Description of problem: When the `chunkLimitSize` for fluentd is configured according to the documentation. The wrong chunk_limit_size is shown in the logs Version-Release number of selected component (if applicable): 4.6 How reproducible: Steps to Reproduce: 1. Deploy cluster-logging 2. Set chunk_size_limit for fluentd ~~~~ # oc edit ClusterLogging instance spec: forwarder: fluentd: buffer: chunkLimitSize: 750k ~~~~ 3. Verify that it is applied ~~~~ # oc extract configmap/fluentd --confirm # grep chunk fluent.conf ~~~~ 4. View logs ~~~~ # oc get pods | grep fluent ✭ NAME READY STATUS RESTARTS AGE fluentd-4t5n7 1/1 Running 0 6m18s fluentd-dkjhr 1/1 Running 0 8m11s fluentd-fzcqb 1/1 Running 0 7m26s fluentd-l5tr5 1/1 Running 0 6m58s fluentd-pf2td 1/1 Running 0 7m50s fluentd-tvjks 1/1 Running 0 6m42s kibana-5db88c8cf-6w4t5 2/2 Running 0 2d15h # oc logs fluentd-4t5n7 | head -3 ✭ Setting each total_size_limit for 2 buffers to 9621286195 bytes Setting queued_chunks_limit_size for each buffer to 1146 Setting chunk_limit_size for each buffer to 8388608 ~~~~ Actual results: Setting each total_size_limit for 2 buffers to 9621286195 bytes Setting queued_chunks_limit_size for each buffer to 1146 Setting chunk_limit_size for each buffer to 8388608 Expected results: Setting each total_size_limit for 2 buffers to 9621286195 bytes Setting queued_chunks_limit_size for each buffer to 1146 Setting chunk_limit_size for each buffer to 768000 Additional info: * `run.sh` is still calculating it depending on the ${BUFFER_SIZE_LIMIT}` variable, which is no longer applied. BUFFER_SIZE_LIMIT=$(echo ${BUFFER_SIZE_LIMIT:-8m} | sed -e "s/[Kk]/*1024/g;s/[Mm]/*1024*1024/g;s/[Gg]/*1024*1024*1024/g;s/i//g" | bc) BUFFER_QUEUE_LIMIT=$(expr $TOTAL_LIMIT_SIZE / $BUFFER_SIZE_LIMIT) echo "Setting queued_chunks_limit_size for each buffer to $BUFFER_QUEUE_LIMIT" export BUFFER_QUEUE_LIMIT echo "Setting chunk_limit_size for each buffer to $BUFFER_SIZE_LIMIT" export BUFFER_SIZE_LIMIT
Verified on cluster-logging.5.0.0-25
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory (Errata Advisory for Openshift Logging 5.0.0), and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2021:0652