Chapter 4. OTLP data ingestion in Loki
You can use an API endpoint by using the OpenTelemetry Protocol (OTLP) with Logging. As OTLP is a standardized format not specifically designed for Loki, OTLP requires an additional Loki configuration to map data format of OpenTelemetry to data model of Loki. OTLP lacks concepts such as stream labels or structured metadata. Instead, OTLP provides metadata about log entries as attributes, grouped into the following three categories:
- Resource
- Scope
- Log
You can set metadata for multiple entries simultaneously or individually as needed.
4.1. Configuring LokiStack for OTLP data ingestion Copy linkLink copied to clipboard!
To configure a LokiStack custom resource (CR) for OTLP ingestion, follow these steps:
Prerequisites
- Ensure that your Loki setup supports structured metadata, introduced in schema version 13 to enable OTLP log ingestion.
Procedure
Set the schema version:
When creating a new
LokiStackCR, setversion: v13in the storage schema configuration.NoteFor existing configurations, add a new schema entry with
version: v13and aneffectiveDatein the future. For more information on updating schema versions, see Upgrading Schemas (Grafana documentation).
Configure the storage schema as follows:
Example configure storage schema
# ... spec: storage: schemas: - version: v13 effectiveDate: 2024-10-25Once the
effectiveDatehas passed, the v13 schema takes effect, enabling yourLokiStackto store structured metadata.If you use the OpenShift Container Platform web console to query logs, enable console labels:
# ... spec: tenants: mode: openshift-logging openshift: otlp: enableConsoleLabels: true
4.2. Attribute mapping Copy linkLink copied to clipboard!
When you set the Loki Operator to the openshift-logging mode, Loki Operator automatically applies a default set of attribute mappings. These mappings align specific OTLP attributes with stream labels and structured metadata of Loki.
For typical setups, these default mappings are sufficient. However, you might need to customize attribute mapping in the following cases:
- Using a custom collector: If your setup includes a custom collector that generates additional attributes that you do not want to store, consider customizing the mapping to ensure these attributes are dropped by Loki.
- Adjusting attribute detail levels: If the default attribute set is more detailed than necessary, you can reduce it to essential attributes only. This can avoid excessive data storage and streamline the logging process.
4.2.1. Custom attribute mapping for OpenShift Copy linkLink copied to clipboard!
When using the Loki Operator in openshift-logging mode, attribute mapping follow OpenShift default values, but you can configure custom mappings to adjust default values. In the openshift-logging mode, you can configure custom attribute mappings globally for all tenants or for individual tenants as needed. When you define custom mappings, they are appended to the OpenShift default values. If you do not need default labels, you can disable them in the tenant configuration.
A major difference between the Loki Operator and Loki lies in inheritance handling. Loki copies only default_resource_attributes_as_index_labels to tenants by default, while the Loki Operator applies the entire global configuration to each tenant in the openshift-logging mode.
Within LokiStack, attribute mapping configuration is managed through the limits setting. See the following example LokiStack configuration:
# ...
spec:
limits:
global:
otlp: {}
tenants:
application:
otlp: {}
You can use both global and per-tenant OTLP configurations for mapping attributes to stream labels.
Stream labels derive only from resource-level attributes, which the LokiStack resource structure reflects. See the following LokiStack example configuration:
spec:
limits:
global:
otlp:
streamLabels:
resourceAttributes:
- name: "k8s.namespace.name"
- name: "k8s.pod.name"
- name: "k8s.container.name"
You can drop attributes of type resource, scope, or log from the log entry.
# ...
spec:
limits:
global:
otlp:
streamLabels:
# ...
drop:
resourceAttributes:
- name: "process.command_line"
- name: "k8s\\.pod\\.labels\\..+"
regex: true
scopeAttributes:
- name: "service.name"
logAttributes:
- name: "http.route"
You can use regular expressions by setting regex: true to apply a configuration for attributes with similar names.
Avoid using regular expressions for stream labels, as this can increase data volume.
Attributes that are not explicitly set as stream labels or dropped from the entry are saved as structured metadata by default.
4.2.2. Required and optional default labels for OpenShift Copy linkLink copied to clipboard!
In the openshift-logging mode, you must not remove certain required attributes because they are essential to OpenShift functions.
If you use the OpenShift Container Platform web console to query logs, enable the attributes called console labels. When you enable console labels, the following attributes are added to stream labels:
-
k8s.container.name -
k8s.cronjob.name -
k8s.daemonset.name -
k8s.deployment.name -
k8s.job.name -
k8s.node.name -
k8s.pod.name -
k8s.statefulset.name -
kubernetes.container_name -
kubernetes.host -
kubernetes.pod_name -
service.name
For information about enabling console attributes, see Configuring LokiStack for OTLP data ingestion.
You can add custom stream labels or drop information from the structured metadata when you use the openshift-logging mode. The custom configuration is merged with the OpenShift defaults.
For information about the attributes, see OpenTelemetry data model attributes.