Loading...

Type: Bug
Resolution: Done-Errata
Priority: Normal
Fix Version/s: Logging 5.8.1
Affects Version/s: None
Component/s: Log Collection
Labels:
- devel_ack+

Blocked:
False
Blocked Reason:
None
Ready:
False
Docs QE Status:
NEW
QE Status:
NEW
Release Note Text:

Hide
Before this change, the ClusterLoggingOperator would undergo numerous reconciliations whenever there was a change in the ClusterLogForwarder. This fix addresses the reconciliations by disregarding status changes in the collector daemonsets that would trigger a reconciliation.

Show
Before this change, the ClusterLoggingOperator would undergo numerous reconciliations whenever there was a change in the ClusterLogForwarder. This fix addresses the reconciliations by disregarding status changes in the collector daemonsets that would trigger a reconciliation.
Release Note Type:
Bug Fix
Intelligence Requested:
Market:

Sprint:
Log Collection - Sprint 244, Log Collection - Sprint 245

SFDC Cases Links:
SFDC Cases Counter:
SFDC Cases Open:

Description of problem:

Whenever a reconcile is needed for ClusterLogForwarder, the forwarder controller reconciles many times (dozens of times) until it settles. The CR status does not get changed in anything except for Status: Conditions: Last Transition Time, which change with every reconcile loop.

Version-Release number of selected component (if applicable):

5.8

How reproducible:

Always

Steps to Reproduce:

Create a ClusterLogging
Create a ClusterLogForwarder
Watch cluster-logging-operator logs

Find attached the log I see, and the CL and CLF I am using. When the ClusterLogging is applied, it runs one or two times the reconcile loop. When the ClusterLogForwarder is created, the reconciler starts running like crazy many times. I am testing with a CLF with HTTP Receiver enabled, but other CLFs I tried had the same behaviour. See also the cluster_log_forwarder_app.yaml.