Details
-
Bug
-
Resolution: Done-Errata
-
Critical
-
Logging 5.8.0
-
False
-
None
-
False
-
NEW
-
NEW
-
Before this update, vector log collector pods were getting stuck in the CrashLoopBackOff state on IBM Power machines. With this fix, vector log collector pods start successfully on IBM Power architecture machines.
-
Bug Fix
-
Proposed
-
This is the collector pod logs with debug enabled on the vector collectorlog.txt
-
Log Collection - Sprint 244, Log Collection - Sprint 245
-
Critical
Description
Description of problem:
Logging 5.8.0 vector pods going into the Crashloopbackoff state on IBM Power on OCP 4.14
Version-Release number of selected component (if applicable):
Logging 5.8.0
How reproducible: Always
Steps to Reproduce:
- Deploy cluster-logging.v5.8.0 operator and elasticsearch-operator.v5.8.0
- create clusterlogging instance with logging type vector and forward logs to default elasticsearch
Actual results:
Collector pods are going into the crashloopbackoff state.
Expected results:
It should create collector pods with running state and forward the logs to default elasticsearch successfully.
Additional info:
Logging 5.7.8 with logging type vector works on the same cluster.
