-
Bug
-
Resolution: Done
-
Undefined
-
Logging 5.4.0
-
False
-
False
-
NEW
-
OBSDA-108 - Distribute an alternate Vector Log Collector
-
VERIFIED
-
-
Logging (Core) - Sprint 215
Logging 5.5 preview
Description of problem:
When a ClusterLogForwarder instance is created to forward logs from same log type and forwarding from different namespaces to multiple log stores, the collector pods fail with Vector configuration error.
Feb 21 09:26:18.244 INFO vector::app: Log level is enabled. level="info" Feb 21 09:26:18.244 INFO vector::sources::host_metrics: PROCFS_ROOT is unset. Using default '/proc' for procfs root. Feb 21 09:26:18.245 INFO vector::sources::host_metrics: SYSFS_ROOT is unset. Using default '/sys' for sysfs root. Feb 21 09:26:18.245 INFO vector::app: Loading configs. path=[("/etc/vector/vector.toml", Some(Toml))] Feb 21 09:26:18.248 ERROR vector::cli: Configuration error. error=redefinition of table `transforms.route_application_logs` for key `transforms.route_application_logs` at line 54 column 1
How reproducible:
Always
Steps to reproduce the issue:
1 Install the Cluster Logging and Elasticsearch 5.5 preview operator.
2 Create a Cluster Logging instance with Vector as collector.
3 Create a ClusterLogForwarder instance to forward logs from same log type and from multiple projects to different log stores.
apiVersion: logging.openshift.io/v1 kind: ClusterLogForwarder metadata: name: instance namespace: openshift-logging spec: outputs: - name: loki-server type: loki url: http://loki-server.loki-aosqe.svc:3100 - name: kafka-server type: kafka url: tls://my-cluster-kafka-bootstrap.amq-aosqe.svc:9092/topic-logging-app inputs: - name: input-kafka application: namespaces: - project-kafka - name: input-loki application: namespaces: - project-loki - name: input-es application: namespaces: - project-es pipelines: - name: to-kafka inputRefs: - input-kafka outputRefs: - kafka-server - name: to-loki inputRefs: - input-loki outputRefs: - loki-server - name: to-es inputRefs: - input-es outputRefs: - default
4 Check the status and logs of the collector pods.
NAME READY STATUS RESTARTS AGE collector-6rfws 1/2 CrashLoopBackOff 8 (2m28s ago) 18m oc logs collector-vplt2 -c collector Feb 21 09:36:38.297 INFO vector::app: Log level is enabled. level="info" Feb 21 09:36:38.298 INFO vector::sources::host_metrics: PROCFS_ROOT is unset. Using default '/proc' for procfs root. Feb 21 09:36:38.298 INFO vector::sources::host_metrics: SYSFS_ROOT is unset. Using default '/sys' for sysfs root. Feb 21 09:36:38.298 INFO vector::app: Loading configs. path=[("/etc/vector/vector.toml", Some(Toml))] Feb 21 09:36:38.302 ERROR vector::cli: Configuration error. error=redefinition of table `transforms.route_application_logs` for key `transforms.route_application_logs` at line 54 column 1
5 Extract and check the Vector config.
oc extract secret/collector-config --confirm From the generated Vector config: [transforms.route_application_logs] type = "route" inputs = ["application"] route.input-kafka = '(.kubernetes.pod_namespace == "project-kafka")' [transforms.route_application_logs] type = "route" inputs = ["application"] route.input-loki = '(.kubernetes.pod_namespace == "project-loki")' [transforms.route_application_logs] type = "route" inputs = ["application"] route.input-es = '(.kubernetes.pod_namespace == "project-es")'
- links to
- mentioned on
(3 mentioned on)