Chapter 1. Red Hat OpenShift Container Platform

Procedure

1. Open a terminal on your workstation, and log in to OpenShift:

   oc login --token=TOKEN --server=SERVER_URL_AND_PORT

   $ oc login --token=sha256~ZvFDBvoIYAbVECixS4-WmkN4RfnNd8Neh3y1WuiFPXC --server=https://example.com:6443

   Note

   You can find your login token and URL for use on the command line from the OpenShift web console. Log in to the OpenShift web console. Click your user name, and click Copy login command. Offer your user name and password again, if asked, and click Display Token to view the command.

2. Create a new bucket:

   export BUCKET=NEW_BUCKET_NAME
   export REGION=AWS_REGION_ID
   export USER=OADP_USER_NAME
   
   aws s3api create-bucket \
   --bucket $BUCKET \
   --region $REGION \
   --create-bucket-configuration LocationConstraint=$REGION

   $ export BUCKET=example-bucket-name
   $ export REGION=us-east-1
   $ export USER=velero
   $
   $ aws s3api create-bucket \
   --bucket $BUCKET \
   --region $REGION \
   --create-bucket-configuration LocationConstraint=$REGION

3. Create a new user:

   $ aws iam create-user --user-name $USER

4. Create a new policy:

   $ cat > velero-policy.json <<EOF
   {
       "Version": "2012-10-17",
       "Statement": [
           {
               "Effect": "Allow",
               "Action": [
                   "ec2:DescribeVolumes",
                   "ec2:DescribeSnapshots",
                   "ec2:CreateTags",
                   "ec2:CreateVolume",
                   "ec2:CreateSnapshot",
                   "ec2:DeleteSnapshot"
               ],
               "Resource": "*"
           },
           {
               "Effect": "Allow",
               "Action": [
                   "s3:GetObject",
                   "s3:DeleteObject",
                   "s3:PutObject",
                   "s3:AbortMultipartUpload",
                   "s3:ListMultipartUploadParts"
               ],
               "Resource": [
                   "arn:aws:s3:::${BUCKET}/*"
               ]
           },
           {
               "Effect": "Allow",
               "Action": [
                   "s3:ListBucket",
                   "s3:GetBucketLocation",
                   "s3:ListBucketMultipartUploads"
               ],
               "Resource": [
                   "arn:aws:s3:::${BUCKET}"
               ]
           }
       ]
   }
   EOF

5. Associate this policy to the new user:

   $ aws iam put-user-policy \
   --user-name $USER \
   --policy-name velero \
   --policy-document file://velero-policy.json

6. Create an access key:

   $ aws iam create-access-key --user-name $USER --output=json | jq -r '.AccessKey | [ "export AWS_ACCESS_KEY_ID=" + .AccessKeyId, "export AWS_SECRET_ACCESS_KEY=" + .SecretAccessKey ] | join("\n")'

7. Create a credentials file with your AWS secret key information:

   $ cat << EOF > ./credentials-velero
   [default]
   aws_access_key_id=$AWS_ACCESS_KEY_ID
   aws_secret_access_key=$AWS_SECRET_ACCESS_KEY
   EOF

8. Log in to the OpenShift web console with a user that has the cluster-admin role.
9. From the Administrator perspective, expand the Operators navigation menu, and click OperatorHub.
10. In the search field, type oadp, and click the OADP Operator tile provided by Red Hat.
11. Click the Install button to show the operator details.
12. Accept the default values, click Install on the Install Operator page, and wait for the installation to finish.

13. After the operator installation finishes, from your workstation terminal, create a secret resource for OpenShift with your AWS credentials:

    $ oc create secret generic cloud-credentials -n openshift-adp --from-file cloud=credentials-velero

14. From the OpenShift web console, click the View Operator button.
15. Click Create instance on the DataProtectionApplication (DPA) tile.
16. On the Create DataProtectionApplication page, select YAML view.

17. Edit the following values in the resource file:

    1. Under the metadata section, replace velero-sample with velero.
    2. Under the spec.configuration.nodeAgent section, replace restic with kopia.
    3. Under the spec.configuration.velero section, add resourceTimeout: 10m.
    4. Under the spec.configuration.velero.defaultPlugins section, add - csi.
    5. Under the spec.snapshotLocations section, replace the us-west-2 value with your AWS regional value.
    6. Under the spec.backupLocations section, replace the us-east-1 value with your AWS regional value.
    7. Under the spec.backupLocations.objectStorage section, replace my-bucket-name with your bucket name. Replace velero with your bucket prefix name, if you use a different prefix.
18. Click the Create button.

Procedure

1. Open a terminal on your workstation, and log in to OpenShift:

   oc login --token=TOKEN --server=SERVER_URL_AND_PORT

   $ oc login --token=sha256~ZvFDBvoIYAbVECixS4-WmkN4RfnNd8Neh3y1WuiFPXC --server=https://example.com:6443

   Note

   You can find your login token and URL for use on the command line from the OpenShift web console. Log in to the OpenShift web console. Click your user name, and click Copy login command. Offer your user name and password again, if asked, and click Display Token to view the command.

2. Find and edit the VolumeSnapshotClass resource:

   $ oc get VolumeSnapshotClass -n openshift-adp
   $ oc edit VolumeSnapshotClass csi-aws-vsc -n openshift-adp

3. Update the following values in the resource file:

   1. Under the metadata.labels section, add the velero.io/csi-volumesnapshot-class: "true" label.
   2. Save your changes, and quit the editor.

4. Create a one-time, initial Backup job resource:

   $ cat <<EOF | oc apply -f -
   apiVersion: velero.io/v1
   kind: Backup
   metadata:
     name: rhtas-backup
     labels:
       velero.io/storage-location: velero-1
     namespace: openshift-adp
   spec:
     hooks: {}
     includedNamespaces:
     - trusted-artifact-signer
     includedResources: []
     excludedResources: []
     snapshotMoveData: true
     storageLocation: velero-1
     ttl: 720h0m0s
   EOF

   By default, all resources are backed up within the trusted-artifact-signer namespace. You can specify what resources you want to include or exclude by using the includeResources or excludedResources properties respectively.

   Important

   Depending on the storage class of the backup target, persistent volumes cannot be actively in-use for the backup to be successful.

5. Create a new schedule for regular backups to occur:

   $ cat << EOF | oc apply -f -
   apiVersion: velero.io/v1
   kind: Schedule
   metadata:
     name: BACKUP_JOB_NAME
     namespace: openshift-adp
   spec:
     schedule: USER_DEFINED_SCHEDULE
     template:
       hooks: {}
       includedNamespaces:
       - trusted-artifact-signer
       storageLocation: velero-1
       defaultVolumesToFsBackup: true
       ttl: 720h0m0s
   EOF

   Replace BACKUP_JOB_NAME with a job name, and replace USER_DEFINED_SCHEDULE with a cron-formatted expression for the schedule. For example, using a cron-formatted schedule of */10 * * * *, this backs up the trusted-artifact-signer namespace and its resources every 10 minutes.

   1. You can verify if this schedule is enabled, and when the last backup job ran. For example:

      $ oc get schedule -n openshift-adp
      
      NAME            STATUS    SCHEDULE       LASTBACKUP   AGE   PAUSED
      rhtas-backups   Enabled   0/10 * * * *   3m11s        16m

Procedure

1. Disable the RHTAS operator:

   $ oc scale deploy rhtas-operator-controller-manager --replicas=0 -n openshift-operators

2. Create the Restore resource:

   $ cat <<EOF | oc apply -f -
   apiVersion: velero.io/v1
   kind: Restore
   metadata:
     name: rhtas-restore
     namespace: openshift-adp
   spec:
     backupName: rhtas-backup
     includedResources: []
     restoreStatus:
       includedResources:
         - securesign.rhtas.redhat.com
         - trillian.rhtas.redhat.com
         - ctlog.rhtas.redhat.com
         - fulcio.rhtas.redhat.com
         - rekor.rhtas.redhat.com
         - tuf.rhtas.redhat.com
         - timestampauthority.rhtas.redhat.com
     excludedResources:
     - pod
     - deployment
     - nodes
     - route
     - service
     - replicaset
     - events
     - cronjob
     - events.events.k8s.io
     - backups.velero.io
     - restores.velero.io
     - resticrepositories.velero.io
     - pods
     - deployments
     restorePVs: true
     existingResourcePolicy: update
   EOF

3. If restoring your RHTAS data to a different OpenShift cluster, do the following steps.

   1. Delete the secret for the Trillian database:

      $ oc delete secret securesign-sample-trillian-db-tls
      $ oc delete pod trillian-db-xxx

      Note

      The RHTAS operator recreates the secret and restarts the pod.

   2. Run the restoreOwnerReferences.sh script.

4. Enable the RHTAS operator:

   $ oc scale deploy rhtas-operator-controller-manager --replicas=1 -n openshift-operators

   Important

   Immediately starting the RHTAS operator after starting the restore ensures the claim of the persistent volume.

Procedure

1. Download the tuftool binary from the OpenShift cluster to your workstation.

   Important

   Currently, the tuftool binary is only available for Linux operating systems on the x86_64 architecture.

   1. From the home page, click the ? icon, click Command line tools, go to the tuftool download section, and click the link for your platform.

   2. Open a terminal on your workstation, decompress the binary .gz file, and set the execution bit:

      $ gunzip tuftool-amd64.gz
      $ chmod +x tuftool-amd64

   3. Move and rename the binary to a location within your $PATH environment:

      $ sudo mv tuftool-amd64 /usr/local/bin/tuftool

2. Log in to OpenShift from the command line:

   oc login --token=TOKEN --server=SERVER_URL_AND_PORT

   $ oc login --token=sha256~ZvFDBvoIYAbVECixS4-WmkN4RfnNd8Neh3y1WuiFPXC --server=https://example.com:6443

   Note

   You can find your login token and URL for use on the command line from the OpenShift web console. Log in to the OpenShift web console. Click your user name, and click Copy login command. Offer your user name and password again, if asked, and click Display Token to view the command.

3. Switch to the RHTAS project:

   $ oc project trusted-artifact-signer

4. Configure your shell environment:

   $ export WORK="${HOME}/trustroot-example"
   $ export ROOT="${WORK}/root/root.json"
   $ export KEYDIR="${WORK}/keys"
   $ export INPUT="${WORK}/input"
   $ export TUF_REPO="${WORK}/tuf-repo"
   $ export TUF_SERVER_POD="$(oc get pod --selector=app.kubernetes.io/component=tuf --no-headers -o custom-columns=":metadata.name")"
   $ export TIMESTAMP_EXPIRATION="in 10 days"
   $ export SNAPSHOT_EXPIRATION="in 26 weeks"
   $ export TARGETS_EXPIRATION="in 26 weeks"
   $ export ROOT_EXPIRATION="in 26 weeks"

   Set the expiration durations according to your requirements.

5. Create a temporary TUF directory structure:

   $ mkdir -p "${WORK}/root/" "${KEYDIR}" "${INPUT}" "${TUF_REPO}"

6. Download the TUF contents to the temporary TUF directory structure:

   $ oc extract --to "${KEYDIR}/" secret/tuf-root-keys
   $ oc cp "${TUF_SERVER_POD}:/var/www/html" "${TUF_REPO}"
   $ cp "${TUF_REPO}/root.json" "${ROOT}"

7. You can update the timestamp, snapshot, and targets metadata all in one command:

   $ tuftool update \
     --root "${ROOT}" \
     --key "${KEYDIR}/timestamp.pem" \
     --key "${KEYDIR}/snapshot.pem" \
     --key "${KEYDIR}/targets.pem" \
     --timestamp-expires "${TIMESTAMP_EXPIRATION}" \
     --snapshot-expires "${SNAPSHOT_EXPIRATION}" \
     --targets-expires "${TARGETS_EXPIRATION}" \
     --outdir "${TUF_REPO}" \
     --metadata-url "file://${TUF_REPO}"

   Note

   You can also run the TUF metadata update on a subset of TUF metadata files. For example, the timestamp.json metadata file expires more often than the other metadata files. Therefore, you can just update the timestamp metadata file by running the following command:

   $ tuftool update \
     --root "${ROOT}" \
     --key "${KEYDIR}/timestamp.pem" \
     --timestamp-expires "${TIMESTAMP_EXPIRATION}" \
     --outdir "${TUF_REPO}" \
     --metadata-url "file://${TUF_REPO}"

8. Only update the root expiration date if it is about to expire:

   $ tuftool root expire "${ROOT}" "${ROOT_EXPIRATION}"

   Note

   You can skip this step if the root file is not close to expiring.

9. Update the root version:

   $ tuftool root bump-version "${ROOT}"

10. Sign the root metadata file again:

    $ tuftool root sign "${ROOT}" -k "${KEYDIR}/root.pem"

11. Set the new root version, and copy the root metadata file in place:

    $ export NEW_ROOT_VERSION=$(cat "${ROOT}" | jq -r ".signed.version")
    $ cp "${ROOT}" "${TUF_REPO}/root.json"
    $ cp "${ROOT}" "${TUF_REPO}/${NEW_ROOT_VERSION}.root.json"

12. Upload these changes to the TUF server:

    $ oc rsync "${TUF_REPO}/" "${TUF_SERVER_POD}:/var/www/html"

Procedure

1. Verify if you need to update and migrate the TUF repository manually.

   1. Check the TUF resource status, and the tuf-version annotation:

      $ oc get tuf securesign-sample

      If you see a status of Ready, then this means no manual update is needed. If you see a status of Failure, then a manual update is needed.

      $ oc get tuf securesign-sample -o jsonpath='{.metadata.annotations}'

      If you see rhtas.redhat.com/tuf-version":"v1, then this means no manual update is needed. If the annotation is missing, or has a different value, then this means a manual update is needed.

   2. Check for TUF error messages:

      $ oc get tuf securesign-sample -o jsonpath='{.status.conditions[?(@.type=="Ready")].message}'

      If you see error messages, such as the following, then you must update the TUF repository manually:

      cannot migrate TUF: root key secret not found

      or

      cannot migrate TUF: root key secret not specified

      Note

      Seeing these error messages does not mean a system outage. The TUF server continues to operate normally, and still handles requests. These error messages means that the TUF resource is not automatically reconciled by the RHTAS Operator, until the manual update and migration is done.

2. Configure your shell environment:

   $ export TUF_URL=$(oc get tuf -o jsonpath='{.items[0].status.url}')
   $ export REKOR_URL=$(oc get rekor -o jsonpath='{.items[0].status.url}')
   $ export FULCIO_URL=$(oc get fulcio -o jsonpath='{.items[0].status.url}')
   $ export TSA_URL=$(oc get timestampauthority -o jsonpath='{.items[0].status.url}')/api/v1/timestamp
   $ export CTLOG_URL=$(oc get svc ctlog -o jsonpath='{"https://"}{.metadata.name}{"."}{.metadata.namespace}{".svc/"}')"$(oc get securesign securesign-sample  -o jsonpath='{.spec.tuf.ctlog.prefix}')"
   $ export OIDC_ISSUERS=$(oc get securesign securesign-sample -o go-template='{{range $i, $val := .spec.fulcio.config.OIDCIssuers}}{{if $i}},{{end}}{{$val.Issuer}}{{end}}')
   $ export OPERATOR_NAME="rhtas.redhat.com"
   $ export WORKDIR="${HOME}/trustroot-migration"
   $ export TUF_REPO="${WORKDIR}/tuf-repo"
   $ export TUF_SERVER_POD="$(oc get pods -l app.kubernetes.io/component=tuf,\!job-name -o jsonpath='{.items[0].metadata.name}')"
   $ export KEYDIR="PATH_TO_YOUR_KEYS"

   Replase PATH_TO_YOUR_KEYS with the path to your signing keys.

3. Make a backup of your TUF repository:

   $ mkdir -p /tmp/tuf-backup
   $ oc rsync $TUF_SERVER_POD:/var/www/html/ /tmp/tuf-backup/

4. Download the TUF repository.

   1. Create a temporary TUF directory structure:

      $ mkdir -p "${TUF_REPO}"

   2. Copy the TUF contents to the temporary TUF directory structure:

      $ oc cp "${TUF_SERVER_POD}:/var/www/html" "${TUF_REPO}"

5. Create the update and migration script.

   1. Create an empty file:

      $ touch tuf_migration.sh

   2. Open the file for editing.
   3. Copy the script contents and paste them into the tuf_migration.sh file.
   4. Save and close the file.

   5. Make the script executable:

      $ chmod +x tuf_migration.sh

   6. Run the script to update the TUF repository:

      $ ./tuf_migration.sh

      Wait for the script to complete. Once it finishes, you should see the message Re-signing and upload complete.

6. Upload the changes back to the TUF server:

   $ oc rsync "${TUF_REPO}/" "${TUF_SERVER_POD}:/var/www/html"

7. Add an annotation to the TUF resource:

   $ oc annotate tuf securesign-sample rhtas.redhat.com/tuf-version=v1

   Now the RHTAS Operator can automatically reconcile the TUF deployment.

8. Verify that the TUF resource is in a Ready status:

   $ oc get tuf securesign-sample

Procedure

1. Download the rekor-cli binary from the OpenShift cluster to your workstation.

   1. Login to the OpenShift web console. From the home page, click the ? icon, click Command line tools, go to the rekor-cli download section, and click the link for your platform.

   2. Open a terminal on your workstation, decompress the binary .gz file, and set the execute bit:

      $ gunzip rekor-cli-amd64.gz
      $ chmod +x rekor-cli-amd64

   3. Move and rename the binary to a location within your $PATH environment:

      $ sudo mv rekor-cli-amd64 /usr/local/bin/rekor-cli

2. Download the tuftool binary from the OpenShift cluster to your workstation.

   Important

   The tuftool binary is only available for Linux operating systems.

   1. From the home page, click the ? icon, click Command line tools, go to the tuftool download section, and click the link for your platform.

   2. From a terminal on your workstation, decompress the binary .gz file, and set the execute bit:

      $ gunzip tuftool-amd64.gz
      $ chmod +x tuftool-amd64

   3. Move and rename the binary to a location within your $PATH environment:

      $ sudo mv tuftool-amd64 /usr/local/bin/tuftool

3. Log in to OpenShift from the command line:

   oc login --token=TOKEN --server=SERVER_URL_AND_PORT

   $ oc login --token=sha256~ZvFDBvoIYAbVECixS4-WmkN4RfnNd8Neh3y1WuiFPXC --server=https://example.com:6443

   Note

   You can find your login token and URL for use on the command line from the OpenShift web console. Log in to the OpenShift web console. Click your user name, and click Copy login command. Offer your user name and password again, if asked, and click Display Token to view the command.

4. Switch to the RHTAS project:

   $ oc project trusted-artifact-signer

5. Get the Rekor URL:

   $ export REKOR_URL=$(oc get rekor -o jsonpath='{.items[0].status.url}')

6. Get the log tree identifier for the active shard:

   $ export OLD_TREE_ID=$(rekor-cli loginfo --rekor_server $REKOR_URL --format json | jq -r .TreeID)

7. Set the log tree to the DRAINING state:

   $ oc run --image registry.redhat.io/rhtas/updatetree-rhel9:1.1.0 --restart=Never --attach=true --rm=true -q -- updatetree --admin_server=trillian-logserver.trusted-artifact-signer.svc:8091 --tree_id=${OLD_TREE_ID} --tree_state=DRAINING --tls_cert_file=/var/run/secrets/kubernetes.io/serviceaccount/service-ca.crt

   While draining, the tree log will not accept any new entries. Watch and wait for the queue to empty.

   Important

   You must wait for the queues to be empty before proceeding to the next step. If leaves are still integrating while draining, then freezing the log tree during this process can cause the log path to exceed the maximum merge delay (MMD) threshold.

8. Freeze the log tree:

   $ oc run --image registry.redhat.io/rhtas/updatetree-rhel9:1.1.0 --restart=Never --attach=true --rm=true -q -- updatetree --admin_server=trillian-logserver.trusted-artifact-signer.svc:8091 --tree_id=${OLD_TREE_ID} --tree_state=FROZEN --tls_cert_file=/var/run/secrets/kubernetes.io/serviceaccount/service-ca.crt

9. Get the length of the frozen log tree:

   $ export OLD_SHARD_LENGTH=$(rekor-cli loginfo --rekor_server $REKOR_URL --format json | jq -r .ActiveTreeSize)

10. Get Rekor’s public key for the old shard:

    $ export OLD_PUBLIC_KEY=$(curl -s $REKOR_URL/api/v1/log/publicKey | base64 | tr -d '\n')

11. Create a new log tree:

    $ export NEW_TREE_ID=$(oc run createtree --image registry.redhat.io/rhtas/createtree-rhel9:1.1.0 --restart=Never --attach=true --rm=true -q -- -logtostderr=false --admin_server=trillian-logserver.trusted-artifact-signer.svc:8091 --display_name=rekor-tree --tls_cert_file=/var/run/secrets/kubernetes.io/serviceaccount/service-ca.crt)

    Now you have two log trees, one frozen tree, and a new tree that will become the active shard.

12. Create a new private key:

    $ openssl ecparam -genkey -name prime256v1 -noout -out new-rekor.pem

    Important

    The new key must have a unique file name.

13. Create a new secret resource with the new signer key:

    $ oc create secret generic rekor-signer-key --from-file=private=new-rekor.pem

14. Update the Securesign Rekor configuration with the new tree identifier and the old sharding information:

    $ read -r -d '' SECURESIGN_PATCH_1 <<EOF
    [
        {
            "op": "replace",
            "path": "/spec/rekor/treeID",
            "value": $NEW_TREE_ID
        },
        {
            "op": "add",
            "path": "/spec/rekor/sharding/-",
            "value": {
                "treeID": $OLD_TREE_ID,
                "treeLength": $OLD_SHARD_LENGTH,
                "encodedPublicKey": "$OLD_PUBLIC_KEY"
            }
        },
        {
            "op": "replace",
            "path": "/spec/rekor/signer/keyRef",
            "value": {"name": "rekor-signer-key", "key": "private"}
        }
    ]
    EOF

    Note

    If you have /spec/rekor/signer/keyPasswordRef set with a value, then create a new separate update to remove it:

    $ read -r -d '' SECURESIGN_PATCH_2 <<EOF
    [
        {
            "op": "remove",
            "path": "/spec/rekor/signer/keyPasswordRef"
        }
    ]
    EOF

    Apply this update after applying the first update.

15. Update the Securesign instance:

    $ oc patch Securesign securesign-sample --type='json' -p="$SECURESIGN_PATCH_1"

16. Wait for the Rekor server to redeploy with the new signer key:

    $ oc wait pod -l app.kubernetes.io/name=rekor-server --for=condition=Ready

17. Get the new public key:

    $ export NEW_KEY_NAME=new-rekor.pub
    $ curl $(oc get rekor -o jsonpath='{.items[0].status.url}')/api/v1/log/publicKey -o $NEW_KEY_NAME

18. Configure The Update Framework (TUF) service to use the new Rekor public key.

    1. Configure your shell environment:

       $ export WORK="${HOME}/trustroot-example"
       $ export ROOT="${WORK}/root/root.json"
       $ export KEYDIR="${WORK}/keys"
       $ export INPUT="${WORK}/input"
       $ export TUF_REPO="${WORK}/tuf-repo"
       $ export TUF_SERVER_POD="$(oc get pods -l app.kubernetes.io/component=tuf,\!job-name -o jsonpath='{.items[0].metadata.name}')"

    2. Create a temporary TUF directory structure:

       $ mkdir -p "${WORK}/root/" "${KEYDIR}" "${INPUT}" "${TUF_REPO}"

    3. Download the TUF contents to the temporary TUF directory structure:

       $ oc extract --to "${KEYDIR}/" secret/tuf-root-keys
       $ oc cp "${TUF_SERVER_POD}:/var/www/html" "${TUF_REPO}"
       $ cp "${TUF_REPO}/root.json" "${ROOT}"

    4. Find the active Rekor signer key file name. Open the latest target file, for example, 1.target.json, within the local TUF repository. In this file you will find the active Rekor signer key file name, for example, rekor.pub. Set an environment variable with this active Rekor signer key file name:

       $ export ACTIVE_KEY_NAME=rekor.pub

    5. Update the Rekor signer key with the old public key:

       $ echo $OLD_PUBLIC_KEY | base64 -d > $ACTIVE_KEY_NAME

    6. Expire the old Rekor signer key:

       $ tuftool rhtas \
         --root "${ROOT}" \
         --key "${KEYDIR}/snapshot.pem" \
         --key "${KEYDIR}/targets.pem" \
         --key "${KEYDIR}/timestamp.pem" \
         --set-rekor-target "${ACTIVE_KEY_NAME}" \
         --rekor-uri "${REKOR_URL}" \
         --rekor-status "Expired" \
         --outdir "${TUF_REPO}" \
         --metadata-url "file://${TUF_REPO}"

    7. Add the new Rekor signer key:

       $ tuftool rhtas \
         --root "${ROOT}" \
         --key "${KEYDIR}/snapshot.pem" \
         --key "${KEYDIR}/targets.pem" \
         --key "${KEYDIR}/timestamp.pem" \
         --set-rekor-target "${NEW_KEY_NAME}" \
         --rekor-uri "${REKOR_URL}" \
         --outdir "${TUF_REPO}" \
         --metadata-url "file://${TUF_REPO}"

    8. Upload these changes to the TUF server:

       $ oc rsync "${TUF_REPO}/" "${TUF_SERVER_POD}:/var/www/html"

    9. Delete the working directory:

       $ rm -r $WORK

19. Update the cosign configuration with the updated TUF configuration:

    $ cosign initialize --mirror=$TUF_URL --root=$TUF_URL/root.json

    Now, you are ready to sign and verify your artifacts with the new Rekor signer key.

Procedure

1. Download the tuftool binary from the OpenShift cluster to your workstation.

   Important

   Currently, the tuftool binary is only available for Linux operating systems on the x86_64 architecture.

   1. From the home page, click the ? icon, click Command line tools, go to the tuftool download section, and click the link for your platform.

   2. Open a terminal on your workstation, decompress the binary .gz file, and set the execution bit:

      $ unzip tuftool-amd64.gz
      $ chmod +x tuftool-amd64

   3. Move and rename the binary to a location within your $PATH environment:

      $ sudo mv tuftool-amd64 /usr/local/bin/tuftool

2. Log in to OpenShift from the command line:

   oc login --token=TOKEN --server=SERVER_URL_AND_PORT

   $ oc login --token=sha256~ZvFDBvoIYAbVECixS4-WmkN4RfnNd8Neh3y1WuiFPXC --server=https://example.com:6443

   Note

   You can find your login token and URL for use on the command line from the OpenShift web console. Log in to the OpenShift web console. Click your user name, and click Copy login command. Offer your user name and password again, if asked, and click Display Token to view the command.

3. Switch to the RHTAS project:

   $ oc project trusted-artifact-signer

4. Make a backup of the current CT log configuration, and keys:

   $ export SERVER_CONFIG_NAME=$(oc get ctlog -o jsonpath='{.items[0].status.serverConfigRef.name}')
   $ oc get secret $SERVER_CONFIG_NAME -o jsonpath="{.data.config}" | base64 --decode > config.txtpb
   $ oc get secret $SERVER_CONFIG_NAME -o jsonpath="{.data.fulcio-0}" | base64 --decode > fulcio-0.pem
   $ oc get secret $SERVER_CONFIG_NAME -o jsonpath="{.data.private}" | base64 --decode > private.pem
   $ oc get secret $SERVER_CONFIG_NAME -o jsonpath="{.data.public}" | base64 --decode > public.pem

5. Capture the current tree identifier:

   $ export OLD_TREE_ID=$(oc get ctlog -o jsonpath='{.items[0].status.treeID}')

6. Set the log tree to the DRAINING state:

   $ oc run --image registry.redhat.io/rhtas/updatetree-rhel9:1.1.0 --restart=Never --attach=true --rm=true -q -- updatetree --admin_server=trillian-logserver.trusted-artifact-signer.svc:8091 --tree_id=${OLD_TREE_ID} --tree_state=DRAINING --tls_cert_file=/var/run/secrets/kubernetes.io/serviceaccount/service-ca.crt

   While draining, the tree log will not accept any new entries. Watch and wait for the queue to empty.

   Important

   You must wait for the queues to be empty before proceeding to the next step. If leaves are still integrating while draining, then freezing the log tree during this process can cause the log path to exceed the maximum merge delay (MMD) threshold.

7. Once the queue has been fully drained, freeze the log:

   $ oc run --image registry.redhat.io/rhtas/updatetree-rhel9:1.1.0 --restart=Never --attach=true --rm=true -q -- updatetree --admin_server=trillian-logserver.trusted-artifact-signer.svc:8091 --tree_id=${OLD_TREE_ID} --tree_state=FROZEN --tls_cert_file=/var/run/secrets/kubernetes.io/serviceaccount/service-ca.crt

8. Create a new Merkle tree, and capture the new tree identifier:

   $ export NEW_TREE_ID=$(kubectl run createtree --image registry.redhat.io/rhtas/createtree-rhel9:1.1.0 --restart=Never --attach=true --rm=true -q -- -logtostderr=false --admin_server=trillian-logserver.trusted-artifact-signer.svc:8091 --display_name=ctlog-tree --tls_cert_file=/var/run/secrets/kubernetes.io/serviceaccount/service-ca.crt)

9. Generate a new certificate, along with new public and private keys:

   $ openssl ecparam -genkey -name prime256v1 -noout -out new-ctlog.pem
   $ openssl ec -in new-ctlog.pem -pubout -out new-ctlog-public.pem
   $ openssl ec -in new-ctlog.pem -out new-ctlog.pass.pem -des3 -passout pass:"CHANGE_ME"

   Replace CHANGE_ME with a new password.

   Important

   The certificate and new keys must have unique file names.

10. Update the CT log configuration.

    1. Open the config.txtpb file for editing.

    2. For the frozen log, add the not_after_limit field to the frozen log entry, rename the prefix value to a unique name, and replace the old path to the private key with /ctfe-keys/private-0:

       ...
       log_configs:{
         # frozen log
         config:{
           log_id:2066075212146181968
           prefix:"trusted-artifact-signer-0"
           roots_pem_file:"/ctfe-keys/fulcio-0"
           private_key:{[type.googleapis.com/keyspb.PEMKeyFile]:{path:"/ctfe-keys/private-0" password:"Example123"}}
           public_key:{der:"0Y0\x13\x06\x07*\x86H\xce=\x02\x01\x06\x08*\x86H\xce=\x03\x01\x07\x03B\x00\x04)'.\xffUJ\xe2s)\xefR\x8a\xfcO\xdcewȶy\xa7\x9d<\x13\xb0\x1c\x99\x96\xe4'\xe3v\x07:\xc8I+\x08J\x9d\x8a\xed\x06\xe4\xaeI:q\x98\xf4\xbc<o4VD\x0cr\xf9\x9c\xecxT\x84"}
           not_after_limit:{seconds:1728056285 nanos:012111000}
           ext_key_usages:"CodeSigning"
           log_backend_name:"trillian"
         }

       Note

       You can get the current time value for seconds and nanoseconds, by running the following commands: date +%s, and date +%N.

       Important

       The not_after_limit field defines the end of the timestamp range for the frozen log only. Certificates beyond this point in time are no longer accepted for inclusion in this log.

    3. Copy and paste the frozen log config block, appending it to the configuration file to create a new entry.

    4. Change the following lines in the new config block. Set the log_id to the new tree identifier, change the prefix to trusted-artifact-signer, change the private_key path to /ctfe-keys/private, remove the public_key line, and change not_after_limit to not_after_start and set the timestamp range:

       ...
       log_configs:{
         # frozen log
         ...
         # new active log
         config:{
       	  log_id: NEW_TREE_ID
       	  prefix:"trusted-artifact-signer"
       	  roots_pem_file:"/ctfe-keys/fulcio-0"
       	  private_key:{[type.googleapis.com/keyspb.PEMKeyFile]:{path:"/ctfe-keys/private" password:"CHANGE_ME"}}
       	  ext_key_usages:"CodeSigning"
       	  not_after_start:{seconds:1713201754 nanos:155663000}
       	  log_backend_name:"trillian"
         }

       Add the NEW_TREE_ID, and replace CHANGE_ME with the new private key password. The password here must match the password used for generating the new private and public keys.

       Important

       The not_after_start field defines the beginning of the timestamp range inclusively. This means the log will start accepting certificates at this point in time.

11. Create a new secret resource:

    $ oc create secret generic ctlog-config \
    --from-file=config=config.txtpb \
    --from-file=private=new-ctlog.pass.pem \
    --from-file=public=new-ctlog-public.pem \
    --from-file=fulcio-0=fulcio-0.pem \
    --from-file=private-0=private.pem \
    --from-file=public-0=public.pem \
    --from-literal=password=CHANGE_ME

    Replace CHANGE_ME with the new private key password.

12. Configure The Update Framework (TUF) service to use the new CT log public key.

    1. Configure your shell environment:

       $ export WORK="${HOME}/trustroot-example"
       $ export ROOT="${WORK}/root/root.json"
       $ export KEYDIR="${WORK}/keys"
       $ export INPUT="${WORK}/input"
       $ export TUF_REPO="${WORK}/tuf-repo"
       $ export TUF_SERVER_POD="$(oc get pods -l app.kubernetes.io/component=tuf,\!job-name -o jsonpath='{.items[0].metadata.name}')"

    2. Create a temporary TUF directory structure:

       $ mkdir -p "${WORK}/root/" "${KEYDIR}" "${INPUT}" "${TUF_REPO}"

    3. Download the TUF contents to the temporary TUF directory structure:

       $ oc extract --to "${KEYDIR}/" secret/tuf-root-keys
       $ oc cp "${TUF_SERVER_POD}:/var/www/html" "${TUF_REPO}"
       $ cp "${TUF_REPO}/root.json" "${ROOT}"

    4. Find the active CT log public key file name. Open the latest target file, for example, 1.targets.json, within the local TUF repository. In this target file you will find the active CT log public key file name, for example, ctfe.pub. Set an environment variable with this active CT log public key file name:

       $ export ACTIVE_CTFE_NAME=ctfe.pub

    5. Extract the active CT log public key from OpenShift:

       $ oc get secret $(oc get ctlog securesign-sample -o jsonpath='{.status.publicKeyRef.name}') -o jsonpath='{.data.public}' | base64 -d > $ACTIVE_CTFE_NAME

    6. Expire the old CT log signer key:

       $ tuftool rhtas \
         --root "${ROOT}" \
         --key "${KEYDIR}/snapshot.pem" \
         --key "${KEYDIR}/targets.pem" \
         --key "${KEYDIR}/timestamp.pem" \
         --set-ctlog-target "$ACTIVE_CTFE_NAME" \
         --ctlog-uri "https://ctlog.rhtas" \
         --ctlog-status "Expired" \
         --outdir "${TUF_REPO}" \
         --metadata-url "file://${TUF_REPO}"

    7. Add the new CT log signer key:

       $ tuftool rhtas \
         --root "${ROOT}" \
         --key "${KEYDIR}/snapshot.pem" \
         --key "${KEYDIR}/targets.pem" \
         --key "${KEYDIR}/timestamp.pem" \
         --set-ctlog-target "new-ctlog-public.pem" \
         --ctlog-uri "https://ctlog.rhtas" \
         --outdir "${TUF_REPO}" \
         --metadata-url "file://${TUF_REPO}"

    8. Upload these changes to the TUF server:

       $ oc rsync "${TUF_REPO}/" "${TUF_SERVER_POD}:/var/www/html"

13. Update the Securesign CT log configuration with the new tree identifier:

    read -r -d '' SECURESIGN_PATCH <<EOF
    [
    	{
        	"op": "replace",
        	"path": "/spec/ctlog/serverConfigRef",
        	"value": {"name": "ctlog-config"}
    	},
        {
            "op": "replace",
            "path": "/spec/ctlog/treeID",
            "value": $NEW_TREE_ID
        },
    	{
        	"op": "replace",
        	"path": "/spec/ctlog/privateKeyRef",
        	"value": {"name": "ctlog-config", "key": "private"}
    	},
        {
            "op": "replace",
            "path": "/spec/ctlog/privateKeyPasswordRef",
            "value": {"name": "ctlog-config", "key": "password"}
        },
    	{
        	"op": "replace",
        	"path": "/spec/ctlog/publicKeyRef",
        	"value": {"name": "ctlog-config", "key": "public"}
    	}
    ]
    EOF

14. Patch the Securesign instance:

    $ oc patch Securesign securesign-sample --type='json' -p="$SECURESIGN_PATCH"

15. Wait for the CT log server to redeploy:

    $ oc wait pod -l app.kubernetes.io/name=ctlog --for=condition=Ready

16. Delete the working directory:

    $ rm -r $WORK

17. Update the cosign configuration with the updated TUF configuration:

    $ cosign initialize --mirror=$TUF_URL --root=$TUF_URL/root.json

    Now, you are ready to sign and verify your artifacts with the new CT log signer key.

Procedure

1. Download the tuftool binary from the OpenShift cluster to your workstation.

   Important

   Currently, the tuftool binary is only available for Linux operating systems on the x86_64 architecture.

   1. From the home page, click the ? icon, click Command line tools, go to the tuftool download section, and click the link for your platform.

   2. Open a terminal on your workstation, decompress the binary .gz file, and set the execution bit:

      $ gunzip tuftool-amd64.gz
      $ chmod +x tuftool-amd64

   3. Move and rename the binary to a location within your $PATH environment:

      $ sudo mv tuftool-amd64 /usr/local/bin/tuftool

2. Log in to OpenShift from the command line:

   oc login --token=TOKEN --server=SERVER_URL_AND_PORT

   $ oc login --token=sha256~ZvFDBvoIYAbVECixS4-WmkN4RfnNd8Neh3y1WuiFPXC --server=https://example.com:6443

   Note

   You can find your login token and URL for use on the command line from the OpenShift web console. Log in to the OpenShift web console. Click your user name, and click Copy login command. Offer your user name and password again, if asked, and click Display Token to view the command.

3. Switch to the RHTAS project:

   $ oc project trusted-artifact-signer

4. Generate a new certificate, along with new public and private keys:

   $ openssl ecparam -genkey -name prime256v1 -noout -out new-fulcio.pem
   $ openssl ec -in new-fulcio.pem -pubout -out new-fulcio-public.pem
   $ openssl ec -in new-fulcio.pem -out new-fulcio.pass.pem -des3 -passout pass:"CHANGE_ME"
   $ openssl req -new -x509 -key new-fulcio.pass.pem -out new-fulcio.cert.pem

   Replace CHANGE_ME with a new password.

   Important

   The certificate and new keys must have unique file names.

5. Create a new secret:

   $ oc create secret generic fulcio-config \
   --from-file=private=new-fulcio.pass.pem \
   --from-file=cert=new-fulcio.cert.pem \
   --from-literal=password=CHANGE_ME

   Replace CHANGE_ME with a new password.

   Note

   The password here must match the password used for generating the new private and public keys.

6. Configure The Update Framework (TUF) service to use the new Fulcio certificate.

   1. Set up your shell environment:

      $ export WORK="${HOME}/trustroot-example"
      $ export ROOT="${WORK}/root/root.json"
      $ export KEYDIR="${WORK}/keys"
      $ export INPUT="${WORK}/input"
      $ export FULCIO_URL=$(oc get fulcio securesign-sample -o jsonpath='{.status.url}')
      $ export TUF_REPO="${WORK}/tuf-repo"
      $ export TUF_SERVER_POD="$(oc get pods -l app.kubernetes.io/component=tuf,\!job-name -o jsonpath='{.items[0].metadata.name}')"

   2. Create a temporary TUF directory structure:

      $ mkdir -p "${WORK}/root/" "${KEYDIR}" "${INPUT}" "${TUF_REPO}"

   3. Download the TUF contents to the temporary TUF directory structure:

      $ oc extract --to "${KEYDIR}/" secret/tuf-root-keys
      $ oc cp "${TUF_SERVER_POD}:/var/www/html" "${TUF_REPO}"
      $ cp "${TUF_REPO}/root.json" "${ROOT}"

   4. Find the active Fulcio certificate file name. Open the latest target file, for example, 1.targets.json, within the local TUF repository. In this file you will find the active Fulcio certificate file name, for example, fulcio_v1.crt.pem. Set an environment variable with this active Fulcio certificate file name:

      $ export ACTIVE_CERT_NAME=fulcio_v1.crt.pem

   5. Extract the active Fulcio certificate from OpenShift:

      $ oc get secret $(oc get fulcio securesign-sample -o jsonpath='{.status.certificate.caRef.name}') -o jsonpath='{.data.cert}' | base64 -d > $ACTIVE_CERT_NAME

   6. Expire the old certificate:

      $ tuftool rhtas \
        --root "${ROOT}" \
        --key "${KEYDIR}/snapshot.pem" \
        --key "${KEYDIR}/targets.pem" \
        --key "${KEYDIR}/timestamp.pem" \
        --set-fulcio-target "$ACTIVE_CERT_NAME" \
        --fulcio-uri "${FULCIO_URL}" \
        --fulcio-status "Expired" \
        --outdir "${TUF_REPO}" \
        --metadata-url "file://${TUF_REPO}"

   7. Add the new Fulcio certificate:

      $ tuftool rhtas \
        --root "${ROOT}" \
        --key "${KEYDIR}/snapshot.pem" \
        --key "${KEYDIR}/targets.pem" \
        --key "${KEYDIR}/timestamp.pem" \
        --set-fulcio-target "new-fulcio.cert.pem" \
        --fulcio-uri "${FULCIO_URL}" \
        --oidc-uri "${OIDC_URI}" \
        --outdir "${TUF_REPO}" \
        --metadata-url "file://${TUF_REPO}"

   8. Upload these changes to the TUF server:

      $ oc rsync "${TUF_REPO}/" "${TUF_SERVER_POD}:/var/www/html"

   9. Delete the working directory:

      $ rm -r $WORK

7. Update the Securesign Fulcio configuration:

   $ read -r -d '' SECURESIGN_PATCH <<EOF
   [
   {
       "op": "replace",
       "path": "/spec/fulcio/certificate/privateKeyRef",
       "value": {"name": "fulcio-config", "key": "private"}
   },
   {
       "op": "replace",
       "path": "/spec/fulcio/certificate/privateKeyPasswordRef",
       "value": {"name": "fulcio-config", "key": "password"}
   },
   {
       "op": "replace",
       "path": "/spec/fulcio/certificate/caRef",
       "value": {"name": "fulcio-config", "key": "cert"}
   },
   {
       "op": "replace",
       "path": "/spec/ctlog/rootCertificates",
       "value": [{"name": "fulcio-config", "key": "cert"}]
   }
   ]
   EOF

8. Patch the Securesign instance:

   $ oc patch Securesign securesign-sample --type='json' -p="$SECURESIGN_PATCH"

9. Wait for the Fulcio server to redeploy:

   $ oc wait pod -l app.kubernetes.io/name=fulcio-server --for=condition=Ready
   $ oc wait pod -l app.kubernetes.io/name=ctlog --for=condition=Ready

10. Update the cosign configuration with the updated TUF configuration:

    $ cosign initialize --mirror=$TUF_URL --root=$TUF_URL/root.json

    Now, you are ready to sign and verify your artifacts with the new Fulcio certificate.

Procedure

1. Download the tuftool binary from the OpenShift cluster to your workstation.

   Important

   Currently, the tuftool binary is only available for Linux operating systems on the x86_64 architecture.

   1. From the home page, click the ? icon, click Command line tools, go to the tuftool download section, and click the link for your platform.

   2. Open a terminal on your workstation, decompress the binary .gz file, and set the execution bit:

      $ gunzip tuftool-amd64.gz
      $ chmod +x tuftool-amd64

   3. Move and rename the binary to a location within your $PATH environment:

      $ sudo mv tuftool-amd64 /usr/local/bin/tuftool

2. Log in to OpenShift from the command line:

   oc login --token=TOKEN --server=SERVER_URL_AND_PORT

   $ oc login --token=sha256~ZvFDBvoIYAbVECixS4-WmkN4RfnNd8Neh3y1WuiFPXC --server=https://example.com:6443

   Note

   You can find your login token and URL for use on the command line from the OpenShift web console. Log in to the OpenShift web console. Click your user name, and click Copy login command. Offer your user name and password again, if asked, and click Display Token to view the command.

3. Switch to the RHTAS project:

   $ oc project trusted-artifact-signer

4. Generate a new certificate chain, and a new signer key.

   Important

   The new certificate and keys must have unique file names.

   1. Create a temporary working directory:

      $ mkdir certs && cd certs

   2. Create the root certificate authority (CA) private key, and set a password:

      $ openssl req -x509 -newkey rsa:2048 -days 365 -sha256 -nodes \
      -keyout rootCA.key.pem -out rootCA.crt.pem \
      -passout pass:"CHANGE_ME" \
      -subj "/C=CC/ST=state/L=Locality/O=RH/OU=RootCA/CN=RootCA" \
      -addext "basicConstraints=CA:true" -addext "keyUsage=cRLSign, keyCertSign"

      Replace CHANGE_ME with a new password.

   3. Create the intermediate CA private key and certificate signing request (CSR), and set a password:

      $ openssl req -newkey rsa:2048 -sha256 \
      -keyout intermediateCA.key.pem -out intermediateCA.csr.pem \
      -passout pass:"CHANGE_ME" \
      -subj "/C=CC/ST=state/L=Locality/O=RH/OU=IntermediateCA/CN=IntermediateCA"

      Replace CHANGE_ME with a new password.

   4. Sign the intermediate CA certificate with the root CA:

      $ openssl x509 -req -in intermediateCA.csr.pem -CA rootCA.crt.pem -CAkey rootCA.key.pem \
      -CAcreateserial -out intermediateCA.crt.pem -days 365 -sha256 \
      -extfile <(echo -e "basicConstraints=CA:true\nkeyUsage=cRLSign, keyCertSign\nextendedKeyUsage=critical,timeStamping") \
      -passin pass:"CHANGE_ME"

      Replace CHANGE_ME with the root CA private key password to sign the intermediate CA certificate.

   5. Create the leaf CA private key and CSR, and set a password:

      $ openssl req -newkey rsa:2048 -sha256 \
      -keyout leafCA.key.pem -out leafCA.csr.pem \
      -passout pass:"CHANGE_ME" \
      -subj "/C=CC/ST=state/L=Locality/O=RH/OU=LeafCA/CN=LeafCA"

   6. Sign the leaf CA certificate with the intermediate CA:

      $ openssl x509 -req -in leafCA.csr.pem -CA intermediateCA.crt.pem -CAkey intermediateCA.key.pem \
        -CAcreateserial -out leafCA.crt.pem -days 365 -sha256 \
        -extfile <(echo -e "basicConstraints=CA:false\nkeyUsage=cRLSign, keyCertSign\nextendedKeyUsage=critical,timeStamping") \
        -passin pass:"CHANGE_ME"

      Replace CHANGE_ME with the intermediate CA private key password to sign the leaf CA certificate.

   7. Create the certificate chain by combining the newly created certificates together:

      $ cat leafCA.crt.pem intermediateCA.crt.pem rootCA.crt.pem > new-tsa.certchain.pem

5. Create a new secret resource with the signer key:

   $ oc create secret generic rotated-signer-key --from-file=rotated-signer-key=certs/leafCA.key.pem

6. Create a new secret resource with the new certificate chain:

   $ oc create secret generic rotated-cert-chain --from-file=rotated-cert-chain=certs/new-tsa.certchain.pem

7. Create a new secret resource with for the password:

   $ oc create secret generic rotated-password --from-literal=rotated-password=CHANGE_ME

   Replace CHANGE_ME with the intermediate CA private key password.

8. Find your active TSA certificate file name, the TSA URL string, and configure your shell environment with these values:

   $ export ACTIVE_CERT_CHAIN_NAME=tsa.certchain.pem
   $ export TSA_URL=$(oc get timestampauthority securesign-sample -o jsonpath='{.status.url}')/api/v1/timestamp
   $ curl $TSA_URL/certchain -o $ACTIVE_CERT_CHAIN_NAME

9. Update the Securesign TSA configuration:

   $ read -r -d '' SECURESIGN_PATCH <<EOF
   [
       {
           "op": "replace",
           "path": "/spec/tsa/signer/certificateChain",
           "value": {
               "certificateChainRef" : {"name": "rotated-cert-chain", "key": "rotated-cert-chain"}
           }
       },
       {
           "op": "replace",
           "path": "/spec/tsa/signer/file",
           "value": {
                   "privateKeyRef": {"name": "rotated-signer-key", "key": "rotated-signer-key"},
                   "passwordRef": {"name": "rotated-password", "key": "rotated-password"}
               }
       }
   ]
   EOF

10. Patch the Securesign instance:

    $ oc patch Securesign securesign-sample --type='json' -p="$SECURESIGN_PATCH"

11. Wait for the TSA server to redeploy with the new signer key and certificate chain:

    $ oc get pods -w -l app.kubernetes.io/name=tsa-server

12. Get the new certificate chain:

    $ export NEW_CERT_CHAIN_NAME=new-tsa.certchain.pem
    $ curl $TSA_URL/certchain -o $NEW_CERT_CHAIN_NAME

13. Configure The Update Framework (TUF) service to use the new TSA certificate chain.

    1. Set up your shell environment:

       $ export WORK="${HOME}/trustroot-example"
       $ export ROOT="${WORK}/root/root.json"
       $ export KEYDIR="${WORK}/keys"
       $ export INPUT="${WORK}/input"
       $ export TUF_REPO="${WORK}/tuf-repo"
       $ export TUF_SERVER_POD="$(oc get pods -l app.kubernetes.io/component=tuf,\!job-name -o jsonpath='{.items[0].metadata.name}')"

    2. Create a temporary TUF directory structure:

       $ mkdir -p "${WORK}/root/" "${KEYDIR}" "${INPUT}" "${TUF_REPO}"

    3. Download the TUF contents to the temporary TUF directory structure:

       $ oc extract --to "${KEYDIR}/" secret/tuf-root-keys
       $ oc cp "${TUF_SERVER_POD}:/var/www/html" "${TUF_REPO}"
       $ cp "${TUF_REPO}/root.json" "${ROOT}"

    4. Expire the old TSA certificate:

       $ tuftool rhtas \
         --root "${ROOT}" \
         --key "${KEYDIR}/snapshot.pem" \
         --key "${KEYDIR}/targets.pem" \
         --key "${KEYDIR}/timestamp.pem" \
         --set-tsa-target "$ACTIVE_CERT_CHAIN_NAME" \
         --tsa-uri "$TSA_URL" \
         --tsa-status "Expired" \
         --outdir "${TUF_REPO}" \
         --metadata-url "file://${TUF_REPO}"

    5. Add the new TSA certificate:

       $ tuftool rhtas \
         --root "${ROOT}" \
         --key "${KEYDIR}/snapshot.pem" \
         --key "${KEYDIR}/targets.pem" \
         --key "${KEYDIR}/timestamp.pem" \
         --set-tsa-target "$NEW_CERT_CHAIN_NAME" \
         --tsa-uri "$TSA_URL" \
         --outdir "${TUF_REPO}" \
         --metadata-url "file://${TUF_REPO}"

    6. Upload these changes to the TUF server:

       $ oc rsync "${TUF_REPO}/" "${TUF_SERVER_POD}:/var/www/html"

    7. Delete the working directory:

       $ rm -r $WORK

14. Update the cosign configuration with the updated TUF configuration:

    $ cosign initialize --mirror=$TUF_URL --root=$TUF_URL/root.json

    Now, you are ready to sign and verify your artifacts that uses the new TSA signer key, and certificate.

Procedure

1. Log in to the OpenShift web console with a user that has the cluster-admin role.
2. From the Administrator perspective, expand the Operators navigation menu, and click OperatorHub.
3. In the search field, type policy-controller, and click the Policy Controller Operator tile provided by Red Hat.
4. Click the Install button to show the operator details.
5. Accept the default values, click Install on the Install Operator page, and wait for the installation to finish.
6. Once the installation finishes, you can create the Policy Controller resources.

Procedure

1. Open a terminal on your workstation, and log in to OpenShift:

   oc login --token=TOKEN --server=SERVER_URL_AND_PORT

   $ oc login --token=sha256~ZvFDBvoIYAbVECixS4-WmkN4RfnNd8Neh3y1WuiFPXC --server=https://example.com:6443

   Note

   You can find your login token and URL for use on the command line from the OpenShift web console. Log in to the OpenShift web console. Click your user name, and click Copy login command. Offer your user name and password again, if asked, and click Display Token to view the command.

2. Create and switch to the policy-controller-operator namespace:

   $ oc new-project policy-controller-operator ; oc project policy-controller-operator

3. Create a basic Policy Controller resource.

   1. Configure the Policy Controller to watch your namespaces that match the defined label selector under spec.policy-controller.webhook.namespaceSelector.matchExpressions:

      ...
      spec:
        policy-controller:
          ...
          webhook:
            ...
            namespaceSelector:
              matchExpressions:
                - key: policy.rhtas.com/include
                  operator: In
                  values: ["true"]
      ...

      $ cat <<EOF | oc apply -f -
      apiVersion: rhtas.charts.redhat.com/v1alpha1
      kind: PolicyController
      metadata:
        name: policycontroller-sample
      spec:
        policy-controller:
          cosign:
            webhookName: "policy.rhtas.com"
          webhook:
            name: webhook
            extraArgs:
              webhook-name: policy.rhtas.com
              mutating-webhook-name: defaulting.clusterimagepolicy.rhtas.com
              validating-webhook-name: validating.clusterimagepolicy.rhtas.com
            failurePolicy: Fail
            namespaceSelector:
              matchExpressions:
                - key: policy.rhtas.com/include
                  operator: In
                  values: ["true"]
            webhookNames:
              defaulting: "defaulting.clusterimagepolicy.rhtas.com"
              validating: "validating.clusterimagepolicy.rhtas.com"
      EOF

      Important

      You must create this resource in the policy-controller-operator namespace.

   2. Add the policy.rhtas.com/include: "true" label to the namespace that you want watched by the Policy Controller:

      apiVersion: v1
      kind: Namespace
      metadata:
        labels:
          policy.rhtas.com/include: "true"
        name: example-namespace

   3. If you have a custom Certificate Authority (CA) bundle or self-signed certificates, then you can add your ConfigMap name and key under the spec.policy-controller.webhook.registryCaBundle section of the Policy Controller resource:

      ...
      spec:
        policy-controller:
          ...
          webhook:
            registryCaBundle:
              name: CONFIGMAP_NAME
              key: CA_BUNDLE_KEY
      ...

4. Create a Trust Root resource. You have three options for creating the Trust Root resource: a custom TUF repository, using your own keys, or using a serialized TUF root.

   1. Configure these environment variables from the RHTAS services:

      $ export TUF_URL="$(oc -n trusted-artifact-signer get tuf -o jsonpath='{.items[0].status.url}')"
      $ export BASE64_TUF_ROOT="$(curl -fsSL "$TUF_URL/root.json" | base64 -w0)"
      $ export FULCIO_URL="$(oc -n trusted-artifact-signer get fulcio -o jsonpath='{.items[0].status.url}')"
      $ export CTLOG_URL="http://ctlog.trusted-artifact-signer.svc.cluster.local"
      $ export REKOR_URL="$(oc -n trusted-artifact-signer get rekor -o jsonpath='{.items[0].status.url}')"
      $ export TSA_URL="$(oc -n trusted-artifact-signer get timestampAuthorities -o jsonpath='{.items[0].status.url}')"

   2. Option 1. Create the TrustRoot resource for a custom TUF repository:

      $ cat <<EOF | oc apply -f -
      apiVersion: policy.sigstore.dev/v1alpha1
      kind: TrustRoot
      metadata:
        name: trust-root
      spec:
        remote:
          mirror: $TUF_URL
          root: |
            $BASE64_TUF_ROOT
      EOF

   3. Option 2. Create a Trust Root with your own keys.

      1. Create and apply the TrustRoot resource using this template:

         apiVersion: policy.sigstore.dev/v1alpha1
         kind: TrustRoot
         metadata:
           name: trust-root
         spec:
           sigstoreKeys:
             certificateAuthorities:
             - subject:
                 organization: fulcio-organization
                 commonName: fulcio-common-name
               uri: $FULCIO_URL
               certChain: |-
                 FULCIO_CERT_CHAIN
             ctLogs:
             - baseURL: $CTLOG_URL
               hashAlgorithm: sha-256
               publicKey: |-
                 CTFE_PUBLIC_KEY
             tLogs:
             - baseURL: $REKOR_URL
               hashAlgorithm: sha-256
               publicKey: |-
                 REKOR_PUBLIC_KEY
             timestampAuthorities:
             - subject:
                 organization: tsa-organization
                 commonName: tsa-common-name
               uri: $TSA_URL
               certChain: |-
                 TSA_CERT_CHAIN

         Note

         Substitute the public keys and certificate chain values with your specific values for your RHTAS environment.

   4. Option 3. Create a Trust Root for a serialized TUF root.

      1. Create a temporary directory to contain a clone of your TUF root:

         $ mkdir -p tuf-repo

      2. Download and clone the TUF repository:

         $ curl -s $TUF_URL/root.json > root.json
         $ tuftool clone --metadata-url=$TUF_URL --metadata-dir=tuf-repo --targets-url=$TUF_URL/targets --targets-dir=tuf-repo/targets --root=root.json

      3. Archive and encode the TUF repository:

         $ tar -C ./tuf-repo -czf tuf-repo.tgz .
         $ export MIRROR_FS=$(base64 -w0 tuf-repo.tgz)

      4. Create the TrustRoot resource:

         $ cat <<EOF | oc apply -f -
         apiVersion: policy.sigstore.dev/v1alpha1
         kind: TrustRoot
         metadata:
           name: trust-root
         spec:
           repository:
             root: |-
               $BASE64_TUF_ROOT
             mirrorFS: |-
               $MIRROR_FS
         EOF

5. Create a basic Policy Controller cluster image policy resource.

   1. Configure these environment variables for Fulcio, Rekor, the Trust Root, and the OpenID Connect (OIDC) issuer and subject:

      $ export FULCIO_URL="$(oc -n trusted-artifact-signer get fulcio -o jsonpath='{.items[0].status.url}')"
      $ export REKOR_URL="$(oc -n trusted-artifact-signer get rekor -o jsonpath='{.items[0].status.url}')"
      $ export TRUST_ROOT_RESOURCE="trust-root"
      $ export OIDC_ISSUER_URL="https://ISSUER_URL"
      $ export OIDC_SUBJECT="SUBJECT"

   2. Create the ClusterImagePolicy resource:

      $ cat <<EOF | oc apply -f -
      apiVersion: policy.sigstore.dev/v1beta1
      kind: ClusterImagePolicy
      metadata:
        name: cluster-image-policy
      spec:
        images:
          - glob: "**"
        authorities:
          - keyless:
              url: $FULCIO_URL
              trustRootRef: $TRUST_ROOT_RESOURCE
              identities:
                - issuer: $OIDC_ISSUER_URL
                  subject: $OIDC_SUBJECT
            ctlog:
              url: $REKOR_URL
              trustRootRef: $TRUST_ROOT_RESOURCE
            rfc3161timestamp:
              trustRootRef: $TRUST_ROOT_RESOURCE
      EOF

      Note

      The glob value of ** evaluates all container images.

Procedure

1. Open a terminal on your workstation, log in to OpenShift from the command line:

   oc login --token=TOKEN --server=SERVER_URL_AND_PORT

   $ oc login --token=sha256~ZvFDBvoIYAbVECixS4-WmkN4RfnNd8Neh3y1WuiFPXC --server=https://example.com:6443

   Note

   You can find your login token and URL for use on the command line from the OpenShift web console. Log in to the OpenShift web console. Click your user name, and click Copy login command. Offer your user name and password again, if asked, and click Display Token to view the command.

2. Configure your shell environment:

   $ export WORK="${HOME}/trustroot-example"
   $ export SIGNED_TRUST_ROOT="${WORK}/root/trusted_root.json"
   $ export TUF_REPO="${WORK}/tuf-repo"
   $ export TUF_SERVER_POD="$(oc get pods -l app.kubernetes.io/component=tuf,\!job-name -o jsonpath='{.items[0].metadata.name}' -n trusted-artifact-signer)"
   $ export CA_URL=$(oc get fulcio -o jsonpath='{.items[0].status.url}' -n trusted-artifact-signer)
   $ export TLOG_URL=$(oc get rekor -o jsonpath='{.items[0].status.url}' -n trusted-artifact-signer)
   $ export OIDC_URL="OIDC_ISSUER_URL"

   Replace OIDC_ISSUER_URL with your OIDC provider’s URL address.

3. Create the temporary TUF directories:

   $ mkdir -p "${WORK}/root/" "${TUF_REPO}"

4. Download the signed target trust root file to the temporary TUF directories:

   $ oc cp "${TUF_SERVER_POD}:/var/www/html" "${TUF_REPO}" -n trusted-artifact-signer
   $ cp "${TUF_REPO}/targets/DIGEST.trusted_root.json" "${SIGNED_TRUST_ROOT}"

   An example signed target trust root file name looks similar to this format, c03afd04e353889093e5b16b019656b23a57.trusted_root.json, where your DIGEST value would be different.

5. Create a script for making the client trust configuration used by the CLI:

   $ cat > make_trust_config.sh <<'EOF'
   #!/bin/bash
   
   # Usage: ./make-trust-config.sh <trusted_root_input.json> <output.json> [caUrl] [oidcUrl] [tlogUrl]
   
   if [ "$#" -lt 2 ]; then
       echo "Usage: $0 <trusted_root_input.json> <output.json> [caUrl] [oidcUrl] [tlogUrl]"
       exit 1
   fi
   
   INPUT_FILE="$1"
   OUTPUT_FILE="$2"
   CA_URL=${3:-${CA_URL:-""}}
   OIDC_URL=${4:-${OIDC_URL:-""}}
   TLOG_URL=${5:-${TLOG_URL:-""}}
   
   # Check for jq
   if ! command -v jq &> /dev/null; then
       echo "Error: 'jq' is required but not installed."
       exit 1
   fi
   
   jq -n \
     --argjson trustedRoot "$(cat $INPUT_FILE)" \
     --arg caUrl "$CA_URL" \
     --arg oidcUrl "$OIDC_URL" \
     --arg tlogUrl "$TLOG_URL" \
     '{
       mediaType: "application/vnd.dev.sigstore.clienttrustconfig.v0.1+json",
       trustedRoot: $trustedRoot,
       signingConfig: {
         caUrl: $caUrl,
         oidcUrl: $oidcUrl,
         tlogUrls: [$tlogUrl]
       }
     }' > "$OUTPUT_FILE"
   EOF

6. Make the script executable:

   $ chmod u+x make_trust_config.sh

7. Run the make_trust_config.sh script:

   $ ./make_trust_config.sh $SIGNED_TRUST_ROOT trust_config.json

   A new trust_config.json file is created in the current working directory.

8. You can now start signing and verifying AL/ML models by using command-line interface.

Procedure

1. Log in to the OpenShift web console with a user that has the cluster-admin role.
2. From the Administrator perspective, expand the Operators navigation menu, and click OperatorHub.
3. In the search field, type Model Validation Operator, and click the tile that is displayed.
4. Click the Install button to show the operator details.
5. Accept the default values, click Install on the Install Operator page, and wait for the installation to finish.
6. Once the installation finishes, click View Operator.
7. Add the AI/ML model, its signature, and your signed Trust Root configuration to the namespace you want validation done. This is typically done by creating a Persistent Volume Claim (PVC) on the OpenShift cluster, and copying these files to the PVC.

8. Create a ModelValidation CR in the namespace you want validation done.

   1. Click the modelvalidation tab, and click the Create modelvalidation button.

   2. On the Create modelvalidation page, select YAML view. Update the YAML file accordingly:

      apiVersion: ml.sigstore.dev/v1alpha1
      kind: ModelValidation
      metadata:
        name: model-validation-example
        namespace: NAMESPACE
      spec:
        config:
          sigstoreConfig:
            certificateIdentity: "IDENTITY"
            certificateOidcIssuer: "OIDC_ISSUER"
          clientTrustConfig:
            trustConfigPath: SIGNED_TRUST_ROOT
        model:
          path: PATH_TO_MODEL
          signaturePath: PATH_TO_MODEL_SIGNATURE
        continuousValidation:
          enabled: true
          interval: "10m"

      Replace NAMESPACE with the same namespace where your workloads run.

      Replace IDENTITY with the signer’s email address.

      Replace OIDC_ISSUER with your OIDC provider’s issuer URL address.

      Replace SIGNED_TRUST_ROOT with the signed Trust Root target file, for example, /data/trust-config.json.

      Replace PATH_TO_MODEL with the path to the model file, for example, /data/model.onnx.

      Replace PATH_TO_MODEL_SIGNATURE with the path to the model’s signature file, for example, /data/model.sig.

      The interval field supports a duration format in seconds (s), minutes (m), or hours (h). In this example, the validation check runs every 10 minutes.

   3. Click the Create button.

9. From your terminal session, create a new pod CR in the namespace you want to trigger a validation check. Update this example YAML file with your specific information:

   apiVersion: v1
   kind: Pod
   metadata:
     name: model-validation-pod-example
     namespace: NAMESPACE
     labels:
       validation.ml.sigstore.dev/ml: "model-validation-example"
   spec:
     containers:
     - name: app
       image: nginx
       ports:
       - containerPort: 80
       volumeMounts:
       - name: model-storage-example
         mountPath: PATH_TO_WORKLOAD_VOLUME
     volumes:
     - name: model-storage-example
       persistentVolumeClaim:
         claimName: PVC_NAME

   You configure the webhook by using this label key, validation.ml.sigstore.dev/ml, with the value of the ModelValidation CR name created earlier, surrounded by double quotes.

   Replace NAMESPACE, PATH_TO_WORKLOAD_VOLUME, and PVC_NAME with values appropriate to your environment.

   1. Create the new pod by applying the CR file:

      oc apply -f _PATH_TO_CR_FILE_

10. Now the webhook can intercept pod create and update requests. Next, the Model Validation Operator injects validation steps that reads the AI/ML model and its signature, and checks it against your Trust Root for RHTAS. If the validation check succeeds, then the pod creation or modification proceeds on.

Procedure

1. Configure your shell environment:

   $ export OIDC_ISSUER="OIDC_ISSUER_URL"
   $ export MODEL_IMAGE="registry.redhat.io/rhtas/model-transparency-rhel9@sha256:6db7fa2b956875a6f507811166b47b164d463dea78ab4403c6d7648d838b8acb"
   $ export MODEL_DIR="PATH_TO_MODEL_DIRECTORY"
   $ export TRUST_CFG="$(pwd)/trust_config.json"
   $ export SIG_PATH="$MODEL_DIR/model.sig"

   Replace OIDC_ISSUER_URL with your OIDC provider URL address.

   Replace PATH_TO_MODEL_DIRECTORY with the absolute path to the directory containing the AI/ML models.

2. There are two options for signing a model by using the CLI. You can use an identity token and a client identifier, or just the client identifier. Using an identity token is the non-interactive way, where as using a client identifier is the interactive way.

   Note

   When using self-signed certificates or a custom Certificate Authority (CA), you have to pass those certificates to the container to successfully sign an AI/ML model.

   1. Option 1. Signing AI/ML models with an identity token:

      $ podman run --rm \
      --userns=keep-id --user "$(id -u)":"$(id -g)" --group-add keep-groups \
      -v "$MODEL_DIR":/model:Z,U \
      -v "$TRUST_CFG":/trust_config.json:Z,ro \
      -w /model "$MODEL_IMAGE" \
      sign sigstore \
      --trust_config /trust_config.json \
      --signature /model/model.sig \
      --identity_token "OIDC_TOKEN" \
      --client_id CLIENT_ID \
      /model

      Replace OIDC_TOKEN with your OIDC authentication token.

      Replace CLIENT_ID with your OIDC client identifier.

   2. Option 2. Signing AI/ML models by using a client identifier:

      $ podman run --rm -it \
      --userns=keep-id --user "$(id -u)":"$(id -g)" --group-add keep-groups \
      -v "$MODEL_DIR":/model:Z,U \
      -v "$TRUST_CFG":/trust_config.json:Z,ro \
      -w /model "$MODEL_IMAGE" \
      sign sigstore \
      --trust_config "/trust_config.json" \
      --signature "/model/model.sig" \
      --client_id CLIENT_ID \
      /model

      Replace CLIENT_ID with your OIDC client identifier.

3. Verify a model signature that uses the CLI:

   $ podman run --rm -it \
   --userns=keep-id --user "$(id -u)":"$(id -g)" --group-add keep-groups \
   -v "$MODEL_DIR":/model:Z,U \
   -v "$TRUST_CFG":/trust_config.json:Z,ro \
   -w /model "$MODEL_IMAGE" \
   verify sigstore \
   --trust_config "/trust_config.json" \
   --signature "/model/model.sig" \
   --identity IDENTITY \
   --identity_provider "$OIDC_ISSUER" \
   /model

   Replace IDENTITY with an email address or with a SPIFFE or URI subject.

Procedure

1. Open a terminal, and install the model transparency CLI Python package to your workstation:

   $ pip install rh-model-signing

2. Download a copy of the trust root from your OpenShift cluster to your workstation:

   $ export TUF_URL=$(oc get tuf -o jsonpath='{.items[0].status.url}' -n trusted-artifact-signer)
   $ curl -o root.json $TUF_URL/root.json

3. Bootstrap the trust root before signing or verifying any models:

   $ rh_model_signing trust-instance $TUF_URL root.json

4. Different ways to sign your AI/ML models:

   Sign a local ML model directory

   $ rh_model_signing sign sigstore PATH_TO_MODEL --instance $TUF_URL --client-id trusted-artifact-signer

   Replace PATH_TO_MODEL with the path to your local ML model directory.

   Sign an OCI model artifact or image with a registry attachment

   $ rh_model_signing sign sigstore PATH_TO_REGISTRY:TAG --instance $TUF_URL --client-id trusted-artifact-signer

   Replace PATH_TO_REGISTRY:TAG with the URL to your OCI image and the tag for the image.

   Sign an OCI model artifact or image but outputting the signature as a separate file

   $ rh_model_signing sign sigstore PATH_TO_REGISTRY:TAG --output-mode file --signature PATH_TO_MODEL_SIGNATURE --instance $TUF_URL --client-id trusted-artifact-signer

   Replace PATH_TO_REGISTRY:TAG with the URL to your OCI image and the tag for the image. Replace PATH_TO_MODEL_SIGNATURE with the path to the file containing the signature for the ML model.

   Sign and attach by using the tag-based method

   $ rh_model_signing sign sigstore PATH_TO_REGISTRY:TAG --attachment-mode tag --instance $TUF_URL --client-id trusted-artifact-signer

   Replace PATH_TO_REGISTRY:TAG with the URL to your OCI image and the tag for the image.

5. Different ways to verify your AI/ML models:

   Verify a local AI/ML model directory

   $ rh_model_signing verify sigstore PATH_TO_MODEL --signature model.sig --identity "jdoe@example.com" --identity-provider "OIDC_ISSUER_URL" --instance $TUF_URL

   Replace PATH_TO_MODEL with the path to your local ML model directory. Replace OIDC_ISSUER_URL with the URL to your OpenID Connect (OIDC) issuer.

   Verify an OCI image from a registry

   $ rh_model_signing verify sigstore PATH_TO_REGISTRY:TAG --identity "jdoe@example.com" --identity-provider “OIDC_ISSUER_URL” --instance $TUF_URL

   Replace PATH_TO_REGISTRY:TAG with the URL to your OCI image and the tag for the image. Replace OIDC_ISSUER_URL with the URL to your OIDC issuer.

   Verify an OCI image with a local signature file

   $ rh_model_signing verify sigstore PATH_TO_REGISTRY:TAG --signature PATH_TO_MODEL_SIGNATURE --identity "jdoe@example.com" --identity-provider "OIDC_ISSUER_URL" --instance $TUF_URL

   Replace PATH_TO_REGISTRY:TAG with the URL to your OCI image and the tag for the image. Replace PATH_TO_MODEL_SIGNATURE with the path to the file containing the signature for the ML model. Replace OIDC_ISSUER_URL with the URL to your OIDC issuer.

   Verify and attach by using the tag-based method

   $ rh_model_signing verify sigstore PATH_TO_REGISTRY:TAG --attachment-mode tag --identity "jdoe@example.com" --identity-provider "OIDC_ISSUER_URL" --instance $TUF_URL

   Replace PATH_TO_REGISTRY:TAG with the URL to your OCI image and the tag for the image. Replace OIDC_ISSUER_URL with the URL to your OIDC issuer.

Procedure

1. Open the Amazon RDS console, and create a new MariaDB instance.

   1. Wait for the MariaDB instance to be deployed, and is available.

2. From your workstation, log in to the new database by providing the regional URL address, the port, and the user name:

   mysql -h FQDN_FOR_RDS_HOSTNAME -P 3306 -u USER_NAME -p

   $ mysql -h exampledb.1234.us-east-1.rds.amazonaws.com -P 3306 -u admin -p

3. Create a new database named trillian:

   create database trillian;

4. Switch to the newly created database:

   use trillian;

5. Create a new database user named trillian, and set a PASSWORD for the newly created user:

   CREATE USER trillian@'%' IDENTIFIED BY 'PASSWORD';
   GRANT ALL PRIVILEGES ON trillian.* TO 'trillian'@'%';
   FLUSH PRIVILEGES;

6. Disconnect from the database:

   EXIT

7. Download the database configuration file:

   $ curl -o dbconfig.sql https://raw.githubusercontent.com/securesign/trillian/main/storage/mysql/schema/storage.sql

8. Apply the database configuration to the new database:

   mysql -h FQDN_FOR_RDS_HOSTNAME -P 3306 -u USER_NAME -p -D DB_NAME < PATH_TO_CONFIG_FILE

   $ mysql -h exampledb.1234.us-east-1.rds.amazonaws.com -P 3306 -u trillian -p -D trillian < dbconfig.sql

9. Open a terminal on your workstation, and log in to OpenShift:

   oc login --token=TOKEN --server=SERVER_URL_AND_PORT

   $ oc login --token=sha256~ZvFDBvoIYAbVECixS4-WmkN4RfnNd8Neh3y1WuiFPXC --server=https://example.com:6443

   Note

   You can find your login token and URL for use on the command line from the OpenShift web console. Log in to the OpenShift web console. Click your user name, and click Copy login command. Offer your user name and password again, if asked, and click Display Token to view the command.

10. Create a new Secret containing the credentials for the Trillian database within the MariaDB instance which was created previously:

    oc create secret generic OBJECT_NAME \
    --from-literal=mysql-database=trillian \
    --from-literal=mysql-host=FQDN_FOR_RDS_HOSTNAME \
    --from-literal=mysql-password=PASSWORD \
    --from-literal=mysql-port=PORT_NUMBER \
    --from-literal=mysql-root-password=PASSWORD \
    --from-literal=mysql-user=USER_NAME

    $ oc create secret generic trillian-db \
    --from-literal=mysql-database=trillian \
    --from-literal=mysql-host=exampledb.1234.us-east-1.rds.amazonaws.com \
    --from-literal=mysql-password=mypassword123 \
    --from-literal=mysql-port=3306 \
    --from-literal=mysql-root-password=myrootpassword123 \
    --from-literal=mysql-user=trillian

    Important

    Verify that your security group rules allow access to the database from the OpenShift cluster.

11. You can now deploy the Trusted Artifact Signer service to use this database. If you were following the Trusted Artifact Signer installation procedure, then you can proceed to the next step.

Procedure

1. Log in to the OpenShift web console where you are deploying the RHTAS service.

2. Select the trusted-artifact-signer project, if the project already exists, else create a new project for the database:

   1. To create a new project, click the drop-down project menu, and click the Create Project button.
   2. Name the new project trusted-artifact-signer, and click the Create button.
3. From the navigation menu, expand Ecosystem, click Software Catalog, then click Databases from the software catalog menu.

4. Select MariaDB, and click the Instantiate Template button.

   Important

   Do not select MariaDB (Ephemeral).

5. On the Instantiate Template page, configure the following fields:

   1. In the MariaDB Database Name field, enter trillian.
   2. In the Volume Capacity field, enter 5Gi.
   3. Click the Create button.

6. Begin a remote shell session:

   1. On the Topology page, selecting the MariaDB pod brings up a side panel, click the Resources tab.
   2. Under the Pods section, click on the MariaDB pod name.
   3. Click the Terminal tab to start a remote shell session to the MariaDB pod.

7. In the remote shell session, verify that you can connect to the Trillian database:

   $ mysql -u $MYSQL_USER -p$MYSQL_PASSWORD -D$MYSQL_DATABASE

   Note

   Credentials are stored in a secret object with the service name (mariadb), and contains the name of the database, and user name, along with the database root password. Make a note of these credentials as they will be used later on when creating the database secret object.

8. Disconnect from the database:

   EXIT

9. Download the database configuration file:

   $ curl -o dbconfig.sql https://raw.githubusercontent.com/securesign/trillian/main/storage/mysql/schema/storage.sql

10. Apply the database configuration to the new database:

    $ mysql -u $MYSQL_USER -p$MYSQL_PASSWORD -D$MYSQL_DATABASE < dbconfig.sql

11. Open a terminal on your workstation, and log in to OpenShift:

    oc login --token=TOKEN --server=SERVER_URL_AND_PORT

    $ oc login --token=sha256~ZvFDBvoIYAbVECixS4-WmkN4RfnNd8Neh3y1WuiFPXC --server=https://example.com:6443

    Note

    You can find your login token and URL for use on the command line from the OpenShift web console. Log in to the OpenShift web console. Click your user name, and click Copy login command. Offer your user name and password again, if asked, and click Display Token to view the command.

12. Create a new Secret containing the credentials for the Trillian database within the MariaDB instance which was created previously:

    oc create secret generic OBJECT_NAME \
    --from-literal=mysql-database=trillian \
    --from-literal=mysql-host=FQDN_or_SERVICE_ADDR \
    --from-literal=mysql-password=PASSWORD \
    --from-literal=mysql-port=PORT_NUMBER \
    --from-literal=mysql-root-password=PASSWORD \
    --from-literal=mysql-user=USER_NAME

    $ oc create secret generic trillian-db \
    --from-literal=mysql-database=trillian \
    --from-literal=mysql-host=mariadb.trusted-artifact-signer.svc.cluster.local \
    --from-literal=mysql-password=mypassword123 \
    --from-literal=mysql-port=3306 \
    --from-literal=mysql-root-password=myrootpassword123 \
    --from-literal=mysql-user=trillian

    You can use an OpenShift internal service name for the MariaDB instance.

13. You can now deploy the Trusted Artifact Signer service to use this database. If you were following the Trusted Artifact Signer installation procedure, then you can proceed to the next step.

Procedure

1. From your workstation, log in to PostgreSQL by providing the hostname, the port, and the user name:

   psql -h HOSTNAME -p 5432 -U USER_NAME

   $ psql -h db.example.com -p 5432 -U postgres

2. Create a new database named trillian:

   CREATE DATABASE trillian;

3. Create a new database user named trillian, and set a PASSWORD for the newly created user:

   CREATE USER trillian WITH PASSWORD 'PASSWORD';
   ALTER DATABASE trillian OWNER TO trillian;
   GRANT ALL ON SCHEMA public TO trillian;

4. Disconnect from the database:

   exit

5. Download the database configuration file:

   $ curl -o dbconfig.sql https://raw.githubusercontent.com/securesign/trillian/main/storage/postgresql/schema/storage.sql

6. Apply the database configuration to the new database:

   psql -h FQDN_or_SERVICE_ADDR -p 5432 -U USER_NAME -d DB_NAME -f PATH_TO_CONFIG_FILE

   $ psql -h db.example.com -p 5432 -U trillian -d trillian -f dbconfig.sql

7. Open a terminal on your workstation, and log in to OpenShift:

   oc login --token=TOKEN --server=SERVER_URL_AND_PORT

   $ oc login --token=sha256~ZvFDBvoIYAbVECixS4-WmkN4RfnNd8Neh3y1WuiFPXC --server=https://example.com:6443

   Note

   You can find your login token and URL for use on the command line from the OpenShift web console. Log in to the OpenShift web console. Click your user name, and click Copy login command. Offer your user name and password again, if asked, and click Display Token to view the command.

8. Create a new Secret containing the credentials for the Trillian database within the PostgreSQL database instance which was created previously:

   oc create secret generic trillian-db-credentials \
     --from-literal=postgresql-host=FQDN \
     --from-literal=postgresql-port=5432 \
     --from-literal=postgresql-user=trillian \
     --from-literal=postgresql-password=PASSWORD \
     --from-literal=postgresql-database=trillian \
     -n trusted-artifact-signer

   $ oc create secret generic trillian-db-credentials \
     --from-literal=postgresql-host=db.example.com \
     --from-literal=postgresql-port=5432 \
     --from-literal=postgresql-user=trillian \
     --from-literal=postgresql-password=mypassword123 \
     --from-literal=postgresql-database=trillian \
     -n trusted-artifact-signer

9. Before deploying the Trusted Artifact Signer service, update the Trillian resource accordingly:

   apiVersion: rhtas.redhat.com/v1alpha1
   kind: Trillian
   metadata:
     name: trillian
     namespace: trusted-artifact-signer
   spec:
     database:
       create: false
       provider: postgresql
       uri: "postgresql://$(DB_USER):$(DB_PASSWORD)@$(DB_HOST):$(DB_PORT)/$(DB_NAME)"
     auth:
       env:
         - name: DB_HOST
           valueFrom:
             secretKeyRef:
               name: trillian-db-credentials
               key: postgresql-host
         - name: DB_PORT
           valueFrom:
             secretKeyRef:
               name: trillian-db-credentials
               key: postgresql-port
         - name: DB_USER
           valueFrom:
             secretKeyRef:
               name: trillian-db-credentials
               key: postgresql-user
         - name: DB_PASSWORD
           valueFrom:
             secretKeyRef:
               name: trillian-db-credentials
               key: postgresql-password
         - name: DB_NAME
           valueFrom:
             secretKeyRef:
               name: trillian-db-credentials
               key: postgresql-database

   If you were following the Trusted Artifact Signer installation procedure, then you can proceed to the next step.

Procedure

1. Open a terminal on your workstation, and log in to OpenShift:

   oc login --token=TOKEN --server=SERVER_URL_AND_PORT

   $ oc login --token=sha256~ZvFDBvoIYAbVECixS4-WmkN4RfnNd8Neh3y1WuiFPXC --server=https://example.com:6443

   Note

   You can find your login token and URL for use on the command line from the OpenShift web console. Log in to the OpenShift web console. Click your user name, and click Copy login command. Offer your user name and password again, if asked, and click Display Token to view the command.

2. Create a secret for your Amazon S3 credentials:

   $ oc create secret generic rekor-s3-credentials \
     --from-literal=AWS_ACCESS_KEY_ID="YOUR_ACCESS_KEY_ID" \
     --from-literal=AWS_SECRET_ACCESS_KEY="YOUR_SECRET_ACCESS_KEY" \
     -n trusted-artifact-signer

   Replace YOUR_ACCESS_KEY_ID and YOUR_SECRET_ACCESS_KEY with your actual AWS access key ID and secret access key.

3. Update the Rekor resource to enable attestation storage by setting spec.attestations.enabled to true, configure the attestation storage url, and set the appropriate environment variables for AWS authentication:

   apiVersion: rhtas.redhat.com/v1alpha1
   kind: Rekor
   metadata:
     name: rekor
     namespace: trusted-artifact-signer
   spec:
     replicas: 3
     attestations:
       enabled: true
       url: "s3://ATTESTATION_BUCKET_NAME?region=AWS_REGION_VALUE"
       maxSize: "100Ki"
     auth:
       env:
         - name: AWS_ACCESS_KEY_ID
           valueFrom:
             secretKeyRef:
               name: rekor-s3-credentials
               key: AWS_ACCESS_KEY_ID
         - name: AWS_SECRET_ACCESS_KEY
           valueFrom:
             secretKeyRef:
               name: rekor-s3-credentials
               key: AWS_SECRET_ACCESS_KEY

   Replace ATTESTATION_BUCKET_NAME and AWS_REGION_VALUE with your actual S3 bucket name and AWS region.

Procedure

1. Create a secret to store your Redis credentials:

   $ oc create secret generic redis-credentials --from-literal=password=PASSWORD -n trusted-artifact-signer

   Replace PASSWORD with the password for your Redis user.

2. Create a Rekor resource to reference the external Redis instance:

   $ cat <<EOF | oc apply -f -
   apiVersion: rhtas.redhat.com/v1alpha1
   kind: Rekor
   metadata:
     name: rekor
     namespace: trusted-artifact-signer
   spec:
     searchIndex:
       create: false
       provider: redis
       url: "redis://:$(REDIS_PASSWORD)@redis.example.com:6379"
     auth:
       env:
         - name: REDIS_PASSWORD
           valueFrom:
             secretKeyRef:
               name: redis-credentials
               key: password
   EOF

   Update the url field to match the connection details of your Redis instance, including the hostname and port. The URL format for Redis is redis://[username:password@]host:port[/database].

   1. Optional. If you are using Transport Layer Security (TLS) with Redis, then you must start the URL string with rediss://, and you must reference your cloud provider’s certificate authority (CA) certificate under the spec section of the Rekor resource:

      apiVersion: rhtas.redhat.com/v1alpha1
      kind: Rekor
      metadata:
        name: rekor
        namespace: trusted-artifact-signer
      spec:
        searchIndex:
          create: false
          provider: redis
          url: "rediss://:$(REDIS_PASSWORD)@redis.example.com:6379"
        auth:
          env:
            - name: REDIS_PASSWORD
              valueFrom:
                secretKeyRef:
                  name: redis-credentials
                  key: password
        trustedCA:
          name: redis-ca-bundle  # ConfigMap containing the CA certificate.

3. Set the Rekor backfill Cron job under the spec section:

   apiVersion: rhtas.redhat.com/v1alpha1
   kind: Rekor
   metadata:
     name: rekor
     namespace: trusted-artifact-signer
   spec:
     backFillRedis:
       enabled: true
       schedule: "0 0 * * *"  # Runs daily at midnight.

   The Rekor backfill Cron job ensures the search index stays synchronized with the transparency log.