialMode: static annotation under spec.storage.secret in the LokiStack CR. (LOG-4540)
  • With this update, the validation of the Azure storage secret is now extended to give early warning for certain error conditions. (LOG-4571)
  • With this update, Loki now adds upstream and downstream support for GCP workload identity federation mechanism. This allows authenticated and authorized access to the corresponding object storage services. (LOG-4754)
1.1.11.4. Bug Fixes
  • Before this update, the logging must-gather could not collect any logs on a FIPS-enabled cluster. With this update, a new oc client is available in cluster-logging-rhel9-operator, and must-gather works properly on FIPS clusters. (LOG-4403)
  • Before this update, the LokiStack ruler pods could not format the IPv6 pod IP in HTTP URLs used for cross-pod communication. This issue caused querying rules and alerts through the Prometheus-compatible API to fail. With this update, the LokiStack ruler pods encapsulate the IPv6 pod IP in square brackets, resolving the problem. Now, querying rules and alerts through the Prometheus-compatible API works just like in IPv4 environments. (LOG-4709)
  • Before this fix, the YAML content from the logging must-gather was exported in a single line, making it unreadable. With this update, the YAML white spaces are preserved, ensuring that the file is properly formatted. (LOG-4792)
  • Before this update, when the ClusterLogForwarder CR was enabled, the Red Hat OpenShift Logging Operator could run into a nil pointer exception when ClusterLogging.Spec.Collection was nil. With this update, the issue is now resolved in the Red Hat OpenShift Logging Operator. (LOG-5006)
  • Before this update, in specific corner cases, replacing the ClusterLogForwarder CR status field caused the resourceVersion to constantly update due to changing timestamps in Status conditions. This condition led to an infinite reconciliation loop. With this update, all status conditions synchronize, so that timestamps remain unchanged if conditions stay the same. (LOG-5007)
  • Before this update, there was an internal buffering behavior to drop_newest to address high memory consumption by the collector resulting in significant log loss. With this update, the behavior reverts to using the collector defaults. (LOG-5123)
  • Before this update, the Loki Operator ServiceMonitor in the openshift-operators-redhat namespace used static token and CA files for authentication, causing errors in the Prometheus Operator in the User Workload Monitoring spec on the ServiceMonitor configuration. With this update, the Loki Operator ServiceMonitor in openshift-operators-redhat namespace now references a service account token secret by a LocalReference object. This approach allows the User Workload Monitoring spec in the Prometheus Operator to handle the Loki Operator ServiceMonitor successfully, enabling Prometheus to scrape the Loki Operator metrics. (LOG-5165)
  • Before this update, the configuration of the Loki Operator ServiceMonitor could match many Kubernetes services, resulting in the Loki Operator metrics being collected multiple times. With this update, the configuration of ServiceMonitor now only matches the dedicated metrics service. (LOG-5212)
1.1.11.5. Known Issues

None.

1.1.11.6. CVEs

1.2. Logging 5.8

Note

Logging is provided as an installable component, with a distinct release cycle from the core OpenShift Container Platform. The Red Hat OpenShift Container Platform Life Cycle Policy outlines release compatibility.

Note

The stable channel only provides updates to the most recent release of logging. To continue receiving updates for prior releases, you must change your subscription channel to stable-x.y, where x.y represents the major and minor version of logging you have installed. For example, stable-5.7.

1.2.1. Logging 5.8.16

This release includes RHBA-2024:10989 and RHBA-2024:143685.

1.2.1.1. Bug fixes
  • Before this update, Loki automatically tried to guess the log level of log messages, which caused confusion because the collector already does this, and Loki and the collector would sometimes come to different results. With this update, the automatic log level discovery in Loki is disabled. LOG-6322.
1.2.1.2. CVEs

1.2.2. Logging 5.8.15

This release includes RHBA-2024:10052 and RHBA-2024:10053.

1.2.2.1. Bug fixes
  • Before this update, Loki did not correctly load some configurations, which caused issues when using Alibaba Cloud or IBM Cloud object storage. This update fixes the configuration-loading code in Loki, resolving the issue. (LOG-6294)
  • Before this update, upgrades to version 6.0 failed with errors if a Log File Metric Exporter instance was present. This update fixes the issue, enabling upgrades to proceed smoothly without errors. (LOG-6328)
1.2.2.2. CVEs

1.2.3. Logging 5.8.14

This release includes OpenShift Logging Bug Fix Release 5.8.14 and OpenShift Logging Bug Fix Release 5.8.14.

1.2.3.1. Bug fixes
  • Before this update, it was possible to set the .containerLimit.maxRecordsPerSecond parameter in the ClusterLogForwarder custom resource to 0, which could lead to an exception during Vector’s startup. With this update, the configuration is validated before being applied, and any invalid values (less than or equal to zero) are rejected. (LOG-4671)
  • Before this update, the Loki Operator did not automatically add the default namespace label to all its alerting rules, which caused Alertmanager instance for user-defined projects to skip routing such alerts. With this update, all alerting and recording rules have the namespace label and Alertmanager now routes these alerts correctly. (LOG-6182)
  • Before this update, the LokiStack ruler component view was not properly initialized, which caused the invalid field error when the ruler component was disabled. With this update, the issue is resolved by the component view being initialized with an empty value. (LOG-6184)
1.2.3.2. CVEs
Note

For detailed information on Red Hat security ratings, review Severity ratings.

1.2.4. Logging 5.8.13

This release includes OpenShift Logging Bug Fix Release 5.8.13 and OpenShift Logging Bug Fix Release 5.8.13.

1.2.4.1. Bug fixes
  • Before this update, the clusterlogforwarder.spec.outputs.http.timeout parameter was not applied to the Fluentd configuration when Fluentd was used as the collector type, causing HTTP timeouts to be misconfigured. With this update, the clusterlogforwarder.spec.outputs.http.timeout parameter is now correctly applied, ensuring that Fluentd honors the specified timeout and handles HTTP connections according to the user’s configuration. (LOG-5210)
  • Before this update, the Elasticsearch Operator did not issue an alert to inform users about the upcoming removal, leaving existing installations unsupported without notice. With this update, the Elasticsearch Operator will trigger a continuous alert on OpenShift Container Platform version 4.16 and later, notifying users of its removal from the catalog in November 2025. (LOG-5966)
  • Before this update, the Red Hat OpenShift Logging Operator was unavailable on OpenShift Container Platform version 4.16 and later, preventing Telco customers from completing their certifications for the upcoming Logging 6.0 release. With this update, the Red Hat OpenShift Logging Operator is now available on OpenShift Container Platform versions 4.16 and 4.17, resolving the issue. (LOG-6103)
  • Before this update, the Elasticsearch Operator was not available in the OpenShift Container Platform versions 4.17 and 4.18, preventing the installation of ServiceMesh, Kiali, and Distributed Tracing. With this update, the Elasticsearch Operator properties have been expanded for OpenShift Container Platform versions 4.17 and 4.18, resolving the issue and allowing ServiceMesh, Kiali, and Distributed Tracing operators to install their stacks. (LOG-6134)
1.2.4.2. CVEs
Note

For detailed information on Red Hat security ratings, review Severity ratings.

1.2.5. Logging 5.8.12

This release includes OpenShift Logging Bug Fix Release 5.8.12 and OpenShift Logging Bug Fix Release 5.8.12.

1.2.5.1. Bug fixes
  • Before this update, the collector used internal buffering with the drop_newest setting to reduce high memory usage, which caused significant log loss. With this update, the collector goes back to its default behavior, where sink<>.buffer is not customized. (LOG-6026)
1.2.5.2. CVEs

1.2.6. Logging 5.8.11

This release includes OpenShift Logging Bug Fix Release 5.8.11 and OpenShift Logging Bug Fix Release 5.8.11.

1.2.6.1. Bug fixes
  • Before this update, the TLS section was added without verifying the broker URL schema, leading to SSL connection errors if the URLs did not start with tls. With this update, the TLS section is added only if broker URLs start with tls, preventing SSL connection errors. (LOG-5139)
  • Before this update, the Loki Operator did not trigger alerts when it dropped log events due to validation failures. With this update, the Loki Operator includes a new alert definition that triggers an alert if Loki drops log events due to validation failures. (LOG-5896)
  • Before this update, the 4.16 GA catalog did not include Elasticsearch Operator 5.8, preventing the installation of products like Service Mesh, Kiali, and Tracing. With this update, Elasticsearch Operator 5.8 is now available on 4.16, resolving the issue and providing support for Elasticsearch storage for these products only. (LOG-5911)
  • Before this update, duplicate conditions in the LokiStack resource status led to invalid metrics from the Loki Operator. With this update, the Operator removes duplicate conditions from the status. (LOG-5857)
  • Before this update, the Loki Operator overwrote user annotations on the LokiStack Route resource, causing customizations to drop. With this update, the Loki Operator no longer overwrites Route annotations, fixing the issue. (LOG-5946)
1.2.6.2. CVEs

1.2.7. Logging 5.8.10

This release includes OpenShift Logging Bug Fix Release 5.8.10 and OpenShift Logging Bug Fix Release 5.8.10.

1.2.7.1. Known issues
  • Before this update, when enabling retention, the Loki Operator produced an invalid configuration. As a result, Loki did not start properly. With this update, Loki pods can set retention. (LOG-5821)
1.2.7.2. Bug fixes
  • Before this update, the ClusterLogForwarder introduced an extra space in the message payload that did not follow the RFC3164 specification. With this update, the extra space has been removed, fixing the issue. (LOG-5647)
1.2.7.3. CVEs

1.2.8. Logging 5.8.9

This release includes OpenShift Logging Bug Fix Release 5.8.9 and OpenShift Logging Bug Fix Release 5.8.9.

1.2.8.1. Bug fixes
  • Before this update, an issue prevented selecting pods that no longer existed, even if they had generated logs. With this update, this issue has been fixed, allowing selection of such pods. (LOG-5698)
  • Before this update, LokiStack was missing a route for the Volume API, which caused the following error: 404 not found. With this update, LokiStack exposes the Volume API, resolving the issue. (LOG-5750)
  • Before this update, the Elasticsearch operator overwrote all service account annotations without considering ownership. As a result, the kube-controller-manager recreated service account secrets because it logged the link to the owning service account. With this update, the Elasticsearch operator merges annotations, resolving the issue. (LOG-5776)
1.2.8.2. CVEs

1.2.9. Logging 5.8.8

This release includes OpenShift Logging Bug Fix Release 5.8.8 and OpenShift Logging Bug Fix Release 5.8.8.

1.2.9.1. Bug fixes
  • Before this update, there was a delay in restarting Ingesters when configuring LokiStack, because the Loki Operator sets the write-ahead log replay_memory_ceiling to zero bytes for the 1x.demo size. With this update, the minimum value used for the replay_memory_ceiling has been increased to avoid delays. (LOG-5615)
1.2.9.2. CVEs

1.2.10. Logging 5.8.7

This release includes OpenShift Logging Bug Fix Release 5.8.7 Security Update and OpenShift Logging Bug Fix Release 5.8.7.

1.2.10.1. Bug fixes
  • Before this update, the elasticsearch-im-<type>-* pods failed if no <type> logs (audit, infrastructure, or application) were collected. With this update, the pods no longer fail when <type> logs are not collected. (LOG-4949)
  • Before this update, the validation feature for output config required an SSL/TLS URL, even for services such as Amazon CloudWatch or Google Cloud Logging where a URL is not needed by design. With this update, the validation logic for services without URLs are improved, and the error message is more informative. (LOG-5467)
  • Before this update, an issue in the metrics collection code of the Logging Operator caused it to report stale telemetry metrics. With this update, the Logging Operator does not report stale telemetry metrics. (LOG-5471)
  • Before this update, changes to the Logging Operator caused an error due to an incorrect configuration in the ClusterLogForwarder CR. As a result, upgrades to logging deleted the daemonset collector. With this update, the Logging Operator re-creates collector daemonsets except when a Not authorized to collect error occurs. (LOG-5514)
1.2.10.2. CVEs

1.2.11. Logging 5.8.6

This release includes OpenShift Logging Bug Fix Release 5.8.6 Security Update and OpenShift Logging Bug Fix Release 5.8.6.

1.2.11.1. Enhancements
  • Before this update, the Loki Operator did not validate the Amazon Simple Storage Service (S3) endpoint used in the storage secret. With this update, the validation process ensures the S3 endpoint is a valid S3 URL, and the LokiStack status updates to indicate any invalid URLs. (LOG-5392)
  • Before this update, the Loki Operator configured Loki to use path-based style access for the Amazon Simple Storage Service (S3), which has been deprecated. With this update, the Loki Operator defaults to virtual-host style without users needing to change their configuration. (LOG-5402)
1.2.11.2. Bug fixes
  • Before this update, the Elastisearch Operator ServiceMonitor in the openshift-operators-redhat namespace used static token and certificate authority (CA) files for authentication, causing errors in the Prometheus Operator in the User Workload Monitoring specification on the ServiceMonitor configuration. With this update, the Elastisearch Operator ServiceMonitor in the openshift-operators-redhat namespace now references a service account token secret by a LocalReference object. This approach allows the User Workload Monitoring specifications in the Prometheus Operator to handle the Elastisearch Operator ServiceMonitor successfully. This enables Prometheus to scrape the Elastisearch Operator metrics. (LOG-5164)
  • Before this update, the Loki Operator did not validate the Amazon Simple Storage Service (S3) endpoint URL format used in the storage secret. With this update, the S3 endpoint URL goes through a validation step that reflects on the status of the LokiStack. (LOG-5398)
1.2.11.3. CVEs

1.2.12. Logging 5.8.5

This release includes OpenShift Logging Bug Fix Release 5.8.5.

1.2.12.1. Bug fixes
  • Before this update, the configuration of the Loki Operator’s ServiceMonitor could match many Kubernetes services, resulting in the Loki Operator’s metrics being collected multiple times. With this update, the configuration of ServiceMonitor now only matches the dedicated metrics service. (LOG-5250)
  • Before this update, the Red Hat build pipeline did not use the existing build details in Loki builds and omitted information such as revision, branch, and version. With this update, the Red Hat build pipeline now adds these details to the Loki builds, fixing the issue. (LOG-5201)
  • Before this update, the Loki Operator checked if the pods were running to decide if the LokiStack was ready. With this update, it also checks if the pods are ready, so that the readiness of the LokiStack reflects the state of its components. (LOG-5171)
  • Before this update, running a query for log metrics caused an error in the histogram. With this update, the histogram toggle function and the chart are disabled and hidden because the histogram doesn’t work with log metrics. (LOG-5044)
  • Before this update, the Loki and Elasticsearch bundle had the wrong maxOpenShiftVersion, resulting in IncompatibleOperatorsInstalled alerts. With this update, including 4.16 as the maxOpenShiftVersion property in the bundle fixes the issue. (LOG-5272)
  • Before this update, the build pipeline did not include linker flags for the build date, causing Loki builds to show empty strings for buildDate and goVersion. With this update, adding the missing linker flags in the build pipeline fixes the issue. (LOG-5274)
  • Before this update, a bug in LogQL parsing left out some line filters from the query. With this update, the parsing now includes all the line filters while keeping the original query unchanged. (LOG-5270)
  • Before this update, the Loki Operator ServiceMonitor in the openshift-operators-redhat namespace used static token and CA files for authentication, causing errors in the Prometheus Operator in the User Workload Monitoring spec on the ServiceMonitor configuration. With this update, the Loki Operator ServiceMonitor in openshift-operators-redhat namespace now references a service account token secret by a LocalReference object. This approach allows the User Workload Monitoring spec in the Prometheus Operator to handle the Loki Operator ServiceMonitor successfully, enabling Prometheus to scrape the Loki Operator metrics. (LOG-5240)
1.2.12.2. CVEs

1.2.13. Logging 5.8.4

This release includes OpenShift Logging Bug Fix Release 5.8.4.

1.2.13.1. Bug fixes
  • Before this update, the developer console’s logs did not account for the current namespace, resulting in query rejection for users without cluster-wide log access. With this update, all supported OCP versions ensure correct namespace inclusion. (LOG-4905)
  • Before this update, the Cluster Logging Operator deployed ClusterRoles supporting LokiStack deployments only when the default log output was LokiStack. With this update, the roles are split into two groups: read and write. The write roles deploys based on the setting of the default log storage, just like all the roles used to do before. The read roles deploys based on whether the logging console plugin is active. (LOG-4987)
  • Before this update, multiple ClusterLogForwarders defining the same input receiver name had their service endlessly reconciled because of changing ownerReferences on one service. With this update, each receiver input will have its own service named with the convention of <CLF.Name>-<input.Name>. (LOG-5009)
  • Before this update, the ClusterLogForwarder did not report errors when forwarding logs to cloudwatch without a secret. With this update, the following error message appears when forwarding logs to cloudwatch without a secret: secret must be provided for cloudwatch output. (LOG-5021)
  • Before this update, the log_forwarder_input_info included application, infrastructure, and audit input metric points. With this update, http is also added as a metric point. (LOG-5043)
1.2.13.2. CVEs

1.2.14. Logging 5.8.3

This release includes Logging Bug Fix 5.8.3 and Logging Security Fix 5.8.3

1.2.14.1. Bug fixes
  • Before this update, when configured to read a custom S3 Certificate Authority the Loki Operator would not automatically update the configuration when the name of the ConfigMap or the contents changed. With this update, the Loki Operator is watching for changes to the ConfigMap and automatically updates the generated configuration. (LOG-4969)
  • Before this update, Loki outputs configured without a valid URL caused the collector pods to crash. With this update, outputs are subject to URL validation, resolving the issue. (LOG-4822)
  • Before this update the Cluster Logging Operator would generate collector configuration fields for outputs that did not specify a secret to use the service account bearer token. With this update, an output does not require authentication, resolving the issue. (LOG-4962)
  • Before this update, the tls.insecureSkipVerify field of an output was not set to a value of true without a secret defined. With this update, a secret is no longer required to set this value. (LOG-4963)
  • Before this update, output configurations allowed the combination of an insecure (HTTP) URL with TLS authentication. With this update, outputs configured for TLS authentication require a secure (HTTPS) URL. (LOG-4893)
1.2.14.2. CVEs

1.2.15. Logging 5.8.2

This release includes OpenShift Logging Bug Fix Release 5.8.2.

1.2.15.1. Bug fixes
  • Before this update, the LokiStack ruler pods would not format the IPv6 pod IP in HTTP URLs used for cross pod communication, causing querying rules and alerts through the Prometheus-compatible API to fail. With this update, the LokiStack ruler pods encapsulate the IPv6 pod IP in square brackets, resolving the issue. (LOG-4890)
  • Before this update, the developer console logs did not account for the current namespace, resulting in query rejection for users without cluster-wide log access. With this update, namespace inclusion has been corrected, resolving the issue. (LOG-4947)
  • Before this update, the logging view plugin of the OpenShift Container Platform web console did not allow for custom node placement and tolerations. With this update, defining custom node placements and tolerations has been added to the logging view plugin of the OpenShift Container Platform web console. (LOG-4912)
1.2.15.2. CVEs

1.2.16. Logging 5.8.1

This release includes OpenShift Logging Bug Fix Release 5.8.1 and OpenShift Logging Bug Fix Release 5.8.1 Kibana.

1.2.16.1. Enhancements
1.2.16.1.1. Log Collection
  • With this update, while configuring Vector as a collector, you can add logic to the Red Hat OpenShift Logging Operator to use a token specified in the secret in place of the token associated with the service account. (LOG-4780)
  • With this update, the BoltDB Shipper Loki dashboards are now renamed to Index dashboards. (LOG-4828)
1.2.16.2. Bug fixes
  • Before this update, the ClusterLogForwarder created empty indices after enabling the parsing of JSON logs, even when the rollover conditions were not met. With this update, the ClusterLogForwarder skips the rollover when the write-index is empty. (LOG-4452)
  • Before this update, the Vector set the default log level incorrectly. With this update, the correct log level is set by improving the enhancement of regular expression, or regexp, for log level detection. (LOG-4480)
  • Before this update, during the process of creating index patterns, the default alias was missing from the initial index in each log output. As a result, Kibana users were unable to create index patterns by using OpenShift Elasticsearch Operator. This update adds the missing aliases to OpenShift Elasticsearch Operator, resolving the issue. Kibana users can now create index patterns that include the {app,infra,audit}-000001 indexes. (LOG-4683)
  • Before this update, Fluentd collector pods were in a CrashLoopBackOff state due to binding of the Prometheus server on IPv6 clusters. With this update, the collectors work properly on IPv6 clusters. (LOG-4706)
  • Before this update, the Red Hat OpenShift Logging Operator would undergo numerous reconciliations whenever there was a change in the ClusterLogForwarder. With this update, the Red Hat OpenShift Logging Operator disregards the status changes in the collector daemonsets that triggered the reconciliations. (LOG-4741)
  • Before this update, the Vector log collector pods were stuck in the CrashLoopBackOff state on IBM Power machines. With this update, the Vector log collector pods start successfully on IBM Power architecture machines. (LOG-4768)
  • Before this update, forwarding with a legacy forwarder to an internal LokiStack would produce SSL certificate errors using Fluentd collector pods. With this update, the log collector service account is used by default for authentication, using the associated token and ca.crt. (LOG-4791)
  • Before this update, forwarding with a legacy forwarder to an internal LokiStack would produce SSL certificate errors using Vector collector pods. With this update, the log collector service account is used by default for authentication and also using the associated token and ca.crt. (LOG-4852)
  • Before this fix, IPv6 addresses would not be parsed correctly after evaluating a host or multiple hosts for placeholders. With this update, IPv6 addresses are correctly parsed. (LOG-4811)
  • Before this update, it was necessary to create a ClusterRoleBinding to collect audit permissions for HTTP receiver inputs. With this update, it is not necessary to create the ClusterRoleBinding because the endpoint already depends upon the cluster certificate authority. (LOG-4815)
  • Before this update, the Loki Operator did not mount a custom CA bundle to the ruler pods. As a result, during the process to evaluate alerting or recording rules, object storage access failed. With this update, the Loki Operator mounts the custom CA bundle to all ruler pods. The ruler pods can download logs from object storage to evaluate alerting or recording rules. (LOG-4836)
  • Before this update, while removing the inputs.receiver section in the ClusterLogForwarder, the HTTP input services and its associated secrets were not deleted. With this update, the HTTP input resources are deleted when not needed. (LOG-4612)
  • Before this update, the ClusterLogForwarder indicated validation errors in the status, but the outputs and the pipeline status did not accurately reflect the specific issues. With this update, the pipeline status displays the validation failure reasons correctly in case of misconfigured outputs, inputs, or filters. (LOG-4821)
  • Before this update, changing a LogQL query that used controls such as time range or severity changed the label matcher operator defining it like a regular expression. With this update, regular expression operators remain unchanged when updating the query. (LOG-4841)
1.2.16.3. CVEs

1.2.17. Logging 5.8.0

This release includes OpenShift Logging Bug Fix Release 5.8.0 and OpenShift Logging Bug Fix Release 5.8.0 Kibana.

1.2.17.1. Deprecation notice

In Logging 5.8, Elasticsearch, Fluentd, and Kibana are deprecated and are planned to be removed in Logging 6.0, which is expected to be shipped alongside a future release of OpenShift Container Platform. Red Hat will provide critical and above CVE bug fixes and support for these components during the current release lifecycle, but these components will no longer receive feature enhancements. The Vector-based collector provided by the Red Hat OpenShift Logging Operator and LokiStack provided by the Loki Operator are the preferred Operators for log collection and storage. We encourage all users to adopt the Vector and Loki log stack, as this will be the stack that will be enhanced going forward.

1.2.17.2. Enhancements
1.2.17.2.1. Log Collection
  • With this update, the LogFileMetricExporter is no longer deployed with the collector by default. You must manually create a LogFileMetricExporter custom resource (CR) to generate metrics from the logs produced by running containers. If you do not create the LogFileMetricExporter CR, you may see a No datapoints found message in the OpenShift Container Platform web console dashboard for Produced Logs. (LOG-3819)
  • With this update, you can deploy multiple, isolated, and RBAC-protected ClusterLogForwarder custom resource (CR) instances in any namespace. This allows independent groups to forward desired logs to any destination while isolating their configuration from other collector deployments. (LOG-1343)

    Important

    In order to support multi-cluster log forwarding in additional namespaces other than the openshift-logging namespace, you must update the Red Hat OpenShift Logging Operator to watch all namespaces. This functionality is supported by default in new Red Hat OpenShift Logging Operator version 5.8 installations.

  • With this update, you can use the flow control or rate limiting mechanism to limit the volume of log data that can be collected or forwarded by dropping excess log records. The input limits prevent poorly-performing containers from overloading the Logging and the output limits put a ceiling on the rate of logs shipped to a given data store. (LOG-884)
  • With this update, you can configure the log collector to look for HTTP connections and receive logs as an HTTP server, also known as a webhook. (LOG-4562)
  • With this update, you can configure audit polices to control which Kubernetes and OpenShift API server events are forwarded by the log collector. (LOG-3982)
1.2.17.2.2. Log Storage
  • With this update, LokiStack administrators can have more fine-grained control over who can access which logs by granting access to logs on a namespace basis. (LOG-3841)
  • With this update, the Loki Operator introduces PodDisruptionBudget configuration on LokiStack deployments to ensure normal operations during OpenShift Container Platform cluster restarts by keeping ingestion and the query path available. (LOG-3839)
  • With this update, the reliability of existing LokiStack installations are seamlessly improved by applying a set of default Affinity and Anti-Affinity policies. (LOG-3840)
  • With this update, you can manage zone-aware data replication as an administrator in LokiStack, in order to enhance reliability in the event of a zone failure. (LOG-3266)
  • With this update, a new supported small-scale LokiStack size of 1x.extra-small is introduced for OpenShift Container Platform clusters hosting a few workloads and smaller ingestion volumes (up to 100GB/day). (LOG-4329)
  • With this update, the LokiStack administrator has access to an official Loki dashboard to inspect the storage performance and the health of each component. (LOG-4327)
1.2.17.2.3. Log Console
  • With this update, you can enable the Logging Console Plugin when Elasticsearch is the default Log Store. (LOG-3856)
  • With this update, OpenShift Container Platform application owners can receive notifications for application log-based alerts on the OpenShift Container Platform web console Developer perspective for OpenShift Container Platform version 4.14 and later. (LOG-3548)
1.2.17.3. Known Issues
  • Currently, Splunk log forwarding might not work after upgrading to version 5.8 of the Red Hat OpenShift Logging Operator. This issue is caused by transitioning from OpenSSL version 1.1.1 to version 3.0.7. In the newer OpenSSL version, there is a default behavior change, where connections to TLS 1.2 endpoints are rejected if they do not expose the RFC 5746 extension.

    As a workaround, enable TLS 1.3 support on the TLS terminating load balancer in front of the Splunk HEC (HTTP Event Collector) endpoint. Splunk is a third-party system and this should be configured from the Splunk end.

  • Currently, there is a flaw in handling multiplexed streams in the HTTP/2 protocol, where you can repeatedly make a request for a new multiplex stream and immediately send an RST_STREAM frame to cancel it. This created extra work for the server set up and tore down the streams, resulting in a denial of service due to server resource consumption. There is currently no workaround for this issue. (LOG-4609)
  • Currently, when using FluentD as the collector, the collector pod cannot start on the OpenShift Container Platform IPv6-enabled cluster. The pod logs produce the fluentd pod [error]: unexpected error error_class=SocketError error="getaddrinfo: Name or service not known error. There is currently no workaround for this issue. (LOG-4706)
  • Currently, the log alert is not available on an IPv6-enabled cluster. There is currently no workaround for this issue. (LOG-4709)
  • Currently, must-gather cannot gather any logs on a FIPS-enabled cluster, because the required OpenSSL library is not available in the cluster-logging-rhel9-operator. There is currently no workaround for this issue. (LOG-4403)
  • Currently, when deploying the logging version 5.8 on a FIPS-enabled cluster, the collector pods cannot start and are stuck in CrashLoopBackOff status, while using FluentD as a collector. There is currently no workaround for this issue. (LOG-3933)
1.2.17.4. CVEs

Chapter 2. Logging 6.0

2.1. Release notes

2.1.1. Logging 6.0.3

This release includes RHBA-2024:10991.

2.1.1.1. New features and enhancements
  • With this update, the Loki Operator supports the configuring of the workload identity federation on the Google Cloud Platform (GCP) by using the Cluster Credential Operator (CCO) in OpenShift Container Platform 4.17 or later. (LOG-6421)
2.1.1.2. Bug fixes
  • Before this update, the collector used the default settings to collect audit logs, which did not account for back pressure from output receivers. With this update, the audit log collection is optimized for file handling and log reading. (LOG-6034)
  • Before this update, any namespace containing openshift or kube was treated as an infrastructure namespace. With this update, only the following namespaces are treated as infrastructure namespaces: default, kube, openshift, and namespaces that begin with openshift- or kube-. (LOG-6204)
  • Before this update, an input receiver service was repeatedly created and deleted, causing issues with mounting the TLS secrets. With this update, the service is created once and only deleted if it is not defined in the ClusterLogForwarder custom resource. (LOG-6343)
  • Before this update, pipeline validation might enter an infinite loop if a name was a substring of another name. With this update, stricter name equality checks prevent the infinite loop. (LOG-6352)
  • Before this update, the collector alerting rules included the summary and message fields. With this update, the collector alerting rules include the summary and description fields. (LOG-6406)
  • Before this update, setting up the custom audit inputs in the ClusterLogForwarder custom resource with configured LokiStack output caused errors due to the nil pointer dereference. With this update, the Operator performs the nil checks, preventing such errors. (LOG-6441)
  • Before this update, the collector did not correctly mount the /var/log/oauth-server/ path, which prevented the collection of the audit logs. With this update, the volume mount is added, and the audit logs are collected as expected. (LOG-6486)
  • Before this update, the collector did not correctly mount the oauth-apiserver audit log file. As a result, such audit logs were not collected. With this update, the volume mount is correctly mounted, and the logs are collected as expected. (LOG-6543)
2.1.1.3. CVEs

2.1.2. Logging 6.0.2

This release includes RHBA-2024:10051.

2.1.2.1. Bug fixes
  • Before this update, Loki did not correctly load some configurations, which caused issues when using Alibaba Cloud or IBM Cloud object storage. This update fixes the configuration-loading code in Loki, resolving the issue. (LOG-5325)
  • Before this update, the collector would discard audit log messages that exceeded the configured threshold. This modifies the audit configuration thresholds for the maximum line size as well as the number of bytes read during a read cycle. (LOG-5998)
  • Before this update, the Cluster Logging Operator did not watch and reconcile resources associated with an instance of a ClusterLogForwarder like it did in prior releases. This update modifies the operator to watch and reconcile all resources it owns and creates. (LOG-6264)
  • Before this update, log events with an unknown severity level sent to Google Cloud Logging would trigger a warning in the vector collector, which would then default the severity to 'DEFAULT'. With this update, log severity levels are now standardized to match Google Cloud Logging specifications, and audit logs are assigned a severity of 'INFO'. (LOG-6296)
  • Before this update, when infrastructure namespaces were included in application inputs, the log_type was set as application. With this update, the log_type of infrastructure namespaces included in application inputs is set to infrastructure. (LOG-6354)
  • Before this update, specifying a value for the syslog.enrichment field of the ClusterLogForwarder added namespace_name, container_name, and pod_name to the messages of non-container logs. With this update, only container logs include namespace_name, container_name, and pod_name in their messages when syslog.enrichment is set. (LOG-6402)
2.1.2.2. CVEs

2.1.3. Logging 6.0.1

This release includes OpenShift Logging Bug Fix Release 6.0.1.

2.1.3.1. Bug fixes
  • With this update, the default memory limit for the collector has been increased from 1024 Mi to 2024 Mi. However, users should always adjust their resource limits according to their cluster specifications and needs. (LOG-6180)
  • Before this update, the Loki Operator failed to add the default namespace label to all AlertingRule resources, which caused the User-Workload-Monitoring Alertmanager to skip routing these alerts. This update adds the rule namespace as a label to all alerting and recording rules, resolving the issue and restoring proper alert routing in Alertmanager. (LOG-6151)
  • Before this update, the LokiStack ruler component view did not initialize properly, causing an invalid field error when the ruler component was disabled. This update ensures that the component view initializes with an empty value, resolving the issue. (LOG-6129)
  • Before this update, it was possible to set log_source in the prune filter, which could lead to inconsistent log data. With this update, the configuration is validated before being applied, and any configuration that includes log_source in the prune filter is rejected. (LOG-6202)
2.1.3.2. CVEs

2.1.4. Logging 6.0.0

This release includes Logging for Red Hat OpenShift Bug Fix Release 6.0.0

Note

Logging is provided as an installable component, with a distinct release cycle from the core OpenShift Container Platform. The Red Hat OpenShift Container Platform Life Cycle Policy outlines release compatibility.

Table 2.1. Upstream component versions
logging VersionComponent Version

Operator

eventrouter

logfilemetricexporter

loki

lokistack-gateway

opa-openshift

vector

6.0

0.4

1.1

3.1.0

0.1

0.1

0.37.1

2.1.5. Removal notice

  • With this release, logging no longer supports the ClusterLogging.logging.openshift.io and ClusterLogForwarder.logging.openshift.io custom resources. Refer to the product documentation for details on the replacement features. (LOG-5803)
  • With this release, logging no longer manages or deploys log storage (such as Elasticsearch), visualization (such as Kibana), or Fluentd-based log collectors. (LOG-5368)
Note

In order to continue to use Elasticsearch and Kibana managed by the elasticsearch-operator, the administrator must modify those object’s ownerRefs before deleting the ClusterLogging resource.

2.1.6. New features and enhancements

  • This feature introduces a new architecture for logging for Red Hat OpenShift by shifting component responsibilities to their relevant Operators, such as for storage, visualization, and collection. It introduces the ClusterLogForwarder.observability.openshift.io API for log collection and forwarding. Support for the ClusterLogging.logging.openshift.io and ClusterLogForwarder.logging.openshift.io APIs, along with the Red Hat managed Elastic stack (Elasticsearch and Kibana), is removed. Users are encouraged to migrate to the Red Hat LokiStack for log storage. Existing managed Elasticsearch deployments can be used for a limited time. Automated migration for log collection is not provided, so administrators need to create a new ClusterLogForwarder.observability.openshift.io specification to replace their previous custom resources. Refer to the official product documentation for more details. (LOG-3493)
  • With this release, the responsibility for deploying the logging view plugin shifts from the Red Hat OpenShift Logging Operator to the Cluster Observability Operator (COO). For new log storage installations that need visualization, the Cluster Observability Operator and the associated UIPlugin resource must be deployed. Refer to the Cluster Observability Operator Overview product documentation for more details. (LOG-5461)
  • This enhancement sets default requests and limits for Vector collector deployments' memory and CPU usage based on Vector documentation recommendations. (LOG-4745)
  • This enhancement updates Vector to align with the upstream version v0.37.1. (LOG-5296)
  • This enhancement introduces an alert that triggers when log collectors buffer logs to a node’s file system and use over 15% of the available space, indicating potential back pressure issues. (LOG-5381)
  • This enhancement updates the selectors for all components to use common Kubernetes labels. (LOG-5906)
  • This enhancement changes the collector configuration to deploy as a ConfigMap instead of a secret, allowing users to view and edit the configuration when the ClusterLogForwarder is set to Unmanaged. (LOG-5599)
  • This enhancement adds the ability to configure the Vector collector log level using an annotation on the ClusterLogForwarder, with options including trace, debug, info, warn, error, or off. (LOG-5372)
  • This enhancement adds validation to reject configurations where Amazon CloudWatch outputs use multiple AWS roles, preventing incorrect log routing. (LOG-5640)
  • This enhancement removes the Log Bytes Collected and Log Bytes Sent graphs from the metrics dashboard. (LOG-5964)
  • This enhancement updates the must-gather functionality to only capture information for inspecting Logging 6.0 components, including Vector deployments from ClusterLogForwarder.observability.openshift.io resources and the Red Hat managed LokiStack. (LOG-5949)
  • This enhancement improves Azure storage secret validation by providing early warnings for specific error conditions. (LOG-4571)

2.1.7. Technology Preview features

  • This release introduces a Technology Preview feature for log forwarding using OpenTelemetry. A new output type,` OTLP`, allows sending JSON-encoded log records using the OpenTelemetry data model and resource semantic conventions. (LOG-4225)

2.1.8. Bug fixes

  • Before this update, the CollectorHighErrorRate and CollectorVeryHighErrorRate alerts were still present. With this update, both alerts are removed in the logging 6.0 release but might return in a future release. (LOG-3432)

2.1.9. CVEs

2.2. Logging 6.0

The ClusterLogForwarder custom resource (CR) is the central configuration point for log collection and forwarding.

2.2.1. Inputs and Outputs

Inputs specify the sources of logs to be forwarded. Logging provides built-in input types: application, infrastructure, and audit, which select logs from different parts of your cluster. You can also define custom inputs based on namespaces or pod labels to fine-tune log selection.

Outputs define the destinations where logs are sent. Each output type has its own set of configuration options, allowing you to customize the behavior and authentication settings.

2.2.2. Receiver Input Type

The receiver input type enables the Logging system to accept logs from external sources. It supports two formats for receiving logs: http and syslog.

The ReceiverSpec defines the configuration for a receiver input.

2.2.3. Pipelines and Filters

Pipelines determine the flow of logs from inputs to outputs. A pipeline consists of one or more input refs, output refs, and optional filter refs. Filters can be used to transform or drop log messages within a pipeline. The order of filters matters, as they are applied sequentially, and earlier filters can prevent log messages from reaching later stages.

2.2.4. Operator Behavior

The Cluster Logging Operator manages the deployment and configuration of the collector based on the managementState field:

  • When set to Managed (default), the operator actively manages the logging resources to match the configuration defined in the spec.
  • When set to Unmanaged, the operator does not take any action, allowing you to manually manage the logging components.

2.2.5. Validation

Logging includes extensive validation rules and default values to ensure a smooth and error-free configuration experience. The ClusterLogForwarder resource enforces validation checks on required fields, dependencies between fields, and the format of input values. Default values are provided for certain fields, reducing the need for explicit configuration in common scenarios.

2.2.5.1. Quick Start

Prerequisites

  • Cluster administrator permissions

Procedure

  1. Install the OpenShift Logging and Loki Operators from OperatorHub.
  2. Create a secret to access an existing object storage bucket:

    Example command for AWS

    $ oc create secret generic logging-loki-s3 \
      --from-literal=bucketnames="<bucket_name>" \
      --from-literal=endpoint="<aws_bucket_endpoint>" \
      --from-literal=access_key_id="<aws_access_key_id>" \
      --from-literal=access_key_secret="<aws_access_key_secret>" \
      --from-literal=region="<aws_region_of_your_bucket>" \
      -n openshift-logging

  3. Create a LokiStack custom resource (CR) in the openshift-logging namespace:

    apiVersion: loki.grafana.com/v1
    kind: LokiStack
    metadata:
      name: logging-loki
      namespace: openshift-logging
    spec:
      managementState: Managed
      size: 1x.extra-small
      storage:
        schemas:
        - effectiveDate: '2022-06-01'
          version: v13
        secret:
          name: logging-loki-s3
          type: s3
      storageClassName: gp3-csi
      tenants:
        mode: openshift-logging
  4. Create a service account for the collector:

    $ oc create sa collector -n openshift-logging
  5. Bind the ClusterRole to the service account:

    $ oc adm policy add-cluster-role-to-user logging-collector-logs-writer -z collector -n openshift-logging
  6. Install the Cluster Observability Operator.
  7. Create a UIPlugin to enable the Log section in the Observe tab:

    apiVersion: observability.openshift.io/v1alpha1
    kind: UIPlugin
    metadata:
      name: logging
    spec:
      type: Logging
      logging:
        lokiStack:
          name: logging-loki
  8. Add additional roles to the collector service account:

    $ oc adm policy add-cluster-role-to-user collect-application-logs -z collector -n openshift-logging
    $ oc adm policy add-cluster-role-to-user collect-audit-logs -z collector -n openshift-logging
    $ oc adm policy add-cluster-role-to-user collect-infrastructure-logs -z collector -n openshift-logging
  9. Create a ClusterLogForwarder CR to configure log forwarding:

    apiVersion: observability.openshift.io/v1
    kind: ClusterLogForwarder
    metadata:
      name: collector
      namespace: openshift-logging
    spec:
      serviceAccount:
        name: collector
      outputs:
      - name: default-lokistack
        type: lokiStack
        lokiStack:
          target:
            name: logging-loki
            namespace: openshift-logging
          authentication:
            token:
              from: serviceAccount
        tls:
          ca:
            key: service-ca.crt
            configMapName: openshift-service-ca.crt
      pipelines:
      - name: default-logstore
        inputRefs:
        - application
        - infrastructure
        outputRefs:
        - default-lokistack
  10. Verify that logs are visible in the Log section of the Observe tab in the OpenShift web console.

2.3. Upgrading to Logging 6.0

Logging v6.0 is a significant upgrade from previous releases, achieving several longstanding goals of Cluster Logging:

  • Introduction of distinct operators to manage logging components (e.g., collectors, storage, visualization).
  • Removal of support for managed log storage and visualization based on Elastic products (i.e., Elasticsearch, Kibana).
  • Deprecation of the Fluentd log collector implementation.
  • Removal of support for ClusterLogging.logging.openshift.io and ClusterLogForwarder.logging.openshift.io resources.
Note

The cluster-logging-operator does not provide an automated upgrade process.

Given the various configurations for log collection, forwarding, and storage, no automated upgrade is provided by the cluster-logging-operator. This documentation assists administrators in converting existing ClusterLogging.logging.openshift.io and ClusterLogForwarder.logging.openshift.io specifications to the new API. Examples of migrated ClusterLogForwarder.observability.openshift.io resources for common use cases are included.

2.3.1. Using the oc explain command

The oc explain command is an essential tool in the OpenShift CLI oc that provides detailed descriptions of the fields within Custom Resources (CRs). This command is invaluable for administrators and developers who are configuring or troubleshooting resources in an OpenShift cluster.

2.3.1.1. Resource Descriptions

oc explain offers in-depth explanations of all fields associated with a specific object. This includes standard resources like pods and services, as well as more complex entities like statefulsets and custom resources defined by Operators.

To view the documentation for the outputs field of the ClusterLogForwarder custom resource, you can use:

$ oc explain clusterlogforwarders.observability.openshift.io.spec.outputs
Note

In place of clusterlogforwarder the short form obsclf can be used.

This will display detailed information about these fields, including their types, default values, and any associated sub-fields.

2.3.1.2. Hierarchical Structure

The command displays the structure of resource fields in a hierarchical format, clarifying the relationships between different configuration options.

For instance, here’s how you can drill down into the storage configuration for a LokiStack custom resource:

$ oc explain lokistacks.loki.grafana.com
$ oc explain lokistacks.loki.grafana.com.spec
$ oc explain lokistacks.loki.grafana.com.spec.storage
$ oc explain lokistacks.loki.grafana.com.spec.storage.schemas

Each command reveals a deeper level of the resource specification, making the structure clear.

2.3.1.3. Type Information

oc explain also indicates the type of each field (such as string, integer, or boolean), allowing you to verify that resource definitions use the correct data types.

For example:

$ oc explain lokistacks.loki.grafana.com.spec.size

This will show that size should be defined using an integer value.

2.3.1.4. Default Values

When applicable, the command shows the default values for fields, providing insights into what values will be used if none are explicitly specified.

Again using lokistacks.loki.grafana.com as an example:

$ oc explain lokistacks.spec.template.distributor.replicas

Example output

GROUP:      loki.grafana.com
KIND:       LokiStack
VERSION:    v1

FIELD: replicas <integer>

DESCRIPTION:
    Replicas defines the number of replica pods of the component.

2.3.2. Log Storage

The only managed log storage solution available in this release is a Lokistack, managed by the loki-operator. This solution, previously available as the preferred alternative to the managed Elasticsearch offering, remains unchanged in its deployment process.

Important

To continue using an existing Red Hat managed Elasticsearch or Kibana deployment provided by the elasticsearch-operator, remove the owner references from the Elasticsearch resource named elasticsearch, and the Kibana resource named kibana in the openshift-logging namespace before removing the ClusterLogging resource named instance in the same namespace.

  1. Temporarily set ClusterLogging to state Unmanaged

    $ oc -n openshift-logging patch clusterlogging/instance -p '{"spec":{"managementState": "Unmanaged"}}' --type=merge
  2. Remove ClusterLogging ownerReferences from the Elasticsearch resource

    The following command ensures that ClusterLogging no longer owns the Elasticsearch resource. Updates to the ClusterLogging resource’s logStore field will no longer affect the Elasticsearch resource.

    $ oc -n openshift-logging patch elasticsearch/elasticsearch -p '{"metadata":{"ownerReferences": []}}' --type=merge
  3. Remove ClusterLogging ownerReferences from the Kibana resource

    The following command ensures that ClusterLogging no longer owns the Kibana resource. Updates to the ClusterLogging resource’s visualization field will no longer affect the Kibana resource.

    $ oc -n openshift-logging patch kibana/kibana -p '{"metadata":{"ownerReferences": []}}' --type=merge
  4. Set ClusterLogging to state Managed
$ oc -n openshift-logging patch clusterlogging/instance -p '{"spec":{"managementState": "Managed"}}' --type=merge

2.3.3. Log Visualization

The OpenShift console UI plugin for log visualization has been moved to the cluster-observability-operator from the cluster-logging-operator.

2.3.4. Log Collection and Forwarding

Log collection and forwarding configurations are now specified under the new API, part of the observability.openshift.io API group. The following sections highlight the differences from the old API resources.

Note

Vector is the only supported collector implementation.

2.3.5. Management, Resource Allocation, and Workload Scheduling

Configuration for management state (e.g., Managed, Unmanaged), resource requests and limits, tolerations, and node selection is now part of the new ClusterLogForwarder API.

Previous Configuration

apiVersion: "logging.openshift.io/v1"
kind: "ClusterLogging"
spec:
  managementState: "Managed"
  collection:
    resources:
      limits: {}
      requests: {}
    nodeSelector: {}
    tolerations: {}

Current Configuration

apiVersion: "observability.openshift.io/v1"
kind: ClusterLogForwarder
spec:
  managementState: Managed
  collector:
    resources:
      limits: {}
      requests: {}
    nodeSelector: {}
    tolerations: {}

2.3.6. Input Specifications

The input specification is an optional part of the ClusterLogForwarder specification. Administrators can continue to use the predefined values of application, infrastructure, and audit to collect these sources.

2.3.6.1. Application Inputs

Namespace and container inclusions and exclusions have been consolidated into a single field.

5.9 Application Input with Namespace and Container Includes and Excludes

apiVersion: "logging.openshift.io/v1"
kind: ClusterLogForwarder
spec:
  inputs:
   - name: application-logs
     type: application
     application:
       namespaces:
       - foo
       - bar
       includes:
       - namespace: my-important
         container: main
       excludes:
       - container: too-verbose

6.0 Application Input with Namespace and Container Includes and Excludes

apiVersion: "observability.openshift.io/v1"
kind: ClusterLogForwarder
spec:
  inputs:
   - name: application-logs
     type: application
     application:
       includes:
       - namespace: foo
       - namespace: bar
       - namespace: my-important
         container: main
       excludes:
       - container: too-verbose

Note

application, infrastructure, and audit are reserved words and cannot be used as names when defining an input.

2.3.6.2. Input Receivers

Changes to input receivers include:

  • Explicit configuration of the type at the receiver level.
  • Port settings moved to the receiver level.

5.9 Input Receivers

apiVersion: "logging.openshift.io/v1"
kind: ClusterLogForwarder
spec:
  inputs:
  - name: an-http
    receiver:
      http:
        port: 8443
        format: kubeAPIAudit
  - name: a-syslog
    receiver:
      type: syslog
      syslog:
        port: 9442

6.0 Input Receivers

apiVersion: "observability.openshift.io/v1"
kind: ClusterLogForwarder
spec:
  inputs:
  - name: an-http
    type: receiver
    receiver:
      type: http
      port: 8443
      http:
        format: kubeAPIAudit
  - name: a-syslog
    type: receiver
    receiver:
      type: syslog
      port: 9442

2.3.7. Output Specifications

High-level changes to output specifications include:

  • URL settings moved to each output type specification.
  • Tuning parameters moved to each output type specification.
  • Separation of TLS configuration from authentication.
  • Explicit configuration of keys and secret/configmap for TLS and authentication.

2.3.8. Secrets and TLS Configuration

Secrets and TLS configurations are now separated into authentication and TLS configuration for each output. They must be explicitly defined in the specification rather than relying on administrators to define secrets with recognized keys. Upgrading TLS and authorization configurations requires administrators to understand previously recognized keys to continue using existing secrets. Examples in the following sections provide details on how to configure ClusterLogForwarder secrets to forward to existing Red Hat managed log storage solutions.

2.3.9. Red Hat Managed Elasticsearch

v5.9 Forwarding to Red Hat Managed Elasticsearch

apiVersion: logging.openshift.io/v1
kind: ClusterLogging
metadata:
  name: instance
  namespace: openshift-logging
spec:
  logStore:
    type: elasticsearch

v6.0 Forwarding to Red Hat Managed Elasticsearch

apiVersion: observability.openshift.io/v1
kind: ClusterLogForwarder
metadata:
  name: instance
  namespace: openshift-logging
spec:
  outputs:
  - name: default-elasticsearch
    type: elasticsearch
    elasticsearch:
      url: https://elasticsearch:9200
      version: 6
      index: <log_type>-write-{+yyyy.MM.dd}
    tls:
      ca:
        key: ca-bundle.crt
        secretName: collector
      certificate:
        key: tls.crt
        secretName: collector
      key:
        key: tls.key
        secretName: collector
  pipelines:
  - outputRefs:
    - default-elasticsearch
  - inputRefs:
    - application
    - infrastructure

Note

In this example, application logs are written to the application-write alias/index instead of app-write.

2.3.10. Red Hat Managed LokiStack

v5.9 Forwarding to Red Hat Managed LokiStack

apiVersion: logging.openshift.io/v1
kind: ClusterLogging
metadata:
  name: instance
  namespace: openshift-logging
spec:
  logStore:
    type: lokistack
    lokistack:
      name: lokistack-dev

v6.0 Forwarding to Red Hat Managed LokiStack

apiVersion: observability.openshift.io/v1
kind: ClusterLogForwarder
metadata:
  name: instance
  namespace: openshift-logging
spec:
  outputs:
  - name: default-lokistack
    type: lokiStack
    lokiStack:
      target:
        name: lokistack-dev
        namespace: openshift-logging
      authentication:
        token:
          from: serviceAccount
    tls:
      ca:
        key: service-ca.crt
        configMapName: openshift-service-ca.crt
  pipelines:
  - outputRefs:
    - default-lokistack
  - inputRefs:
    - application
    - infrastructure

2.3.11. Filters and Pipeline Configuration

Pipeline configurations now define only the routing of input sources to their output destinations, with any required transformations configured separately as filters. All attributes of pipelines from previous releases have been converted to filters in this release. Individual filters are defined in the filters specification and referenced by a pipeline.

5.9 Filters

apiVersion: logging.openshift.io/v1
kind: ClusterLogForwarder
spec:
  pipelines:
   - name: application-logs
     parse: json
     labels:
       foo: bar
     detectMultilineErrors: true

6.0 Filter Configuration

apiVersion: observability.openshift.io/v1
kind: ClusterLogForwarder
spec:
  filters:
  - name: detectexception
    type: detectMultilineException
  - name: parse-json
    type: parse
  - name: labels
    type: openshiftLabels
    openshiftLabels:
      foo: bar
  pipelines:
  - name: application-logs
    filterRefs:
    - detectexception
    - labels
    - parse-json

2.3.12. Validation and Status

Most validations are enforced when a resource is created or updated, providing immediate feedback. This is a departure from previous releases, where validation occurred post-creation and required inspecting the resource status. Some validation still occurs post-creation for cases where it is not possible to validate at creation or update time.

Instances of the ClusterLogForwarder.observability.openshift.io must satisfy the following conditions before the operator will deploy the log collector: Authorized, Valid, Ready. An example of these conditions is:

6.0 Status Conditions

apiVersion: observability.openshift.io/v1
kind: ClusterLogForwarder
status:
  conditions:
  - lastTransitionTime: "2024-09-13T03:28:44Z"
    message: 'permitted to collect log types: [application]'
    reason: ClusterRolesExist
    status: "True"
    type: observability.openshift.io/Authorized
  - lastTransitionTime: "2024-09-13T12:16:45Z"
    message: ""
    reason: ValidationSuccess
    status: "True"
    type: observability.openshift.io/Valid
  - lastTransitionTime: "2024-09-13T12:16:45Z"
    message: ""
    reason: ReconciliationComplete
    status: "True"
    type: Ready
  filterConditions:
  - lastTransitionTime: "2024-09-13T13:02:59Z"
    message: filter "detectexception" is valid
    reason: ValidationSuccess
    status: "True"
    type: observability.openshift.io/ValidFilter-detectexception
  - lastTransitionTime: "2024-09-13T13:02:59Z"
    message: filter "parse-json" is valid
    reason: ValidationSuccess
    status: "True"
    type: observability.openshift.io/ValidFilter-parse-json
  inputConditions:
  - lastTransitionTime: "2024-09-13T12:23:03Z"
    message: input "application1" is valid
    reason: ValidationSuccess
    status: "True"
    type: observability.openshift.io/ValidInput-application1
  outputConditions:
  - lastTransitionTime: "2024-09-13T13:02:59Z"
    message: output "default-lokistack-application1" is valid
    reason: ValidationSuccess
    status: "True"
    type: observability.openshift.io/ValidOutput-default-lokistack-application1
  pipelineConditions:
  - lastTransitionTime: "2024-09-13T03:28:44Z"
    message: pipeline "default-before" is valid
    reason: ValidationSuccess
    status: "True"
    type: observability.openshift.io/ValidPipeline-default-before

Note

Conditions that are satisfied and applicable have a "status" value of "True". Conditions with a status other than "True" provide a reason and a message explaining the issue.

2.4. Configuring log forwarding

The ClusterLogForwarder (CLF) allows users to configure forwarding of logs to various destinations. It provides a flexible way to select log messages from different sources, send them through a pipeline that can transform or filter them, and forward them to one or more outputs.

Key Functions of the ClusterLogForwarder

  • Selects log messages using inputs
  • Forwards logs to external destinations using outputs
  • Filters, transforms, and drops log messages using filters
  • Defines log forwarding pipelines connecting inputs, filters and outputs

2.4.1. Setting up log collection

This release of Cluster Logging requires administrators to explicitly grant log collection permissions to the service account associated with ClusterLogForwarder. This was not required in previous releases for the legacy logging scenario consisting of a ClusterLogging and, optionally, a ClusterLogForwarder.logging.openshift.io resource.

The Red Hat OpenShift Logging Operator provides collect-audit-logs, collect-application-logs, and collect-infrastructure-logs cluster roles, which enable the collector to collect audit logs, application logs, and infrastructure logs respectively.

Setup log collection by binding the required cluster roles to your service account.

2.4.1.1. Legacy service accounts

To use the existing legacy service account logcollector, create the following ClusterRoleBinding:

$ oc adm policy add-cluster-role-to-user collect-application-logs system:serviceaccount:openshift-logging:logcollector
$ oc adm policy add-cluster-role-to-user collect-infrastructure-logs system:serviceaccount:openshift-logging:logcollector

Additionally, create the following ClusterRoleBinding if collecting audit logs:

$ oc adm policy add-cluster-role-to-user collect-audit-logs system:serviceaccount:openshift-logging:logcollector
2.4.1.2. Creating service accounts

Prerequisites

  • The Red Hat OpenShift Logging Operator is installed in the openshift-logging namespace.
  • You have administrator permissions.

Procedure

  1. Create a service account for the collector. If you want to write logs to storage that requires a token for authentication, you must include a token in the service account.
  2. Bind the appropriate cluster roles to the service account:

    Example binding command

    $ oc adm policy add-cluster-role-to-user <cluster_role_name> system:serviceaccount:<namespace_name>:<service_account_name>

2.4.1.2.1. Cluster Role Binding for your Service Account

The role_binding.yaml file binds the ClusterLogging operator’s ClusterRole to a specific ServiceAccount, allowing it to manage Kubernetes resources cluster-wide.

apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
  name: manager-rolebinding
roleRef:                                           1
  apiGroup: rbac.authorization.k8s.io              2
  kind: ClusterRole                                3
  name: cluster-logging-operator                   4
subjects:                                          5
  - kind: ServiceAccount                           6
    name: cluster-logging-operator                 7
    namespace: openshift-logging                   8
1
roleRef: References the ClusterRole to which the binding applies.
2
apiGroup: Indicates the RBAC API group, specifying that the ClusterRole is part of Kubernetes' RBAC system.
3
kind: Specifies that the referenced role is a ClusterRole, which applies cluster-wide.
4
name: The name of the ClusterRole being bound to the ServiceAccount, here cluster-logging-operator.
5
subjects: Defines the entities (users or service accounts) that are being granted the permissions from the ClusterRole.
6
kind: Specifies that the subject is a ServiceAccount.
7
Name: The name of the ServiceAccount being granted the permissions.
8
namespace: Indicates the namespace where the ServiceAccount is located.
2.4.1.2.2. Writing application logs

The write-application-logs-clusterrole.yaml file defines a ClusterRole that grants permissions to write application logs to the Loki logging application.

apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
  name: cluster-logging-write-application-logs
rules:                                              1
  - apiGroups:                                      2
      - loki.grafana.com                            3
    resources:                                      4
      - application                                 5
    resourceNames:                                  6
      - logs                                        7
    verbs:                                          8
      - create                                      9
Annotations
<1> rules: Specifies the permissions granted by this ClusterRole.
<2> apiGroups: Refers to the API group loki.grafana.com, which relates to the Loki logging system.
<3> loki.grafana.com: The API group for managing Loki-related resources.
<4> resources: The resource type that the ClusterRole grants permission to interact with.
<5> application: Refers to the application resources within the Loki logging system.
<6> resourceNames: Specifies the names of resources that this role can manage.
<7> logs: Refers to the log resources that can be created.
<8> verbs: The actions allowed on the resources.
<9> create: Grants permission to create new logs in the Loki system.
2.4.1.2.3. Writing audit logs

The write-audit-logs-clusterrole.yaml file defines a ClusterRole that grants permissions to create audit logs in the Loki logging system.

apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
  name: cluster-logging-write-audit-logs
rules:                                              1
  - apiGroups:                                      2
      - loki.grafana.com                            3
    resources:                                      4
      - audit                                       5
    resourceNames:                                  6
      - logs                                        7
    verbs:                                          8
      - create                                      9
1 1
rules: Defines the permissions granted by this ClusterRole.
2 2
apiGroups: Specifies the API group loki.grafana.com.
3 3
loki.grafana.com: The API group responsible for Loki logging resources.
4 4
resources: Refers to the resource type this role manages, in this case, audit.
5 5
audit: Specifies that the role manages audit logs within Loki.
6 6
resourceNames: Defines the specific resources that the role can access.
7 7
logs: Refers to the logs that can be managed under this role.
8 8
verbs: The actions allowed on the resources.
9 9
create: Grants permission to create new audit logs.
2.4.1.2.4. Writing infrastructure logs

The write-infrastructure-logs-clusterrole.yaml file defines a ClusterRole that grants permission to create infrastructure logs in the Loki logging system.

Sample YAML

apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
  name: cluster-logging-write-infrastructure-logs
rules:                                              1
  - apiGroups:                                      2
      - loki.grafana.com                            3
    resources:                                      4
      - infrastructure                              5
    resourceNames:                                  6
      - logs                                        7
    verbs:                                          8
      - create                                      9

1
rules: Specifies the permissions this ClusterRole grants.
2
apiGroups: Specifies the API group for Loki-related resources.
3
loki.grafana.com: The API group managing the Loki logging system.
4
resources: Defines the resource type that this role can interact with.
5
infrastructure: Refers to infrastructure-related resources that this role manages.
6
resourceNames: Specifies the names of resources this role can manage.
7
logs: Refers to the log resources related to infrastructure.
8
verbs: The actions permitted by this role.
9
create: Grants permission to create infrastructure logs in the Loki system.
2.4.1.2.5. ClusterLogForwarder editor role

The clusterlogforwarder-editor-role.yaml file defines a ClusterRole that allows users to manage ClusterLogForwarders in OpenShift.

apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
  name: clusterlogforwarder-editor-role
rules:                                              1
  - apiGroups:                                      2
      - observability.openshift.io                  3
    resources:                                      4
      - clusterlogforwarders                        5
    verbs:                                          6
      - create                                      7
      - delete                                      8
      - get                                         9
      - list                                        10
      - patch                                       11
      - update                                      12
      - watch                                       13
1
rules: Specifies the permissions this ClusterRole grants.
2
apiGroups: Refers to the OpenShift-specific API group
3
obervability.openshift.io: The API group for managing observability resources, like logging.
4
resources: Specifies the resources this role can manage.
5
clusterlogforwarders: Refers to the log forwarding resources in OpenShift.
6
verbs: Specifies the actions allowed on the ClusterLogForwarders.
7
create: Grants permission to create new ClusterLogForwarders.
8
delete: Grants permission to delete existing ClusterLogForwarders.
9
get: Grants permission to retrieve information about specific ClusterLogForwarders.
10
list: Allows listing all ClusterLogForwarders.
11
patch: Grants permission to partially modify ClusterLogForwarders.
12
update: Grants permission to update existing ClusterLogForwarders.
13
watch: Grants permission to monitor changes to ClusterLogForwarders.

2.4.2. Modifying log level in collector

To modify the log level in the collector, you can set the observability.openshift.io/log-level annotation to trace, debug, info, warn, error, and off.

Example log level annotation

apiVersion: observability.openshift.io/v1
kind: ClusterLogForwarder
metadata:
  name: collector
  annotations:
    observability.openshift.io/log-level: debug
# ...

2.4.3. Managing the Operator

The ClusterLogForwarder resource has a managementState field that controls whether the operator actively manages its resources or leaves them Unmanaged:

Managed
(default) The operator will drive the logging resources to match the desired state in the CLF spec.
Unmanaged
The operator will not take any action related to the logging components.

This allows administrators to temporarily pause log forwarding by setting managementState to Unmanaged.

2.4.4. Structure of the ClusterLogForwarder

The CLF has a spec section that contains the following key components:

Inputs
Select log messages to be forwarded. Built-in input types application, infrastructure and audit forward logs from different parts of the cluster. You can also define custom inputs.
Outputs
Define destinations to forward logs to. Each output has a unique name and type-specific configuration.
Pipelines
Define the path logs take from inputs, through filters, to outputs. Pipelines have a unique name and consist of a list of input, output and filter names.
Filters
Transform or drop log messages in the pipeline. Users can define filters that match certain log fields and drop or modify the messages. Filters are applied in the order specified in the pipeline.
2.4.4.1. Inputs

Inputs are configured in an array under spec.inputs. There are three built-in input types:

application
Selects logs from all application containers, excluding those in infrastructure namespaces such as default, openshift, or any namespace with the kube- or openshift- prefix.
infrastructure
Selects logs from infrastructure components running in default and openshift namespaces and node logs.
audit
Selects logs from the OpenShift API server audit logs, Kubernetes API server audit logs, ovn audit logs, and node audit logs from auditd.

Users can define custom inputs of type application that select logs from specific namespaces or using pod labels.

2.4.4.2. Outputs

Outputs are configured in an array under spec.outputs. Each output must have a unique name and a type. Supported types are:

azureMonitor
Forwards logs to Azure Monitor.
cloudwatch
Forwards logs to AWS CloudWatch.
elasticsearch
Forwards logs to an external Elasticsearch instance.
googleCloudLogging
Forwards logs to Google Cloud Logging.
http
Forwards logs to a generic HTTP endpoint.
kafka
Forwards logs to a Kafka broker.
loki
Forwards logs to a Loki logging backend.
lokistack
Forwards logs to the logging supported combination of Loki and web proxy with OpenShift Container Platform authentication integration. LokiStack’s proxy uses OpenShift Container Platform authentication to enforce multi-tenancy
otlp
Forwards logs using the OpenTelemetry Protocol.
splunk
Forwards logs to Splunk.
syslog
Forwards logs to an external syslog server.

Each output type has its own configuration fields.

2.4.4.3. Pipelines

Pipelines are configured in an array under spec.pipelines. Each pipeline must have a unique name and consists of:

inputRefs
Names of inputs whose logs should be forwarded to this pipeline.
outputRefs
Names of outputs to send logs to.
filterRefs
(optional) Names of filters to apply.

The order of filterRefs matters, as they are applied sequentially. Earlier filters can drop messages that will not be processed by later filters.

2.4.4.4. Filters

Filters are configured in an array under spec.filters. They can match incoming log messages based on the value of structured fields and modify or drop them.

Administrators can configure the following types of filters:

2.4.4.5. Enabling multi-line exception detection

Enables multi-line error detection of container logs.

Warning

Enabling this feature could have performance implications and may require additional computing resources or alternate logging solutions.

Log parsers often incorrectly identify separate lines of the same exception as separate exceptions. This leads to extra log entries and an incomplete or inaccurate view of the traced information.

Example java exception

java.lang.NullPointerException: Cannot invoke "String.toString()" because "<param1>" is null
    at testjava.Main.handle(Main.java:47)
    at testjava.Main.printMe(Main.java:19)
    at testjava.Main.main(Main.java:10)

  • To enable logging to detect multi-line exceptions and reassemble them into a single log entry, ensure that the ClusterLogForwarder Custom Resource (CR) contains a detectMultilineErrors field under the .spec.filters.

Example ClusterLogForwarder CR

apiVersion: "observability.openshift.io/v1"
kind: ClusterLogForwarder
metadata:
  name: <log_forwarder_name>
  namespace: <log_forwarder_namespace>
spec:
  serviceAccount:
    name: <service_account_name>
  filters:
  - name: <name>
    type: detectMultilineException
  pipelines:
    - inputRefs:
        - <input-name>
      name: <pipeline-name>
      filterRefs:
        - <filter-name>
      outputRefs:
        - <output-name>

2.4.4.5.1. Details

When log messages appear as a consecutive sequence forming an exception stack trace, they are combined into a single, unified log record. The first log message’s content is replaced with the concatenated content of all the message fields in the sequence.

The collector supports the following languages:

  • Java
  • JS
  • Ruby
  • Python
  • Golang
  • PHP
  • Dart
2.4.4.6. Configuring content filters to drop unwanted log records

When the drop filter is configured, the log collector evaluates log streams according to the filters before forwarding. The collector drops unwanted log records that match the specified configuration.

Procedure

  1. Add a configuration for a filter to the filters spec in the ClusterLogForwarder CR.

    The following example shows how to configure the ClusterLogForwarder CR to drop log records based on regular expressions:

    Example ClusterLogForwarder CR

    apiVersion: observability.openshift.io/v1
    kind: ClusterLogForwarder
    metadata:
    # ...
    spec:
      serviceAccount:
        name: <service_account_name>
      filters:
      - name: <filter_name>
        type: drop 1
        drop: 2
        - test: 3
          - field: .kubernetes.labels."foo-bar/baz" 4
            matches: .+ 5
          - field: .kubernetes.pod_name
            notMatches: "my-pod" 6
      pipelines:
      - name: <pipeline_name> 7
        filterRefs: ["<filter_name>"]
    # ...

    1
    Specifies the type of filter. The drop filter drops log records that match the filter configuration.
    2
    Specifies configuration options for applying the drop filter.
    3
    Specifies the configuration for tests that are used to evaluate whether a log record is dropped.
    • If all the conditions specified for a test are true, the test passes and the log record is dropped.
    • When multiple tests are specified for the drop filter configuration, if any of the tests pass, the record is dropped.
    • If there is an error evaluating a condition, for example, the field is missing from the log record being evaluated, that condition evaluates to false.
    4
    Specifies a dot-delimited field path, which is a path to a field in the log record. The path can contain alpha-numeric characters and underscores (a-zA-Z0-9_), for example, .kubernetes.namespace_name. If segments contain characters outside of this range, the segment must be in quotes, for example, .kubernetes.labels."foo.bar-bar/baz". You can include multiple field paths in a single test configuration, but they must all evaluate to true for the test to pass and the drop filter to be applied.
    5
    Specifies a regular expression. If log records match this regular expression, they are dropped. You can set either the matches or notMatches condition for a single field path, but not both.
    6
    Specifies a regular expression. If log records do not match this regular expression, they are dropped. You can set either the matches or notMatches condition for a single field path, but not both.
    7
    Specifies the pipeline that the drop filter is applied to.
  2. Apply the ClusterLogForwarder CR by running the following command:

    $ oc apply -f <filename>.yaml

Additional examples

The following additional example shows how you can configure the drop filter to only keep higher priority log records:

apiVersion: observability.openshift.io/v1
kind: ClusterLogForwarder
metadata:
# ...
spec:
  serviceAccount:
    name: <service_account_name>
  filters:
  - name: important
    type: drop
    drop:
    - test:
      - field: .message
        notMatches: "(?i)critical|error"
      - field: .level
        matches: "info|warning"
# ...

In addition to including multiple field paths in a single test configuration, you can also include additional tests that are treated as OR checks. In the following example, records are dropped if either test configuration evaluates to true. However, for the second test configuration, both field specs must be true for it to be evaluated to true:

apiVersion: observability.openshift.io/v1
kind: ClusterLogForwarder
metadata:
# ...
spec:
  serviceAccount:
    name: <service_account_name>
  filters:
  - name: important
    type: drop
    drop:
    - test:
      - field: .kubernetes.namespace_name
        matches: "^open"
    - test:
      - field: .log_type
        matches: "application"
      - field: .kubernetes.pod_name
        notMatches: "my-pod"
# ...
2.4.4.7. Overview of API audit filter

OpenShift API servers generate audit events for each API call, detailing the request, response, and the identity of the requester, leading to large volumes of data. The API Audit filter uses rules to enable the exclusion of non-essential events and the reduction of event size, facilitating a more manageable audit trail. Rules are checked in order, and checking stops at the first match. The amount of data that is included in an event is determined by the value of the level field:

  • None: The event is dropped.
  • Metadata: Audit metadata is included, request and response bodies are removed.
  • Request: Audit metadata and the request body are included, the response body is removed.
  • RequestResponse: All data is included: metadata, request body and response body. The response body can be very large. For example, oc get pods -A generates a response body containing the YAML description of every pod in the cluster.

The ClusterLogForwarder custom resource (CR) uses the same format as the standard Kubernetes audit policy, while providing the following additional functions:

Wildcards
Names of users, groups, namespaces, and resources can have a leading or trailing * asterisk character. For example, the namespace openshift-\* matches openshift-apiserver or openshift-authentication. Resource \*/status matches Pod/status or Deployment/status.
Default Rules

Events that do not match any rule in the policy are filtered as follows:

  • Read-only system events such as get, list, and watch are dropped.
  • Service account write events that occur within the same namespace as the service account are dropped.
  • All other events are forwarded, subject to any configured rate limits.

To disable these defaults, either end your rules list with a rule that has only a level field or add an empty rule.

Omit Response Codes
A list of integer status codes to omit. You can drop events based on the HTTP status code in the response by using the OmitResponseCodes field, which lists HTTP status codes for which no events are created. The default value is [404, 409, 422, 429]. If the value is an empty list, [], then no status codes are omitted.

The ClusterLogForwarder CR audit policy acts in addition to the OpenShift Container Platform audit policy. The ClusterLogForwarder CR audit filter changes what the log collector forwards and provides the ability to filter by verb, user, group, namespace, or resource. You can create multiple filters to send different summaries of the same audit stream to different places. For example, you can send a detailed stream to the local cluster log store and a less detailed stream to a remote site.

Note

You must have a cluster role collect-audit-logs to collect the audit logs. The following example provided is intended to illustrate the range of rules possible in an audit policy and is not a recommended configuration.

Example audit policy

apiVersion: observability.openshift.io/v1
kind: ClusterLogForwarder
metadata:
  name: <log_forwarder_name>
  namespace: <log_forwarder_namespace>
spec:
  serviceAccount:
    name: <service_account_name>
  pipelines:
    - name: my-pipeline
      inputRefs: audit 1
      filterRefs: my-policy 2
  filters:
    - name: my-policy
      type: kubeAPIAudit
      kubeAPIAudit:
        # Don't generate audit events for all requests in RequestReceived stage.
        omitStages:
          - "RequestReceived"

        rules:
          # Log pod changes at RequestResponse level
          - level: RequestResponse
            resources:
            - group: ""
              resources: ["pods"]

          # Log "pods/log", "pods/status" at Metadata level
          - level: Metadata
            resources:
            - group: ""
              resources: ["pods/log", "pods/status"]

          # Don't log requests to a configmap called "controller-leader"
          - level: None
            resources:
            - group: ""
              resources: ["configmaps"]
              resourceNames: ["controller-leader"]

          # Don't log watch requests by the "system:kube-proxy" on endpoints or services
          - level: None
            users: ["system:kube-proxy"]
            verbs: ["watch"]
            resources:
            - group: "" # core API group
              resources: ["endpoints", "services"]

          # Don't log authenticated requests to certain non-resource URL paths.
          - level: None
            userGroups: ["system:authenticated"]
            nonResourceURLs:
            - "/api*" # Wildcard matching.
            - "/version"

          # Log the request body of configmap changes in kube-system.
          - level: Request
            resources:
            - group: "" # core API group
              resources: ["configmaps"]
            # This rule only applies to resources in the "kube-system" namespace.
            # The empty string "" can be used to select non-namespaced resources.
            namespaces: ["kube-system"]

          # Log configmap and secret changes in all other namespaces at the Metadata level.
          - level: Metadata
            resources:
            - group: "" # core API group
              resources: ["secrets", "configmaps"]

          # Log all other resources in core and extensions at the Request level.
          - level: Request
            resources:
            - group: "" # core API group
            - group: "extensions" # Version of group should NOT be included.

          # A catch-all rule to log all other requests at the Metadata level.
          - level: Metadata

1
The log types that are collected. The value for this field can be audit for audit logs, application for application logs, infrastructure for infrastructure logs, or a named input that has been defined for your application.
2
The name of your audit policy.
2.4.4.8. Filtering application logs at input by including the label expressions or a matching label key and values

You can include the application logs based on the label expressions or a matching label key and its values by using the input selector.

Procedure

  1. Add a configuration for a filter to the input spec in the ClusterLogForwarder CR.

    The following example shows how to configure the ClusterLogForwarder CR to include logs based on label expressions or matched label key/values:

    Example ClusterLogForwarder CR

    apiVersion: observability.openshift.io/v1
    kind: ClusterLogForwarder
    # ...
    spec:
      serviceAccount:
        name: <service_account_name>
      inputs:
        - name: mylogs
          application:
            selector:
              matchExpressions:
              - key: env 1
                operator: In 2
                values: ["prod", "qa"] 3
              - key: zone
                operator: NotIn
                values: ["east", "west"]
              matchLabels: 4
                app: one
                name: app1
          type: application
    # ...

    1
    Specifies the label key to match.
    2
    Specifies the operator. Valid values include: In, NotIn, Exists, and DoesNotExist.
    3
    Specifies an array of string values. If the operator value is either Exists or DoesNotExist, the value array must be empty.
    4
    Specifies an exact key or value mapping.
  2. Apply the ClusterLogForwarder CR by running the following command:

    $ oc apply -f <filename>.yaml
2.4.4.9. Configuring content filters to prune log records

When the prune filter is configured, the log collector evaluates log streams according to the filters before forwarding. The collector prunes log records by removing low value fields such as pod annotations.

Procedure

  1. Add a configuration for a filter to the prune spec in the ClusterLogForwarder CR.

    The following example shows how to configure the ClusterLogForwarder CR to prune log records based on field paths:

    Important

    If both are specified, records are pruned based on the notIn array first, which takes precedence over the in array. After records have been pruned by using the notIn array, they are then pruned by using the in array.

    Example ClusterLogForwarder CR

    apiVersion: observability.openshift.io/v1
    kind: ClusterLogForwarder
    metadata:
    # ...
    spec:
      serviceAccount:
        name: <service_account_name>
      filters:
      - name: <filter_name>
        type: prune 1
        prune: 2
          in: [.kubernetes.annotations, .kubernetes.namespace_id] 3
          notIn: [.kubernetes,.log_type,.message,."@timestamp"] 4
      pipelines:
      - name: <pipeline_name> 5
        filterRefs: ["<filter_name>"]
    # ...

    1
    Specify the type of filter. The prune filter prunes log records by configured fields.
    2
    Specify configuration options for applying the prune filter. The in and notIn fields are specified as arrays of dot-delimited field paths, which are paths to fields in log records. These paths can contain alpha-numeric characters and underscores (a-zA-Z0-9_), for example, .kubernetes.namespace_name. If segments contain characters outside of this range, the segment must be in quotes, for example, .kubernetes.labels."foo.bar-bar/baz".
    3
    Optional: Any fields that are specified in this array are removed from the log record.
    4
    Optional: Any fields that are not specified in this array are removed from the log record.
    5
    Specify the pipeline that the prune filter is applied to.
    Note

    The filters exempts the log_type, .log_source, and .message fields.

  2. Apply the ClusterLogForwarder CR by running the following command:

    $ oc apply -f <filename>.yaml

2.4.5. Filtering the audit and infrastructure log inputs by source

You can define the list of audit and infrastructure sources to collect the logs by using the input selector.

Procedure

  1. Add a configuration to define the audit and infrastructure sources in the ClusterLogForwarder CR.

    The following example shows how to configure the ClusterLogForwarder CR to define audit and infrastructure sources:

    Example ClusterLogForwarder CR

    apiVersion: observability.openshift.io/v1
    kind: ClusterLogForwarder
    # ...
    spec:
      serviceAccount:
        name: <service_account_name>
      inputs:
        - name: mylogs1
          type: infrastructure
          infrastructure:
            sources: 1
              - node
        - name: mylogs2
          type: audit
          audit:
            sources: 2
              - kubeAPI
              - openshiftAPI
              - ovn
    # ...

    1
    Specifies the list of infrastructure sources to collect. The valid sources include:
    • node: Journal log from the node
    • container: Logs from the workloads deployed in the namespaces
    2
    Specifies the list of audit sources to collect. The valid sources include:
    • kubeAPI: Logs from the Kubernetes API servers
    • openshiftAPI: Logs from the OpenShift API servers
    • auditd: Logs from a node auditd service
    • ovn: Logs from an open virtual network service
  2. Apply the ClusterLogForwarder CR by running the following command:

    $ oc apply -f <filename>.yaml

2.4.6. Filtering application logs at input by including or excluding the namespace or container name

You can include or exclude the application logs based on the namespace and container name by using the input selector.

Procedure

  1. Add a configuration to include or exclude the namespace and container names in the ClusterLogForwarder CR.

    The following example shows how to configure the ClusterLogForwarder CR to include or exclude namespaces and container names:

    Example ClusterLogForwarder CR

    apiVersion: observability.openshift.io/v1
    kind: ClusterLogForwarder
    # ...
    spec:
      serviceAccount:
        name: <service_account_name>
      inputs:
        - name: mylogs
          application:
            includes:
              - namespace: "my-project" 1
                container: "my-container" 2
            excludes:
              - container: "other-container*" 3
                namespace: "other-namespace" 4
          type: application
    # ...

    1
    Specifies that the logs are only collected from these namespaces.
    2
    Specifies that the logs are only collected from these containers.
    3
    Specifies the pattern of namespaces to ignore when collecting the logs.
    4
    Specifies the set of containers to ignore when collecting the logs.
    Note

    The excludes field takes precedence over the includes field.

  2. Apply the ClusterLogForwarder CR by running the following command:

    $ oc apply -f <filename>.yaml

2.5. Storing logs with LokiStack

You can configure a LokiStack CR to store application, audit, and infrastructure-related logs.

Loki is a horizontally scalable, highly available, multi-tenant log aggregation system offered as a GA log store for logging for Red Hat OpenShift that can be visualized with the OpenShift Observability UI. The Loki configuration provided by OpenShift Logging is a short-term log store designed to enable users to perform fast troubleshooting with the collected logs. For that purpose, the logging for Red Hat OpenShift configuration of Loki has short-term storage, and is optimized for very recent queries.

Important

For long-term storage or queries over a long time period, users should look to log stores external to their cluster. Loki sizing is only tested and supported for short term storage, for a maximum of 30 days.

2.5.1. Prerequisites

  • You have installed the Loki Operator by using the CLI or web console.
  • You have a serviceAccount in the same namespace in which you create the ClusterLogForwarder.
  • The serviceAccount is assigned collect-audit-logs, collect-application-logs, and collect-infrastructure-logs cluster roles.

2.5.2. Core Setup and Configuration

Role-based access controls, basic monitoring, and pod placement to deploy Loki.

2.5.3. Loki deployment sizing

Sizing for Loki follows the format of 1x.<size> where the value 1x is number of instances and <size> specifies performance capabilities.

The 1x.pico configuration defines a single Loki deployment with minimal resource and limit requirements, offering high availability (HA) support for all Loki components. This configuration is suited for deployments that do not require a single replication factor or auto-compaction.

Disk requests are similar across size configurations, allowing customers to test different sizes to determine the best fit for their deployment needs.

Important

It is not possible to change the number 1x for the deployment size.

Table 2.2. Loki sizing
 1x.demo1x.pico [6.1+ only]1x.extra-small1x.small1x.medium

Data transfer

Demo use only

50GB/day

100GB/day

500GB/day

2TB/day

Queries per second (QPS)

Demo use only

1-25 QPS at 200ms

1-25 QPS at 200ms

25-50 QPS at 200ms

25-75 QPS at 200ms

Replication factor

None

2

2

2

2

Total CPU requests

None

7 vCPUs

14 vCPUs

34 vCPUs

54 vCPUs

Total CPU requests if using the ruler

None

8 vCPUs

16 vCPUs

42 vCPUs

70 vCPUs

Total memory requests

None

17Gi

31Gi

67Gi

139Gi

Total memory requests if using the ruler

None

18Gi

35Gi

83Gi

171Gi

Total disk requests

40Gi

590Gi

430Gi

430Gi

590Gi

Total disk requests if using the ruler

80Gi

910Gi

750Gi

750Gi

910Gi

2.5.4. Authorizing LokiStack rules RBAC permissions

Administrators can allow users to create and manage their own alerting and recording rules by binding cluster roles to usernames. Cluster roles are defined as ClusterRole objects that contain necessary role-based access control (RBAC) permissions for users.

The following cluster roles for alerting and recording rules are available for LokiStack:

Rule nameDescription

alertingrules.loki.grafana.com-v1-admin

Users with this role have administrative-level access to manage alerting rules. This cluster role grants permissions to create, read, update, delete, list, and watch AlertingRule resources within the loki.grafana.com/v1 API group.

alertingrules.loki.grafana.com-v1-crdview

Users with this role can view the definitions of Custom Resource Definitions (CRDs) related to AlertingRule resources within the loki.grafana.com/v1 API group, but do not have permissions for modifying or managing these resources.

alertingrules.loki.grafana.com-v1-edit

Users with this role have permission to create, update, and delete AlertingRule resources.

alertingrules.loki.grafana.com-v1-view

Users with this role can read AlertingRule resources within the loki.grafana.com/v1 API group. They can inspect configurations, labels, and annotations for existing alerting rules but cannot make any modifications to them.

recordingrules.loki.grafana.com-v1-admin

Users with this role have administrative-level access to manage recording rules. This cluster role grants permissions to create, read, update, delete, list, and watch RecordingRule resources within the loki.grafana.com/v1 API group.

recordingrules.loki.grafana.com-v1-crdview

Users with this role can view the definitions of Custom Resource Definitions (CRDs) related to RecordingRule resources within the loki.grafana.com/v1 API group, but do not have permissions for modifying or managing these resources.

recordingrules.loki.grafana.com-v1-edit

Users with this role have permission to create, update, and delete RecordingRule resources.

recordingrules.loki.grafana.com-v1-view

Users with this role can read RecordingRule resources within the loki.grafana.com/v1 API group. They can inspect configurations, labels, and annotations for existing alerting rules but cannot make any modifications to them.

2.5.4.1. Examples

To apply cluster roles for a user, you must bind an existing cluster role to a specific username.

Cluster roles can be cluster or namespace scoped, depending on which type of role binding you use. When a RoleBinding object is used, as when using the oc adm policy add-role-to-user command, the cluster role only applies to the specified namespace. When a ClusterRoleBinding object is used, as when using the oc adm policy add-cluster-role-to-user command, the cluster role applies to all namespaces in the cluster.

The following example command gives the specified user create, read, update and delete (CRUD) permissions for alerting rules in a specific namespace in the cluster:

Example cluster role binding command for alerting rule CRUD permissions in a specific namespace

$ oc adm policy add-role-to-user alertingrules.loki.grafana.com-v1-admin -n <namespace> <username>

The following command gives the specified user administrator permissions for alerting rules in all namespaces:

Example cluster role binding command for administrator permissions

$ oc adm policy add-cluster-role-to-user alertingrules.loki.grafana.com-v1-admin <username>

2.5.5. Creating a log-based alerting rule with Loki

The AlertingRule CR contains a set of specifications and webhook validation definitions to declare groups of alerting rules for a single LokiStack instance. In addition, the webhook validation definition provides support for rule validation conditions:

  • If an AlertingRule CR includes an invalid interval period, it is an invalid alerting rule
  • If an AlertingRule CR includes an invalid for period, it is an invalid alerting rule.
  • If an AlertingRule CR includes an invalid LogQL expr, it is an invalid alerting rule.
  • If an AlertingRule CR includes two groups with the same name, it is an invalid alerting rule.
  • If none of the above applies, an alerting rule is considered valid.
Table 2.3. AlertingRule definitions
Tenant typeValid namespaces for AlertingRule CRs

application

<your_application_namespace>

audit

openshift-logging

infrastructure

openshift-/*, kube-/\*, default

Procedure

  1. Create an AlertingRule custom resource (CR):

    Example infrastructure AlertingRule CR

      apiVersion: loki.grafana.com/v1
      kind: AlertingRule
      metadata:
        name: loki-operator-alerts
        namespace: openshift-operators-redhat 1
        labels: 2
          openshift.io/<label_name>: "true"
      spec:
        tenantID: "infrastructure" 3
        groups:
          - name: LokiOperatorHighReconciliationError
            rules:
              - alert: HighPercentageError
                expr: | 4
                  sum(rate({kubernetes_namespace_name="openshift-operators-redhat", kubernetes_pod_name=~"loki-operator-controller-manager.*"} |= "error" [1m])) by (job)
                    /
                  sum(rate({kubernetes_namespace_name="openshift-operators-redhat", kubernetes_pod_name=~"loki-operator-controller-manager.*"}[1m])) by (job)
                    > 0.01
                for: 10s
                labels:
                  severity: critical 5
                annotations:
                  summary: High Loki Operator Reconciliation Errors 6
                  description: High Loki Operator Reconciliation Errors 7

    1
    The namespace where this AlertingRule CR is created must have a label matching the LokiStack spec.rules.namespaceSelector definition.
    2
    The labels block must match the LokiStack spec.rules.selector definition.
    3
    AlertingRule CRs for infrastructure tenants are only supported in the openshift-*, kube-\*, or default namespaces.
    4
    The value for kubernetes_namespace_name: must match the value for metadata.namespace.
    5
    The value of this mandatory field must be critical, warning, or info.
    6
    This field is mandatory.
    7
    This field is mandatory.

    Example application AlertingRule CR

      apiVersion: loki.grafana.com/v1
      kind: AlertingRule
      metadata:
        name: app-user-workload
        namespace: app-ns 1
        labels: 2
          openshift.io/<label_name>: "true"
      spec:
        tenantID: "application"
        groups:
          - name: AppUserWorkloadHighError
            rules:
              - alert:
                expr: | 3
                  sum(rate({kubernetes_namespace_name="app-ns", kubernetes_pod_name=~"podName.*"} |= "error" [1m])) by (job)
                for: 10s
                labels:
                  severity: critical 4
                annotations:
                  summary:  5
                  description:  6

    1
    The namespace where this AlertingRule CR is created must have a label matching the LokiStack spec.rules.namespaceSelector definition.
    2
    The labels block must match the LokiStack spec.rules.selector definition.
    3
    Value for kubernetes_namespace_name: must match the value for metadata.namespace.
    4
    The value of this mandatory field must be critical, warning, or info.
    5
    The value of this mandatory field is a summary of the rule.
    6
    The value of this mandatory field is a detailed description of the rule.
  2. Apply the AlertingRule CR:

    $ oc apply -f <filename>.yaml

2.5.6. Configuring Loki to tolerate memberlist creation failure

In an OpenShift Container Platform cluster, administrators generally use a non-private IP network range. As a result, the LokiStack memberlist configuration fails because, by default, it only uses private IP networks.

As an administrator, you can select the pod network for the memberlist configuration. You can modify the LokiStack custom resource (CR) to use the podIP address in the hashRing spec. To configure the LokiStack CR, use the following command:

$ oc patch LokiStack logging-loki -n openshift-logging  --type=merge -p '{"spec": {"hashRing":{"memberlist":{"instanceAddrType":"podIP"},"type":"memberlist"}}}'

Example LokiStack to include podIP

apiVersion: loki.grafana.com/v1
kind: LokiStack
metadata:
  name: logging-loki
  namespace: openshift-logging
spec:
# ...
  hashRing:
    type: memberlist
    memberlist:
      instanceAddrType: podIP
# ...

2.5.7. Enabling stream-based retention with Loki

You can configure retention policies based on log streams. Rules for these may be set globally, per-tenant, or both. If you configure both, tenant rules apply before global rules.

Important

If there is no retention period defined on the s3 bucket or in the LokiStack custom resource (CR), then the logs are not pruned and they stay in the s3 bucket forever, which might fill up the s3 storage.

Note

Schema v13 is recommended.

Procedure

  1. Create a LokiStack CR:

    • Enable stream-based retention globally as shown in the following example:

      Example global stream-based retention for AWS

      apiVersion: loki.grafana.com/v1
      kind: LokiStack
      metadata:
        name: logging-loki
        namespace: openshift-logging
      spec:
        limits:
         global: 1
            retention: 2
              days: 20
              streams:
              - days: 4
                priority: 1
                selector: '{kubernetes_namespace_name=~"test.+"}' 3
              - days: 1
                priority: 1
                selector: '{log_type="infrastructure"}'
        managementState: Managed
        replicationFactor: 1
        size: 1x.small
        storage:
          schemas:
          - effectiveDate: "2020-10-11"
            version: v13
          secret:
            name: logging-loki-s3
            type: aws
        storageClassName: gp3-csi
        tenants:
          mode: openshift-logging

      1
      Sets retention policy for all log streams. Note: This field does not impact the retention period for stored logs in object storage.
      2
      Retention is enabled in the cluster when this block is added to the CR.
      3
      Contains the LogQL query used to define the log stream.spec: limits:
    • Enable stream-based retention per-tenant basis as shown in the following example:

      Example per-tenant stream-based retention for AWS

      apiVersion: loki.grafana.com/v1
      kind: LokiStack
      metadata:
        name: logging-loki
        namespace: openshift-logging
      spec:
        limits:
          global:
            retention:
              days: 20
          tenants: 1
            application:
              retention:
                days: 1
                streams:
                  - days: 4
                    selector: '{kubernetes_namespace_name=~"test.+"}' 2
            infrastructure:
              retention:
                days: 5
                streams:
                  - days: 1
                    selector: '{kubernetes_namespace_name=~"openshift-cluster.+"}'
        managementState: Managed
        replicationFactor: 1
        size: 1x.small
        storage:
          schemas:
          - effectiveDate: "2020-10-11"
            version: v13
          secret:
            name: logging-loki-s3
            type: aws
        storageClassName: gp3-csi
        tenants:
          mode: openshift-logging

      1
      Sets retention policy by tenant. Valid tenant types are application, audit, and infrastructure.
      2
      Contains the LogQL query used to define the log stream.
  2. Apply the LokiStack CR:

    $ oc apply -f <filename>.yaml

2.5.8. Loki pod placement

You can control which nodes the Loki pods run on, and prevent other workloads from using those nodes, by using tolerations or node selectors on the pods.

You can apply tolerations to the log store pods with the LokiStack custom resource (CR) and apply taints to a node with the node specification. A taint on a node is a key:value pair that instructs the node to repel all pods that do not allow the taint. Using a specific key:value pair that is not on other pods ensures that only the log store pods can run on that node.

Example LokiStack with node selectors

apiVersion: loki.grafana.com/v1
kind: LokiStack
metadata:
  name: logging-loki
  namespace: openshift-logging
spec:
# ...
  template:
    compactor: 1
      nodeSelector:
        node-role.kubernetes.io/infra: "" 2
    distributor:
      nodeSelector:
        node-role.kubernetes.io/infra: ""
    gateway:
      nodeSelector:
        node-role.kubernetes.io/infra: ""
    indexGateway:
      nodeSelector:
        node-role.kubernetes.io/infra: ""
    ingester:
      nodeSelector:
        node-role.kubernetes.io/infra: ""
    querier:
      nodeSelector:
        node-role.kubernetes.io/infra: ""
    queryFrontend:
      nodeSelector:
        node-role.kubernetes.io/infra: ""
    ruler:
      nodeSelector:
        node-role.kubernetes.io/infra: ""
# ...

1
Specifies the component pod type that applies to the node selector.
2
Specifies the pods that are moved to nodes containing the defined label.

Example LokiStack CR with node selectors and tolerations

apiVersion: loki.grafana.com/v1
kind: LokiStack
metadata:
  name: logging-loki
  namespace: openshift-logging
spec:
# ...
  template:
    compactor:
      nodeSelector:
        node-role.kubernetes.io/infra: ""
      tolerations:
      - effect: NoSchedule
        key: node-role.kubernetes.io/infra
        value: reserved
      - effect: NoExecute
        key: node-role.kubernetes.io/infra
        value: reserved
    distributor:
      nodeSelector:
        node-role.kubernetes.io/infra: ""
      tolerations:
      - effect: NoSchedule
        key: node-role.kubernetes.io/infra
        value: reserved
      - effect: NoExecute
        key: node-role.kubernetes.io/infra
        value: reserved
      nodeSelector:
        node-role.kubernetes.io/infra: ""
      tolerations:
      - effect: NoSchedule
        key: node-role.kubernetes.io/infra
        value: reserved
      - effect: NoExecute
        key: node-role.kubernetes.io/infra
        value: reserved
    indexGateway:
      nodeSelector:
        node-role.kubernetes.io/infra: ""
      tolerations:
      - effect: NoSchedule
        key: node-role.kubernetes.io/infra
        value: reserved
      - effect: NoExecute
        key: node-role.kubernetes.io/infra
        value: reserved
    ingester:
      nodeSelector:
        node-role.kubernetes.io/infra: ""
      tolerations:
      - effect: NoSchedule
        key: node-role.kubernetes.io/infra
        value: reserved
      - effect: NoExecute
        key: node-role.kubernetes.io/infra
        value: reserved
    querier:
      nodeSelector:
        node-role.kubernetes.io/infra: ""
      tolerations:
      - effect: NoSchedule
        key: node-role.kubernetes.io/infra
        value: reserved
      - effect: NoExecute
        key: node-role.kubernetes.io/infra
        value: reserved
    queryFrontend:
      nodeSelector:
        node-role.kubernetes.io/infra: ""
      tolerations:
      - effect: NoSchedule
        key: node-role.kubernetes.io/infra
        value: reserved
      - effect: NoExecute
        key: node-role.kubernetes.io/infra
        value: reserved
    ruler:
      nodeSelector:
        node-role.kubernetes.io/infra: ""
      tolerations:
      - effect: NoSchedule
        key: node-role.kubernetes.io/infra
        value: reserved
      - effect: NoExecute
        key: node-role.kubernetes.io/infra
        value: reserved
    gateway:
      nodeSelector:
        node-role.kubernetes.io/infra: ""
      tolerations:
      - effect: NoSchedule
        key: node-role.kubernetes.io/infra
        value: reserved
      - effect: NoExecute
        key: node-role.kubernetes.io/infra
        value: reserved
# ...

To configure the nodeSelector and tolerations fields of the LokiStack (CR), you can use the oc explain command to view the description and fields for a particular resource:

$ oc explain lokistack.spec.template

Example output

KIND:     LokiStack
VERSION:  loki.grafana.com/v1

RESOURCE: template <Object>

DESCRIPTION:
     Template defines the resource/limits/tolerations/nodeselectors per
     component

FIELDS:
   compactor	<Object>
     Compactor defines the compaction component spec.

   distributor	<Object>
     Distributor defines the distributor component spec.
...

For more detailed information, you can add a specific field:

$ oc explain lokistack.spec.template.compactor

Example output

KIND:     LokiStack
VERSION:  loki.grafana.com/v1

RESOURCE: compactor <Object>

DESCRIPTION:
     Compactor defines the compaction component spec.

FIELDS:
   nodeSelector	<map[string]string>
     NodeSelector defines the labels required by a node to schedule the
     component onto it.
...

2.5.8.1. Enhanced Reliability and Performance

Configurations to ensure Loki’s reliability and efficiency in production.

2.5.8.2. Enabling authentication to cloud-based log stores using short-lived tokens

Workload identity federation enables authentication to cloud-based log stores using short-lived tokens.

Procedure

  • Use one of the following options to enable authentication:

    • If you use the OpenShift Container Platform web console to install the Loki Operator, clusters that use short-lived tokens are automatically detected. You are prompted to create roles and supply the data required for the Loki Operator to create a CredentialsRequest object, which populates a secret.
    • If you use the OpenShift CLI (oc) to install the Loki Operator, you must manually create a Subscription object using the appropriate template for your storage provider, as shown in the following examples. This authentication strategy is only supported for the storage providers indicated.

      Example Azure sample subscription

      apiVersion: operators.coreos.com/v1alpha1
      kind: Subscription
      metadata:
        name: loki-operator
        namespace: openshift-operators-redhat
      spec:
        channel: "stable-6.0"
        installPlanApproval: Manual
        name: loki-operator
        source: redhat-operators
        sourceNamespace: openshift-marketplace
        config:
          env:
            - name: CLIENTID
              value: <your_client_id>
            - name: TENANTID
              value: <your_tenant_id>
            - name: SUBSCRIPTIONID
              value: <your_subscription_id>
            - name: REGION
              value: <your_region>

      Example AWS sample subscription

      apiVersion: operators.coreos.com/v1alpha1
      kind: Subscription
      metadata:
        name: loki-operator
        namespace: openshift-operators-redhat
      spec:
        channel: "stable-6.0"
        installPlanApproval: Manual
        name: loki-operator
        source: redhat-operators
        sourceNamespace: openshift-marketplace
        config:
          env:
          - name: ROLEARN
            value: <role_ARN>

2.5.8.3. Configuring Loki to tolerate node failure

The Loki Operator supports setting pod anti-affinity rules to request that pods of the same component are scheduled on different available nodes in the cluster.

Affinity is a property of pods that controls the nodes on which they prefer to be scheduled. Anti-affinity is a property of pods that prevents a pod from being scheduled on a node.

In OpenShift Container Platform, pod affinity and pod anti-affinity allow you to constrain which nodes your pod is eligible to be scheduled on based on the key-value labels on other pods.

The Operator sets default, preferred podAntiAffinity rules for all Loki components, which includes the compactor, distributor, gateway, indexGateway, ingester, querier, queryFrontend, and ruler components.

You can override the preferred podAntiAffinity settings for Loki components by configuring required settings in the requiredDuringSchedulingIgnoredDuringExecution field:

Example user settings for the ingester component

apiVersion: loki.grafana.com/v1
kind: LokiStack
metadata:
  name: logging-loki
  namespace: openshift-logging
spec:
# ...
  template:
    ingester:
      podAntiAffinity:
      # ...
        requiredDuringSchedulingIgnoredDuringExecution: 1
        - labelSelector:
            matchLabels: 2
              app.kubernetes.io/component: ingester
          topologyKey: kubernetes.io/hostname
# ...

1
The stanza to define a required rule.
2
The key-value pair (label) that must be matched to apply the rule.
2.5.8.4. LokiStack behavior during cluster restarts

When an OpenShift Container Platform cluster is restarted, LokiStack ingestion and the query path continue to operate within the available CPU and memory resources available for the node. This means that there is no downtime for the LokiStack during OpenShift Container Platform cluster updates. This behavior is achieved by using PodDisruptionBudget resources. The Loki Operator provisions PodDisruptionBudget resources for Loki, which determine the minimum number of pods that must be available per component to ensure normal operations under certain conditions.

2.5.8.5. Advanced Deployment and Scalability

Specialized configurations for high availability, scalability, and error handling.

2.5.8.6. Zone aware data replication

The Loki Operator offers support for zone-aware data replication through pod topology spread constraints. Enabling this feature enhances reliability and safeguards against log loss in the event of a single zone failure. When configuring the deployment size as 1x.extra-small, 1x.small, or 1x.medium, the replication.factor field is automatically set to 2.

To ensure proper replication, you need to have at least as many availability zones as the replication factor specifies. While it is possible to have more availability zones than the replication factor, having fewer zones can lead to write failures. Each zone should host an equal number of instances for optimal operation.

Example LokiStack CR with zone replication enabled

apiVersion: loki.grafana.com/v1
kind: LokiStack
metadata:
 name: logging-loki
 namespace: openshift-logging
spec:
 replicationFactor: 2 1
 replication:
   factor: 2 2
   zones:
   -  maxSkew: 1 3
      topologyKey: topology.kubernetes.io/zone 4

1
Deprecated field, values entered are overwritten by replication.factor.
2
This value is automatically set when deployment size is selected at setup.
3
The maximum difference in number of pods between any two topology domains. The default is 1, and you cannot specify a value of 0.
4
Defines zones in the form of a topology key that corresponds to a node label.
2.5.8.7. Recovering Loki pods from failed zones

In OpenShift Container Platform a zone failure happens when specific availability zone resources become inaccessible. Availability zones are isolated areas within a cloud provider’s data center, aimed at enhancing redundancy and fault tolerance. If your OpenShift Container Platform cluster is not configured to handle this, a zone failure can lead to service or data loss.

Loki pods are part of a StatefulSet, and they come with Persistent Volume Claims (PVCs) provisioned by a StorageClass object. Each Loki pod and its PVCs reside in the same zone. When a zone failure occurs in a cluster, the StatefulSet controller automatically attempts to recover the affected pods in the failed zone.

Warning

The following procedure will delete the PVCs in the failed zone, and all data contained therein. To avoid complete data loss the replication factor field of the LokiStack CR should always be set to a value greater than 1 to ensure that Loki is replicating.

Prerequisites

  • Verify your LokiStack CR has a replication factor greater than 1.
  • Zone failure detected by the control plane, and nodes in the failed zone are marked by cloud provider integration.

The StatefulSet controller automatically attempts to reschedule pods in a failed zone. Because the associated PVCs are also in the failed zone, automatic rescheduling to a different zone does not work. You must manually delete the PVCs in the failed zone to allow successful re-creation of the stateful Loki Pod and its provisioned PVC in the new zone.

Procedure

  1. List the pods in Pending status by running the following command:

    $ oc get pods --field-selector status.phase==Pending -n openshift-logging

    Example oc get pods output

    NAME                           READY   STATUS    RESTARTS   AGE 1
    logging-loki-index-gateway-1   0/1     Pending   0          17m
    logging-loki-ingester-1        0/1     Pending   0          16m
    logging-loki-ruler-1           0/1     Pending   0          16m

    1
    These pods are in Pending status because their corresponding PVCs are in the failed zone.
  2. List the PVCs in Pending status by running the following command:

    $ oc get pvc -o=json -n openshift-logging | jq '.items[] | select(.status.phase == "Pending") | .metadata.name' -r

    Example oc get pvc output

    storage-logging-loki-index-gateway-1
    storage-logging-loki-ingester-1
    wal-logging-loki-ingester-1
    storage-logging-loki-ruler-1
    wal-logging-loki-ruler-1

  3. Delete the PVC(s) for a pod by running the following command:

    $ oc delete pvc <pvc_name>  -n openshift-logging
  4. Delete the pod(s) by running the following command:

    $ oc delete pod <pod_name>  -n openshift-logging

    Once these objects have been successfully deleted, they should automatically be rescheduled in an available zone.

2.5.8.7.1. Troubleshooting PVC in a terminating state

The PVCs might hang in the terminating state without being deleted, if PVC metadata finalizers are set to kubernetes.io/pv-protection. Removing the finalizers should allow the PVCs to delete successfully.

  • Remove the finalizer for each PVC by running the command below, then retry deletion.

    $ oc patch pvc <pvc_name> -p '{"metadata":{"finalizers":null}}' -n openshift-logging
2.5.8.8. Troubleshooting Loki rate limit errors

If the Log Forwarder API forwards a large block of messages that exceeds the rate limit to Loki, Loki generates rate limit (429) errors.

These errors can occur during normal operation. For example, when adding the logging to a cluster that already has some logs, rate limit errors might occur while the logging tries to ingest all of the existing log entries. In this case, if the rate of addition of new logs is less than the total rate limit, the historical data is eventually ingested, and the rate limit errors are resolved without requiring user intervention.

In cases where the rate limit errors continue to occur, you can fix the issue by modifying the LokiStack custom resource (CR).

Important

The LokiStack CR is not available on Grafana-hosted Loki. This topic does not apply to Grafana-hosted Loki servers.

Conditions

  • The Log Forwarder API is configured to forward logs to Loki.
  • Your system sends a block of messages that is larger than 2 MB to Loki. For example:

    "values":[["1630410392689800468","{\"kind\":\"Event\",\"apiVersion\":\
    .......
    ......
    ......
    ......
    \"received_at\":\"2021-08-31T11:46:32.800278+00:00\",\"version\":\"1.7.4 1.6.0\"}},\"@timestamp\":\"2021-08-31T11:46:32.799692+00:00\",\"viaq_index_name\":\"audit-write\",\"viaq_msg_id\":\"MzFjYjJkZjItNjY0MC00YWU4LWIwMTEtNGNmM2E5ZmViMGU4\",\"log_type\":\"audit\"}"]]}]}
  • After you enter oc logs -n openshift-logging -l component=collector, the collector logs in your cluster show a line containing one of the following error messages:

    429 Too Many Requests Ingestion rate limit exceeded

    Example Vector error message

    2023-08-25T16:08:49.301780Z  WARN sink{component_kind="sink" component_id=default_loki_infra component_type=loki component_name=default_loki_infra}: vector::sinks::util::retries: Retrying after error. error=Server responded with an error: 429 Too Many Requests internal_log_rate_limit=true

    Example Fluentd error message

    2023-08-30 14:52:15 +0000 [warn]: [default_loki_infra] failed to flush the buffer. retry_times=2 next_retry_time=2023-08-30 14:52:19 +0000 chunk="604251225bf5378ed1567231a1c03b8b" error_class=Fluent::Plugin::LokiOutput::LogPostError error="429 Too Many Requests Ingestion rate limit exceeded for user infrastructure (limit: 4194304 bytes/sec) while attempting to ingest '4082' lines totaling '7820025' bytes, reduce log volume or contact your Loki administrator to see if the limit can be increased\n"

    The error is also visible on the receiving end. For example, in the LokiStack ingester pod:

    Example Loki ingester error message

    level=warn ts=2023-08-30T14:57:34.155592243Z caller=grpc_logging.go:43 duration=1.434942ms method=/logproto.Pusher/Push err="rpc error: code = Code(429) desc = entry with timestamp 2023-08-30 14:57:32.012778399 +0000 UTC ignored, reason: 'Per stream rate limit exceeded (limit: 3MB/sec) while attempting to ingest for stream

Procedure

  • Update the ingestionBurstSize and ingestionRate fields in the LokiStack CR:

    apiVersion: loki.grafana.com/v1
    kind: LokiStack
    metadata:
      name: logging-loki
      namespace: openshift-logging
    spec:
      limits:
        global:
          ingestion:
            ingestionBurstSize: 16 1
            ingestionRate: 8 2
    # ...
    1
    The ingestionBurstSize field defines the maximum local rate-limited sample size per distributor replica in MB. This value is a hard limit. Set this value to at least the maximum logs size expected in a single push request. Single requests that are larger than the ingestionBurstSize value are not permitted.
    2
    The ingestionRate field is a soft limit on the maximum amount of ingested samples per second in MB. Rate limit errors occur if the rate of logs exceeds the limit, but the collector retries sending the logs. As long as the total average is lower than the limit, the system recovers and errors are resolved without user intervention.

2.6. Visualization for logging

Visualization for logging is provided by deploying the Logging UI Plugin of the Cluster Observability Operator, which requires Operator installation.

Important

Until the approaching General Availability (GA) release of the Cluster Observability Operator (COO), which is currently in Technology Preview (TP), Red Hat provides support to customers who are using Logging 6.0 or later with the COO for its Logging UI Plugin on OpenShift Container Platform 4.14 or later. This support exception is temporary as the COO includes several independent features, some of which are still TP features, but the Logging UI Plugin is ready for GA.

Chapter 3. Logging 6.1

3.1. Logging 6.1

3.1.1. Logging 6.1.1 Release Notes

This release includes Logging for Red Hat OpenShift Bug Fix Release 6.1.1.

3.1.1.1. New Features and Enhancements
  • With this update, the Loki Operator supports configuring the workload identity federation on the Google Cloud Platform (GCP) by using the Cluster Credential Operator (CCO) in OpenShift Container Platform 4.17 or later. (LOG-6420)
3.1.1.2. Bug Fixes
  • Before this update, the collector was discarding longer audit log messages with the following error message: Internal log [Found line that exceeds max_line_bytes; discarding.]. With this update, the discarding of longer audit messages is avoided by increasing the audit configuration thresholds: The maximum line size, max_line_bytes, is 3145728 bytes. The maximum number of bytes read during a read cycle, max_read_bytes, is 262144 bytes. (LOG-6379)
  • Before this update, an input receiver service was repeatedly created and deleted, causing issues with mounting the TLS secrets. With this update, the service is created once and only deleted if it is not defined in the ClusterLogForwarder custom resource. (LOG-6383)
  • Before this update, pipeline validation might have entered an infinite loop if a name was a substring of another name. With this update, stricter name equality checks prevent the infinite loop. (LOG-6405)
  • Before this update, the collector alerting rules included the summary and message fields. With this update, the collector alerting rules include the summary and description fields. (LOG-6407)
  • Before this update, setting up the custom audit inputs in the ClusterLogForwarder custom resource with configured LokiStack output caused errors due to the nil pointer dereference. With this update, the Operator performs the nil checks, preventing such errors. (LOG-6449)
  • Before this update, the ValidLokistackOTLPOutputs condition appeared in the status of the ClusterLogForwarder custom resource even when the output type is not LokiStack. With this update, the ValidLokistackOTLPOutputs condition is removed, and the validation messages for the existing output conditions are corrected. (LOG-6469)
  • Before this update, the collector did not correctly mount the /var/log/oauth-server/ path, which prevented the collection of the audit logs. With this update, the volume mount is added, and the audit logs are collected as expected. (LOG-6484)
  • Before this update, the must-gather script of the Red Hat OpenShift Logging Operator might have failed to gather the LokiStack data. With this update, the must-gather script is fixed, and the LokiStack data is gathered reliably. (LOG-6498)
  • Before this update, the collector did not correctly mount the oauth-apiserver audit log file. As a result, such audit logs were not collected. With this update, the volume mount is correctly mounted, and the logs are collected as expected. (LOG-6533)
3.1.1.3. CVEs

3.1.2. Logging 6.1.0 Release Notes

This release includes Logging for Red Hat OpenShift Bug Fix Release 6.1.0.

3.1.2.1. New Features and Enhancements
3.1.2.1.1. Log Collection
  • This enhancement adds the source iostream to the attributes sent from collected container logs. The value is set to either stdout or stderr based on how the collector received it. (LOG-5292)
  • With this update, the default memory limit for the collector increases from 1024 Mi to 2048 Mi. Users should adjust resource limits based on their cluster’s specific needs and specifications. (LOG-6072)
  • With this update, users can now set the syslog output delivery mode of the ClusterLogForwarder CR to either AtLeastOnce or AtMostOnce. (LOG-6355)
3.1.2.1.2. Log Storage
  • With this update, the new 1x.pico LokiStack size supports clusters with fewer workloads and lower log volumes (up to 50GB/day). (LOG-5939)
3.1.2.2. Technology Preview
Important

The OpenTelemetry Protocol (OTLP) output log forwarder is a Technology Preview feature only. Technology Preview features are not supported with Red Hat production service level agreements (SLAs) and might not be functionally complete. Red Hat does not recommend using them in production. These features provide early access to upcoming product features, enabling customers to test functionality and provide feedback during the development process.

For more information about the support scope of Red Hat Technology Preview features, see Technology Preview Features Support Scope.

  • With this update, OpenTelemetry logs can now be forwarded using the OTel (OpenTelemetry) data model to a Red Hat Managed LokiStack instance. To enable this feature, add the observability.openshift.io/tech-preview-otlp-output: "enabled" annotation to your ClusterLogForwarder configuration. For additional configuration information, see OTLP Forwarding.
  • With this update, a dataModel field has been added to the lokiStack output specification. Set the dataModel to Otel to configure log forwarding using the OpenTelemetry data format. The default is set to Viaq. For information about data mapping see OTLP Specification.
3.1.2.3. Bug Fixes

None.

3.1.2.4. CVEs

3.2. Logging 6.1

context: logging-6x-6.1

The ClusterLogForwarder custom resource (CR) is the central configuration point for log collection and forwarding.

3.2.1. Inputs and outputs

Inputs specify the sources of logs to be forwarded. Logging provides built-in input types: application, receiver, infrastructure, and audit, which select logs from different parts of your cluster. You can also define custom inputs based on namespaces or pod labels to fine-tune log selection.

Outputs define the destinations where logs are sent. Each output type has its own set of configuration options, allowing you to customize the behavior and authentication settings.

3.2.2. Receiver input type

The receiver input type enables the Logging system to accept logs from external sources. It supports two formats for receiving logs: http and syslog.

The ReceiverSpec defines the configuration for a receiver input.

3.2.3. Pipelines and filters

Pipelines determine the flow of logs from inputs to outputs. A pipeline consists of one or more input refs, output refs, and optional filter refs. Filters can be used to transform or drop log messages within a pipeline. The order of filters matters, as they are applied sequentially, and earlier filters can prevent log messages from reaching later stages.

3.2.4. Operator behavior

The Cluster Logging Operator manages the deployment and configuration of the collector based on the managementState field of the ClusterLogForwarder resource:

  • When set to Managed (default), the operator actively manages the logging resources to match the configuration defined in the spec.
  • When set to Unmanaged, the operator does not take any action, allowing you to manually manage the logging components.

3.2.5. Validation

Logging includes extensive validation rules and default values to ensure a smooth and error-free configuration experience. The ClusterLogForwarder resource enforces validation checks on required fields, dependencies between fields, and the format of input values. Default values are provided for certain fields, reducing the need for explicit configuration in common scenarios.

3.2.6. Quick start

OpenShift Logging supports two data models:

  • ViaQ (General Availability)
  • OpenTelemetry (Technology Preview)

You can select either of these data models based on your requirement by configuring the lokiStack.dataModel field in the ClusterLogForwarder. ViaQ is the default data model when forwarding logs to LokiStack.

Note

In future releases of OpenShift Logging, the default data model will change from ViaQ to OpenTelemetry.

3.2.6.1. Quick start with ViaQ

To use the default ViaQ data model, follow these steps:

Prerequisites

  • Cluster administrator permissions

Procedure

  1. Install the Red Hat OpenShift Logging Operator, Loki Operator, and Cluster Observability Operator (COO) from OperatorHub.
  2. Create a LokiStack custom resource (CR) in the openshift-logging namespace:

    apiVersion: loki.grafana.com/v1
    kind: LokiStack
    metadata:
      name: logging-loki
      namespace: openshift-logging
    spec:
      managementState: Managed
      size: 1x.extra-small
      storage:
        schemas:
        - effectiveDate: '2024-10-01'
          version: v13
        secret:
          name: logging-loki-s3
          type: s3
      storageClassName: gp3-csi
      tenants:
        mode: openshift-logging
    Note

    Ensure that the logging-loki-s3 secret is created beforehand. The contents of this secret vary depending on the object storage in use. For more information, see Secrets and TLS Configuration.

  3. Create a service account for the collector:

    $ oc create sa collector -n openshift-logging
  4. Allow the collector’s service account to write data to the LokiStack CR:

    $ oc adm policy add-cluster-role-to-user logging-collector-logs-writer -z collector
    Note

    The ClusterRole resource is created automatically during the Cluster Logging Operator installation and does not need to be created manually.

  5. Allow the collector’s service account to collect logs:

    $ oc project openshift-logging
    $ oc adm policy add-cluster-role-to-user collect-application-logs -z collector
    $ oc adm policy add-cluster-role-to-user collect-audit-logs -z collector
    $ oc adm policy add-cluster-role-to-user collect-infrastructure-logs -z collector
    Note

    The example binds the collector to all three roles (application, infrastructure, and audit), but by default, only application and infrastructure logs are collected. To collect audit logs, update your ClusterLogForwarder configuration to include them. Assign roles based on the specific log types required for your environment.

  6. Create a UIPlugin CR to enable the Log section in the Observe tab:

    apiVersion: observability.openshift.io/v1alpha1
    kind: UIPlugin
    metadata:
      name: logging
    spec:
      type: Logging
      logging:
        lokiStack:
          name: logging-loki
  7. Create a ClusterLogForwarder CR to configure log forwarding:

    apiVersion: observability.openshift.io/v1
    kind: ClusterLogForwarder
    metadata:
      name: collector
      namespace: openshift-logging
    spec:
      serviceAccount:
        name: collector
      outputs:
      - name: default-lokistack
        type: lokiStack
        lokiStack:
          authentication:
            token:
              from: serviceAccount
          target:
            name: logging-loki
            namespace: openshift-logging
        tls:
          ca:
            key: service-ca.crt
            configMapName: openshift-service-ca.crt
      pipelines:
      - name: default-logstore
        inputRefs:
        - application
        - infrastructure
        outputRefs:
        - default-lokistack
    Note

    The dataModel field is optional and left unset (dataModel: "") by default. This allows the Cluster Logging Operator (CLO) to automatically select a data model. Currently, the CLO defaults to the ViaQ model when the field is unset, but this will change in future releases. Specifying dataModel: ViaQ ensures the configuration remains compatible if the default changes.

Verification

  • Verify that logs are visible in the Log section of the Observe tab in the OpenShift web console.
3.2.6.2. Quick start with OpenTelemetry
Important

The OpenTelemetry Protocol (OTLP) output log forwarder is a Technology Preview feature only. Technology Preview features are not supported with Red Hat production service level agreements (SLAs) and might not be functionally complete. Red Hat does not recommend using them in production. These features provide early access to upcoming product features, enabling customers to test functionality and provide feedback during the development process.

For more information about the support scope of Red Hat Technology Preview features, see Technology Preview Features Support Scope.

To configure OTLP ingestion and enable the OpenTelemetry data model, follow these steps:

Prerequisites

  • Cluster administrator permissions

Procedure

  1. Install the Red Hat OpenShift Logging Operator, Loki Operator, and Cluster Observability Operator (COO) from OperatorHub.
  2. Create a LokiStack custom resource (CR) in the openshift-logging namespace:

    apiVersion: loki.grafana.com/v1
    kind: LokiStack
    metadata:
      name: logging-loki
      namespace: openshift-logging
    spec:
      managementState: Managed
      size: 1x.extra-small
      storage:
        schemas:
        - effectiveDate: '2024-10-01'
          version: v13
        secret:
          name: logging-loki-s3
          type: s3
      storageClassName: gp3-csi
      tenants:
        mode: openshift-logging
    Note

    Ensure that the logging-loki-s3 secret is created beforehand. The contents of this secret vary depending on the object storage in use. For more information, see "Secrets and TLS Configuration".

  3. Create a service account for the collector:

    $ oc create sa collector -n openshift-logging
  4. Allow the collector’s service account to write data to the LokiStack CR:

    $ oc adm policy add-cluster-role-to-user logging-collector-logs-writer -z collector
    Note

    The ClusterRole resource is created automatically during the Cluster Logging Operator installation and does not need to be created manually.

  5. Allow the collector’s service account to collect logs:

    $ oc project openshift-logging
    $ oc adm policy add-cluster-role-to-user collect-application-logs -z collector
    $ oc adm policy add-cluster-role-to-user collect-audit-logs -z collector
    $ oc adm policy add-cluster-role-to-user collect-infrastructure-logs -z collector
    Note

    The example binds the collector to all three roles (application, infrastructure, and audit). By default, only application and infrastructure logs are collected. To collect audit logs, update your ClusterLogForwarder configuration to include them. Assign roles based on the specific log types required for your environment.

  6. Create a UIPlugin CR to enable the Log section in the Observe tab:

    apiVersion: observability.openshift.io/v1alpha1
    kind: UIPlugin
    metadata:
      name: logging
    spec:
      type: Logging
      logging:
        lokiStack:
          name: logging-loki
  7. Create a ClusterLogForwarder CR to configure log forwarding:

    apiVersion: observability.openshift.io/v1
    kind: ClusterLogForwarder
    metadata:
      name: collector
      namespace: openshift-logging
      annotations:
        observability.openshift.io/tech-preview-otlp-output: "enabled" 1
    spec:
      serviceAccount:
        name: collector
      outputs:
      - name: loki-otlp
        type: lokiStack 2
        lokiStack:
          target:
            name: logging-loki
            namespace: openshift-logging
          dataModel: Otel 3
          authentication:
            token:
              from: serviceAccount
        tls:
          ca:
            key: service-ca.crt
            configMapName: openshift-service-ca.crt
      pipelines:
      - name: my-pipeline
        inputRefs:
        - application
        - infrastructure
        outputRefs:
        - loki-otlp
    1
    Use the annotation to enable the Otel data model, which is a Technology Preview feature.
    2
    Define the output type as lokiStack.
    3
    Specifies the OpenTelemetry data model.
    Note

    You cannot use lokiStack.labelKeys when dataModel is Otel. To achieve similar functionality when dataModel is Otel, refer to "Configuring LokiStack for OTLP data ingestion".

Verification

  • Verify that OTLP is functioning correctly by going to ObserveOpenShift LoggingLokiStackWrites in the OpenShift web console, and checking Distributor - Structured Metadata.

3.3. Configuring log forwarding

The ClusterLogForwarder (CLF) allows users to configure forwarding of logs to various destinations. It provides a flexible way to select log messages from different sources, send them through a pipeline that can transform or filter them, and forward them to one or more outputs.

Key Functions of the ClusterLogForwarder

  • Selects log messages using inputs
  • Forwards logs to external destinations using outputs
  • Filters, transforms, and drops log messages using filters
  • Defines log forwarding pipelines connecting inputs, filters and outputs

3.3.1. Setting up log collection

This release of Cluster Logging requires administrators to explicitly grant log collection permissions to the service account associated with ClusterLogForwarder. This was not required in previous releases for the legacy logging scenario consisting of a ClusterLogging and, optionally, a ClusterLogForwarder.logging.openshift.io resource.

The Red Hat OpenShift Logging Operator provides collect-audit-logs, collect-application-logs, and collect-infrastructure-logs cluster roles, which enable the collector to collect audit logs, application logs, and infrastructure logs respectively.

Setup log collection by binding the required cluster roles to your service account.

3.3.1.1. Legacy service accounts

To use the existing legacy service account logcollector, create the following ClusterRoleBinding:

$ oc adm policy add-cluster-role-to-user collect-application-logs system:serviceaccount:openshift-logging:logcollector
$ oc adm policy add-cluster-role-to-user collect-infrastructure-logs system:serviceaccount:openshift-logging:logcollector

Additionally, create the following ClusterRoleBinding if collecting audit logs:

$ oc adm policy add-cluster-role-to-user collect-audit-logs system:serviceaccount:openshift-logging:logcollector
3.3.1.2. Creating service accounts

Prerequisites

  • The Red Hat OpenShift Logging Operator is installed in the openshift-logging namespace.
  • You have administrator permissions.

Procedure

  1. Create a service account for the collector. If you want to write logs to storage that requires a token for authentication, you must include a token in the service account.
  2. Bind the appropriate cluster roles to the service account:

    Example binding command

    $ oc adm policy add-cluster-role-to-user <cluster_role_name> system:serviceaccount:<namespace_name>:<service_account_name>

3.3.1.2.1. Cluster Role Binding for your Service Account

The role_binding.yaml file binds the ClusterLogging operator’s ClusterRole to a specific ServiceAccount, allowing it to manage Kubernetes resources cluster-wide.

apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
  name: manager-rolebinding
roleRef:                                           1
  apiGroup: rbac.authorization.k8s.io              2
  kind: ClusterRole                                3
  name: cluster-logging-operator                   4
subjects:                                          5
  - kind: ServiceAccount                           6
    name: cluster-logging-operator                 7
    namespace: openshift-logging                   8
1
roleRef: References the ClusterRole to which the binding applies.
2
apiGroup: Indicates the RBAC API group, specifying that the ClusterRole is part of Kubernetes' RBAC system.
3
kind: Specifies that the referenced role is a ClusterRole, which applies cluster-wide.
4
name: The name of the ClusterRole being bound to the ServiceAccount, here cluster-logging-operator.
5
subjects: Defines the entities (users or service accounts) that are being granted the permissions from the ClusterRole.
6
kind: Specifies that the subject is a ServiceAccount.
7
Name: The name of the ServiceAccount being granted the permissions.
8
namespace: Indicates the namespace where the ServiceAccount is located.
3.3.1.2.2. Writing application logs

The write-application-logs-clusterrole.yaml file defines a ClusterRole that grants permissions to write application logs to the Loki logging application.

apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
  name: cluster-logging-write-application-logs
rules:                                              1
  - apiGroups:                                      2
      - loki.grafana.com                            3
    resources:                                      4
      - application                                 5
    resourceNames:                                  6
      - logs                                        7
    verbs:                                          8
      - create                                      9
Annotations
<1> rules: Specifies the permissions granted by this ClusterRole.
<2> apiGroups: Refers to the API group loki.grafana.com, which relates to the Loki logging system.
<3> loki.grafana.com: The API group for managing Loki-related resources.
<4> resources: The resource type that the ClusterRole grants permission to interact with.
<5> application: Refers to the application resources within the Loki logging system.
<6> resourceNames: Specifies the names of resources that this role can manage.
<7> logs: Refers to the log resources that can be created.
<8> verbs: The actions allowed on the resources.
<9> create: Grants permission to create new logs in the Loki system.
3.3.1.2.3. Writing audit logs

The write-audit-logs-clusterrole.yaml file defines a ClusterRole that grants permissions to create audit logs in the Loki logging system.

apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
  name: cluster-logging-write-audit-logs
rules:                                              1
  - apiGroups:                                      2
      - loki.grafana.com                            3
    resources:                                      4
      - audit                                       5
    resourceNames:                                  6
      - logs                                        7
    verbs:                                          8
      - create                                      9
1 1
rules: Defines the permissions granted by this ClusterRole.
2 2
apiGroups: Specifies the API group loki.grafana.com.
3 3
loki.grafana.com: The API group responsible for Loki logging resources.
4 4
resources: Refers to the resource type this role manages, in this case, audit.
5 5
audit: Specifies that the role manages audit logs within Loki.
6 6
resourceNames: Defines the specific resources that the role can access.
7 7
logs: Refers to the logs that can be managed under this role.
8 8
verbs: The actions allowed on the resources.
9 9
create: Grants permission to create new audit logs.
3.3.1.2.4. Writing infrastructure logs

The write-infrastructure-logs-clusterrole.yaml file defines a ClusterRole that grants permission to create infrastructure logs in the Loki logging system.

Sample YAML

apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
  name: cluster-logging-write-infrastructure-logs
rules:                                              1
  - apiGroups:                                      2
      - loki.grafana.com                            3
    resources:                                      4
      - infrastructure                              5
    resourceNames:                                  6
      - logs                                        7
    verbs:                                          8
      - create                                      9

1
rules: Specifies the permissions this ClusterRole grants.
2
apiGroups: Specifies the API group for Loki-related resources.
3
loki.grafana.com: The API group managing the Loki logging system.
4
resources: Defines the resource type that this role can interact with.
5
infrastructure: Refers to infrastructure-related resources that this role manages.
6
resourceNames: Specifies the names of resources this role can manage.
7
logs: Refers to the log resources related to infrastructure.
8
verbs: The actions permitted by this role.
9
create: Grants permission to create infrastructure logs in the Loki system.
3.3.1.2.5. ClusterLogForwarder editor role

The clusterlogforwarder-editor-role.yaml file defines a ClusterRole that allows users to manage ClusterLogForwarders in OpenShift.

apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
  name: clusterlogforwarder-editor-role
rules:                                              1
  - apiGroups:                                      2
      - observability.openshift.io                  3
    resources:                                      4
      - clusterlogforwarders                        5
    verbs:                                          6
      - create                                      7
      - delete                                      8
      - get                                         9
      - list                                        10
      - patch                                       11
      - update                                      12
      - watch                                       13
1
rules: Specifies the permissions this ClusterRole grants.
2
apiGroups: Refers to the OpenShift-specific API group
3
obervability.openshift.io: The API group for managing observability resources, like logging.
4
resources: Specifies the resources this role can manage.
5
clusterlogforwarders: Refers to the log forwarding resources in OpenShift.
6
verbs: Specifies the actions allowed on the ClusterLogForwarders.
7
create: Grants permission to create new ClusterLogForwarders.
8
delete: Grants permission to delete existing ClusterLogForwarders.
9
get: Grants permission to retrieve information about specific ClusterLogForwarders.
10
list: Allows listing all ClusterLogForwarders.
11
patch: Grants permission to partially modify ClusterLogForwarders.
12
update: Grants permission to update existing ClusterLogForwarders.
13
watch: Grants permission to monitor changes to ClusterLogForwarders.

3.3.2. Modifying log level in collector

To modify the log level in the collector, you can set the observability.openshift.io/log-level annotation to trace, debug, info, warn, error, and off.

Example log level annotation

apiVersion: observability.openshift.io/v1
kind: ClusterLogForwarder
metadata:
  name: collector
  annotations:
    observability.openshift.io/log-level: debug
# ...

3.3.3. Managing the Operator

The ClusterLogForwarder resource has a managementState field that controls whether the operator actively manages its resources or leaves them Unmanaged:

Managed
(default) The operator will drive the logging resources to match the desired state in the CLF spec.
Unmanaged
The operator will not take any action related to the logging components.

This allows administrators to temporarily pause log forwarding by setting managementState to Unmanaged.

3.3.4. Structure of the ClusterLogForwarder

The CLF has a spec section that contains the following key components:

Inputs
Select log messages to be forwarded. Built-in input types application, infrastructure and audit forward logs from different parts of the cluster. You can also define custom inputs.
Outputs
Define destinations to forward logs to. Each output has a unique name and type-specific configuration.
Pipelines
Define the path logs take from inputs, through filters, to outputs. Pipelines have a unique name and consist of a list of input, output and filter names.
Filters
Transform or drop log messages in the pipeline. Users can define filters that match certain log fields and drop or modify the messages. Filters are applied in the order specified in the pipeline.
3.3.4.1. Inputs

Inputs are configured in an array under spec.inputs. There are three built-in input types:

application
Selects logs from all application containers, excluding those in infrastructure namespaces such as default, openshift, or any namespace with the kube- or openshift- prefix.
infrastructure
Selects logs from infrastructure components running in default and openshift namespaces and node logs.
audit
Selects logs from the OpenShift API server audit logs, Kubernetes API server audit logs, ovn audit logs, and node audit logs from auditd.

Users can define custom inputs of type application that select logs from specific namespaces or using pod labels.

3.3.4.2. Outputs

Outputs are configured in an array under spec.outputs. Each output must have a unique name and a type. Supported types are:

azureMonitor
Forwards logs to Azure Monitor.
cloudwatch
Forwards logs to AWS CloudWatch.
elasticsearch
Forwards logs to an external Elasticsearch instance.
googleCloudLogging
Forwards logs to Google Cloud Logging.
http
Forwards logs to a generic HTTP endpoint.
kafka
Forwards logs to a Kafka broker.
loki
Forwards logs to a Loki logging backend.
lokistack
Forwards logs to the logging supported combination of Loki and web proxy with OpenShift Container Platform authentication integration. LokiStack’s proxy uses OpenShift Container Platform authentication to enforce multi-tenancy
otlp
Forwards logs using the OpenTelemetry Protocol.
splunk
Forwards logs to Splunk.
syslog
Forwards logs to an external syslog server.

Each output type has its own configuration fields.

3.3.5. Configuring OTLP output

Cluster administrators can use the OpenTelemetry Protocol (OTLP) output to collect and forward logs to OTLP receivers. The OTLP output uses the specification defined by the OpenTelemetry Observability framework to send data over HTTP with JSON encoding.

Important

The OpenTelemetry Protocol (OTLP) output log forwarder is a Technology Preview feature only. Technology Preview features are not supported with Red Hat production service level agreements (SLAs) and might not be functionally complete. Red Hat does not recommend using them in production. These features provide early access to upcoming product features, enabling customers to test functionality and provide feedback during the development process.

For more information about the support scope of Red Hat Technology Preview features, see Technology Preview Features Support Scope.

Procedure

  • Create or edit a ClusterLogForwarder custom resource (CR) to enable forwarding using OTLP by adding the following annotation:

    Example ClusterLogForwarder CR

    apiVersion: observability.openshift.io/v1
    kind: ClusterLogForwarder
    metadata:
      annotations:
        observability.openshift.io/tech-preview-otlp-output: "enabled" 1
      name: clf-otlp
    spec:
      serviceAccount:
        name: <service_account_name>
      outputs:
      - name: otlp
        type: otlp
        otlp:
          tuning:
            compression: gzip
            deliveryMode: AtLeastOnce
            maxRetryDuration: 20
            maxWrite: 10M
            minRetryDuration: 5
          url: <otlp_url> 2
      pipelines:
      - inputRefs:
        - application
        - infrastructure
        - audit
        name: otlp-logs
        outputRefs:
        - otlp

    1
    Use this annotation to enable the OpenTelemetry Protocol (OTLP) output, which is a Technology Preview feature.
    2
    This URL must be absolute and is a placeholder for the OTLP endpoint where logs are sent.
Note

The OTLP output uses the OpenTelemetry data model, which is different from the ViaQ data model that is used by other output types. It adheres to the OTLP using OpenTelemetry Semantic Conventions defined by the OpenTelemetry Observability framework.

3.3.5.1. Pipelines

Pipelines are configured in an array under spec.pipelines. Each pipeline must have a unique name and consists of:

inputRefs
Names of inputs whose logs should be forwarded to this pipeline.
outputRefs
Names of outputs to send logs to.
filterRefs
(optional) Names of filters to apply.

The order of filterRefs matters, as they are applied sequentially. Earlier filters can drop messages that will not be processed by later filters.

3.3.5.2. Filters

Filters are configured in an array under spec.filters. They can match incoming log messages based on the value of structured fields and modify or drop them.

Administrators can configure the following types of filters:

3.3.5.3. Enabling multi-line exception detection

Enables multi-line error detection of container logs.

Warning

Enabling this feature could have performance implications and may require additional computing resources or alternate logging solutions.

Log parsers often incorrectly identify separate lines of the same exception as separate exceptions. This leads to extra log entries and an incomplete or inaccurate view of the traced information.

Example java exception

java.lang.NullPointerException: Cannot invoke "String.toString()" because "<param1>" is null
    at testjava.Main.handle(Main.java:47)
    at testjava.Main.printMe(Main.java:19)
    at testjava.Main.main(Main.java:10)

  • To enable logging to detect multi-line exceptions and reassemble them into a single log entry, ensure that the ClusterLogForwarder Custom Resource (CR) contains a detectMultilineErrors field under the .spec.filters.

Example ClusterLogForwarder CR

apiVersion: "observability.openshift.io/v1"
kind: ClusterLogForwarder
metadata:
  name: <log_forwarder_name>
  namespace: <log_forwarder_namespace>
spec:
  serviceAccount:
    name: <service_account_name>
  filters:
  - name: <name>
    type: detectMultilineException
  pipelines:
    - inputRefs:
        - <input-name>
      name: <pipeline-name>
      filterRefs:
        - <filter-name>
      outputRefs:
        - <output-name>

3.3.5.3.1. Details

When log messages appear as a consecutive sequence forming an exception stack trace, they are combined into a single, unified log record. The first log message’s content is replaced with the concatenated content of all the message fields in the sequence.

The collector supports the following languages:

  • Java
  • JS
  • Ruby
  • Python
  • Golang
  • PHP
  • Dart
3.3.5.4. Configuring content filters to drop unwanted log records

When the drop filter is configured, the log collector evaluates log streams according to the filters before forwarding. The collector drops unwanted log records that match the specified configuration.

Procedure

  1. Add a configuration for a filter to the filters spec in the ClusterLogForwarder CR.

    The following example shows how to configure the ClusterLogForwarder CR to drop log records based on regular expressions:

    Example ClusterLogForwarder CR

    apiVersion: observability.openshift.io/v1
    kind: ClusterLogForwarder
    metadata:
    # ...
    spec:
      serviceAccount:
        name: <service_account_name>
      filters:
      - name: <filter_name>
        type: drop 1
        drop: 2
        - test: 3
          - field: .kubernetes.labels."foo-bar/baz" 4
            matches: .+ 5
          - field: .kubernetes.pod_name
            notMatches: "my-pod" 6
      pipelines:
      - name: <pipeline_name> 7
        filterRefs: ["<filter_name>"]
    # ...

    1
    Specifies the type of filter. The drop filter drops log records that match the filter configuration.
    2
    Specifies configuration options for applying the drop filter.
    3
    Specifies the configuration for tests that are used to evaluate whether a log record is dropped.
    • If all the conditions specified for a test are true, the test passes and the log record is dropped.
    • When multiple tests are specified for the drop filter configuration, if any of the tests pass, the record is dropped.
    • If there is an error evaluating a condition, for example, the field is missing from the log record being evaluated, that condition evaluates to false.
    4
    Specifies a dot-delimited field path, which is a path to a field in the log record. The path can contain alpha-numeric characters and underscores (a-zA-Z0-9_), for example, .kubernetes.namespace_name. If segments contain characters outside of this range, the segment must be in quotes, for example, .kubernetes.labels."foo.bar-bar/baz". You can include multiple field paths in a single test configuration, but they must all evaluate to true for the test to pass and the drop filter to be applied.
    5
    Specifies a regular expression. If log records match this regular expression, they are dropped. You can set either the matches or notMatches condition for a single field path, but not both.
    6
    Specifies a regular expression. If log records do not match this regular expression, they are dropped. You can set either the matches or notMatches condition for a single field path, but not both.
    7
    Specifies the pipeline that the drop filter is applied to.
  2. Apply the ClusterLogForwarder CR by running the following command:

    $ oc apply -f <filename>.yaml

Additional examples

The following additional example shows how you can configure the drop filter to only keep higher priority log records:

apiVersion: observability.openshift.io/v1
kind: ClusterLogForwarder
metadata:
# ...
spec:
  serviceAccount:
    name: <service_account_name>
  filters:
  - name: important
    type: drop
    drop:
    - test:
      - field: .message
        notMatches: "(?i)critical|error"
      - field: .level
        matches: "info|warning"
# ...

In addition to including multiple field paths in a single test configuration, you can also include additional tests that are treated as OR checks. In the following example, records are dropped if either test configuration evaluates to true. However, for the second test configuration, both field specs must be true for it to be evaluated to true:

apiVersion: observability.openshift.io/v1
kind: ClusterLogForwarder
metadata:
# ...
spec:
  serviceAccount:
    name: <service_account_name>
  filters:
  - name: important
    type: drop
    drop:
    - test:
      - field: .kubernetes.namespace_name
        matches: "^open"
    - test:
      - field: .log_type
        matches: "application"
      - field: .kubernetes.pod_name
        notMatches: "my-pod"
# ...
3.3.5.5. Overview of API audit filter

OpenShift API servers generate audit events for each API call, detailing the request, response, and the identity of the requester, leading to large volumes of data. The API Audit filter uses rules to enable the exclusion of non-essential events and the reduction of event size, facilitating a more manageable audit trail. Rules are checked in order, and checking stops at the first match. The amount of data that is included in an event is determined by the value of the level field:

  • None: The event is dropped.
  • Metadata: Audit metadata is included, request and response bodies are removed.
  • Request: Audit metadata and the request body are included, the response body is removed.
  • RequestResponse: All data is included: metadata, request body and response body. The response body can be very large. For example, oc get pods -A generates a response body containing the YAML description of every pod in the cluster.

The ClusterLogForwarder custom resource (CR) uses the same format as the standard Kubernetes audit policy, while providing the following additional functions:

Wildcards
Names of users, groups, namespaces, and resources can have a leading or trailing * asterisk character. For example, the namespace openshift-\* matches openshift-apiserver or openshift-authentication. Resource \*/status matches Pod/status or Deployment/status.
Default Rules

Events that do not match any rule in the policy are filtered as follows:

  • Read-only system events such as get, list, and watch are dropped.
  • Service account write events that occur within the same namespace as the service account are dropped.
  • All other events are forwarded, subject to any configured rate limits.

To disable these defaults, either end your rules list with a rule that has only a level field or add an empty rule.

Omit Response Codes
A list of integer status codes to omit. You can drop events based on the HTTP status code in the response by using the OmitResponseCodes field, which lists HTTP status codes for which no events are created. The default value is [404, 409, 422, 429]. If the value is an empty list, [], then no status codes are omitted.

The ClusterLogForwarder CR audit policy acts in addition to the OpenShift Container Platform audit policy. The ClusterLogForwarder CR audit filter changes what the log collector forwards and provides the ability to filter by verb, user, group, namespace, or resource. You can create multiple filters to send different summaries of the same audit stream to different places. For example, you can send a detailed stream to the local cluster log store and a less detailed stream to a remote site.

Note

You must have a cluster role collect-audit-logs to collect the audit logs. The following example provided is intended to illustrate the range of rules possible in an audit policy and is not a recommended configuration.

Example audit policy

apiVersion: observability.openshift.io/v1
kind: ClusterLogForwarder
metadata:
  name: <log_forwarder_name>
  namespace: <log_forwarder_namespace>
spec:
  serviceAccount:
    name: <service_account_name>
  pipelines:
    - name: my-pipeline
      inputRefs: audit 1
      filterRefs: my-policy 2
  filters:
    - name: my-policy
      type: kubeAPIAudit
      kubeAPIAudit:
        # Don't generate audit events for all requests in RequestReceived stage.
        omitStages:
          - "RequestReceived"

        rules:
          # Log pod changes at RequestResponse level
          - level: RequestResponse
            resources:
            - group: ""
              resources: ["pods"]

          # Log "pods/log", "pods/status" at Metadata level
          - level: Metadata
            resources:
            - group: ""
              resources: ["pods/log", "pods/status"]

          # Don't log requests to a configmap called "controller-leader"
          - level: None
            resources:
            - group: ""
              resources: ["configmaps"]
              resourceNames: ["controller-leader"]

          # Don't log watch requests by the "system:kube-proxy" on endpoints or services
          - level: None
            users: ["system:kube-proxy"]
            verbs: ["watch"]
            resources:
            - group: "" # core API group
              resources: ["endpoints", "services"]

          # Don't log authenticated requests to certain non-resource URL paths.
          - level: None
            userGroups: ["system:authenticated"]
            nonResourceURLs:
            - "/api*" # Wildcard matching.
            - "/version"

          # Log the request body of configmap changes in kube-system.
          - level: Request
            resources:
            - group: "" # core API group
              resources: ["configmaps"]
            # This rule only applies to resources in the "kube-system" namespace.
            # The empty string "" can be used to select non-namespaced resources.
            namespaces: ["kube-system"]

          # Log configmap and secret changes in all other namespaces at the Metadata level.
          - level: Metadata
            resources:
            - group: "" # core API group
              resources: ["secrets", "configmaps"]

          # Log all other resources in core and extensions at the Request level.
          - level: Request
            resources:
            - group: "" # core API group
            - group: "extensions" # Version of group should NOT be included.

          # A catch-all rule to log all other requests at the Metadata level.
          - level: Metadata

1
The log types that are collected. The value for this field can be audit for audit logs, application for application logs, infrastructure for infrastructure logs, or a named input that has been defined for your application.
2
The name of your audit policy.
3.3.5.6. Filtering application logs at input by including the label expressions or a matching label key and values

You can include the application logs based on the label expressions or a matching label key and its values by using the input selector.

Procedure

  1. Add a configuration for a filter to the input spec in the ClusterLogForwarder CR.

    The following example shows how to configure the ClusterLogForwarder CR to include logs based on label expressions or matched label key/values:

    Example ClusterLogForwarder CR

    apiVersion: observability.openshift.io/v1
    kind: ClusterLogForwarder
    # ...
    spec:
      serviceAccount:
        name: <service_account_name>
      inputs:
        - name: mylogs
          application:
            selector:
              matchExpressions:
              - key: env 1
                operator: In 2
                values: ["prod", "qa"] 3
              - key: zone
                operator: NotIn
                values: ["east", "west"]
              matchLabels: 4
                app: one
                name: app1
          type: application
    # ...

    1
    Specifies the label key to match.
    2
    Specifies the operator. Valid values include: In, NotIn, Exists, and DoesNotExist.
    3
    Specifies an array of string values. If the operator value is either Exists or DoesNotExist, the value array must be empty.
    4
    Specifies an exact key or value mapping.
  2. Apply the ClusterLogForwarder CR by running the following command:

    $ oc apply -f <filename>.yaml
3.3.5.7. Configuring content filters to prune log records

When the prune filter is configured, the log collector evaluates log streams according to the filters before forwarding. The collector prunes log records by removing low value fields such as pod annotations.

Procedure

  1. Add a configuration for a filter to the prune spec in the ClusterLogForwarder CR.

    The following example shows how to configure the ClusterLogForwarder CR to prune log records based on field paths:

    Important

    If both are specified, records are pruned based on the notIn array first, which takes precedence over the in array. After records have been pruned by using the notIn array, they are then pruned by using the in array.

    Example ClusterLogForwarder CR

    apiVersion: observability.openshift.io/v1
    kind: ClusterLogForwarder
    metadata:
    # ...
    spec:
      serviceAccount:
        name: <service_account_name>
      filters:
      - name: <filter_name>
        type: prune 1
        prune: 2
          in: [.kubernetes.annotations, .kubernetes.namespace_id] 3
          notIn: [.kubernetes,.log_type,.message,."@timestamp"] 4
      pipelines:
      - name: <pipeline_name> 5
        filterRefs: ["<filter_name>"]
    # ...

    1
    Specify the type of filter. The prune filter prunes log records by configured fields.
    2
    Specify configuration options for applying the prune filter. The in and notIn fields are specified as arrays of dot-delimited field paths, which are paths to fields in log records. These paths can contain alpha-numeric characters and underscores (a-zA-Z0-9_), for example, .kubernetes.namespace_name. If segments contain characters outside of this range, the segment must be in quotes, for example, .kubernetes.labels."foo.bar-bar/baz".
    3
    Optional: Any fields that are specified in this array are removed from the log record.
    4
    Optional: Any fields that are not specified in this array are removed from the log record.
    5
    Specify the pipeline that the prune filter is applied to.
    Note

    The filters exempts the log_type, .log_source, and .message fields.

  2. Apply the ClusterLogForwarder CR by running the following command:

    $ oc apply -f <filename>.yaml

3.3.6. Filtering the audit and infrastructure log inputs by source

You can define the list of audit and infrastructure sources to collect the logs by using the input selector.

Procedure

  1. Add a configuration to define the audit and infrastructure sources in the ClusterLogForwarder CR.

    The following example shows how to configure the ClusterLogForwarder CR to define audit and infrastructure sources:

    Example ClusterLogForwarder CR

    apiVersion: observability.openshift.io/v1
    kind: ClusterLogForwarder
    # ...
    spec:
      serviceAccount:
        name: <service_account_name>
      inputs:
        - name: mylogs1
          type: infrastructure
          infrastructure:
            sources: 1
              - node
        - name: mylogs2
          type: audit
          audit:
            sources: 2
              - kubeAPI
              - openshiftAPI
              - ovn
    # ...

    1
    Specifies the list of infrastructure sources to collect. The valid sources include:
    • node: Journal log from the node
    • container: Logs from the workloads deployed in the namespaces
    2
    Specifies the list of audit sources to collect. The valid sources include:
    • kubeAPI: Logs from the Kubernetes API servers
    • openshiftAPI: Logs from the OpenShift API servers
    • auditd: Logs from a node auditd service
    • ovn: Logs from an open virtual network service
  2. Apply the ClusterLogForwarder CR by running the following command:

    $ oc apply -f <filename>.yaml

3.3.7. Filtering application logs at input by including or excluding the namespace or container name

You can include or exclude the application logs based on the namespace and container name by using the input selector.

Procedure

  1. Add a configuration to include or exclude the namespace and container names in the ClusterLogForwarder CR.

    The following example shows how to configure the ClusterLogForwarder CR to include or exclude namespaces and container names:

    Example ClusterLogForwarder CR

    apiVersion: observability.openshift.io/v1
    kind: ClusterLogForwarder
    # ...
    spec:
      serviceAccount:
        name: <service_account_name>
      inputs:
        - name: mylogs
          application:
            includes:
              - namespace: "my-project" 1
                container: "my-container" 2
            excludes:
              - container: "other-container*" 3
                namespace: "other-namespace" 4
          type: application
    # ...

    1
    Specifies that the logs are only collected from these namespaces.
    2
    Specifies that the logs are only collected from these containers.
    3
    Specifies the pattern of namespaces to ignore when collecting the logs.
    4
    Specifies the set of containers to ignore when collecting the logs.
    Note

    The excludes field takes precedence over the includes field.

  2. Apply the ClusterLogForwarder CR by running the following command:

    $ oc apply -f <filename>.yaml

3.4. Storing logs with LokiStack

You can configure a LokiStack CR to store application, audit, and infrastructure-related logs.

Loki is a horizontally scalable, highly available, multi-tenant log aggregation system offered as a GA log store for logging for Red Hat OpenShift that can be visualized with the OpenShift Observability UI. The Loki configuration provided by OpenShift Logging is a short-term log store designed to enable users to perform fast troubleshooting with the collected logs. For that purpose, the logging for Red Hat OpenShift configuration of Loki has short-term storage, and is optimized for very recent queries.

Important

For long-term storage or queries over a long time period, users should look to log stores external to their cluster. Loki sizing is only tested and supported for short term storage, for a maximum of 30 days.

3.4.1. Loki deployment sizing

Sizing for Loki follows the format of 1x.<size> where the value 1x is number of instances and <size> specifies performance capabilities.

The 1x.pico configuration defines a single Loki deployment with minimal resource and limit requirements, offering high availability (HA) support for all Loki components. This configuration is suited for deployments that do not require a single replication factor or auto-compaction.

Disk requests are similar across size configurations, allowing customers to test different sizes to determine the best fit for their deployment needs.

Important

It is not possible to change the number 1x for the deployment size.

Table 3.1. Loki sizing
 1x.demo1x.pico [6.1+ only]1x.extra-small1x.small1x.medium

Data transfer

Demo use only

50GB/day

100GB/day

500GB/day

2TB/day

Queries per second (QPS)

Demo use only

1-25 QPS at 200ms

1-25 QPS at 200ms

25-50 QPS at 200ms

25-75 QPS at 200ms

Replication factor

None

2

2

2

2

Total CPU requests

None

7 vCPUs

14 vCPUs

34 vCPUs

54 vCPUs

Total CPU requests if using the ruler

None

8 vCPUs

16 vCPUs

42 vCPUs

70 vCPUs

Total memory requests

None

17Gi

31Gi

67Gi

139Gi

Total memory requests if using the ruler

None

18Gi

35Gi

83Gi

171Gi

Total disk requests

40Gi

590Gi

430Gi

430Gi

590Gi

Total disk requests if using the ruler

80Gi

910Gi

750Gi

750Gi

910Gi

3.4.2. Prerequisites

  • You have installed the Loki Operator by using the CLI or web console.
  • You have a serviceAccount in the same namespace in which you create the ClusterLogForwarder.
  • The serviceAccount is assigned collect-audit-logs, collect-application-logs, and collect-infrastructure-logs cluster roles.

3.4.3. Core Setup and Configuration

Role-based access controls, basic monitoring, and pod placement to deploy Loki.

3.4.4. Authorizing LokiStack rules RBAC permissions

Administrators can allow users to create and manage their own alerting and recording rules by binding cluster roles to usernames. Cluster roles are defined as ClusterRole objects that contain necessary role-based access control (RBAC) permissions for users.

The following cluster roles for alerting and recording rules are available for LokiStack:

Rule nameDescription

alertingrules.loki.grafana.com-v1-admin

Users with this role have administrative-level access to manage alerting rules. This cluster role grants permissions to create, read, update, delete, list, and watch AlertingRule resources within the loki.grafana.com/v1 API group.

alertingrules.loki.grafana.com-v1-crdview

Users with this role can view the definitions of Custom Resource Definitions (CRDs) related to AlertingRule resources within the loki.grafana.com/v1 API group, but do not have permissions for modifying or managing these resources.

alertingrules.loki.grafana.com-v1-edit

Users with this role have permission to create, update, and delete AlertingRule resources.

alertingrules.loki.grafana.com-v1-view

Users with this role can read AlertingRule resources within the loki.grafana.com/v1 API group. They can inspect configurations, labels, and annotations for existing alerting rules but cannot make any modifications to them.

recordingrules.loki.grafana.com-v1-admin

Users with this role have administrative-level access to manage recording rules. This cluster role grants permissions to create, read, update, delete, list, and watch RecordingRule resources within the loki.grafana.com/v1 API group.

recordingrules.loki.grafana.com-v1-crdview

Users with this role can view the definitions of Custom Resource Definitions (CRDs) related to RecordingRule resources within the loki.grafana.com/v1 API group, but do not have permissions for modifying or managing these resources.

recordingrules.loki.grafana.com-v1-edit

Users with this role have permission to create, update, and delete RecordingRule resources.

recordingrules.loki.grafana.com-v1-view

Users with this role can read RecordingRule resources within the loki.grafana.com/v1 API group. They can inspect configurations, labels, and annotations for existing alerting rules but cannot make any modifications to them.

3.4.4.1. Examples

To apply cluster roles for a user, you must bind an existing cluster role to a specific username.

Cluster roles can be cluster or namespace scoped, depending on which type of role binding you use. When a RoleBinding object is used, as when using the oc adm policy add-role-to-user command, the cluster role only applies to the specified namespace. When a ClusterRoleBinding object is used, as when using the oc adm policy add-cluster-role-to-user command, the cluster role applies to all namespaces in the cluster.

The following example command gives the specified user create, read, update and delete (CRUD) permissions for alerting rules in a specific namespace in the cluster:

Example cluster role binding command for alerting rule CRUD permissions in a specific namespace

$ oc adm policy add-role-to-user alertingrules.loki.grafana.com-v1-admin -n <namespace> <username>

The following command gives the specified user administrator permissions for alerting rules in all namespaces:

Example cluster role binding command for administrator permissions

$ oc adm policy add-cluster-role-to-user alertingrules.loki.grafana.com-v1-admin <username>

3.4.5. Creating a log-based alerting rule with Loki

The AlertingRule CR contains a set of specifications and webhook validation definitions to declare groups of alerting rules for a single LokiStack instance. In addition, the webhook validation definition provides support for rule validation conditions:

  • If an AlertingRule CR includes an invalid interval period, it is an invalid alerting rule
  • If an AlertingRule CR includes an invalid for period, it is an invalid alerting rule.
  • If an AlertingRule CR includes an invalid LogQL expr, it is an invalid alerting rule.
  • If an AlertingRule CR includes two groups with the same name, it is an invalid alerting rule.
  • If none of the above applies, an alerting rule is considered valid.
Table 3.2. AlertingRule definitions
Tenant typeValid namespaces for AlertingRule CRs

application

<your_application_namespace>

audit

openshift-logging

infrastructure

openshift-/*, kube-/\*, default

Procedure

  1. Create an AlertingRule custom resource (CR):

    Example infrastructure AlertingRule CR

      apiVersion: loki.grafana.com/v1
      kind: AlertingRule
      metadata:
        name: loki-operator-alerts
        namespace: openshift-operators-redhat