Bug 1894640 - Fluent stops sending logs even though logging stack seems functional
Summary: Fluent stops sending logs even though logging stack seems functional
Keywords:
Status: CLOSED NEXTRELEASE
Alias: None
Product: OpenShift Container Platform
Classification: Red Hat
Component: Logging
Version: 4.5
Hardware: Unspecified
OS: Unspecified
unspecified
medium
Target Milestone: ---
: 4.5.z
Assignee: Jeff Cantrill
QA Contact: Anping Li
URL:
Whiteboard: logging-core
Depends On: 1894639
Blocks:
TreeView+ depends on / blocked
 
Reported: 2020-11-04 16:58 UTC by Jeff Cantrill
Modified: 2021-02-03 16:14 UTC (History)
4 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of: 1894639
Environment:
Last Closed: 2021-02-03 16:14:06 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)

Description Jeff Cantrill 2020-11-04 16:58:18 UTC
+++ This bug was initially created as a clone of Bug #1894639 +++

+++ This bug was initially created as a clone of Bug #1894634 +++

Description of problem:
Fluent seems functional but is not shipping logs and it has a log entry of:

failed to write data into buffer by buffer overflow action=:block

Once the pods are restarted logs start flowing again

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


How reproducible:


Steps to Reproduce:
1.
2.
3.

Actual results:


Expected results:


Additional info:

failed to write data into buffer by buffer overflow action=:block

--- Additional comment from Jeff Cantrill on 2020-11-04 16:55:22 UTC ---

This is modifiable via changes to CL api https://issues.redhat.com/browse/LOG-742

--- Additional comment from Jeff Cantrill on 2020-11-04 16:57:29 UTC ---

This is modifiable via changes to CL api https://issues.redhat.com/browse/LOG-742

Comment 1 Jeff Cantrill 2021-01-28 18:13:42 UTC
closing next release as this was determined jit would not be pulled into 4.5

Comment 2 Jeff Cantrill 2021-02-03 16:14:06 UTC
Fixed in the next release.  This can be worked around by either:

* deleting the fluent pods (oc -n openshift-logging delete pods -l component=fluentd)
* Setting clusterlogging instance to "Unmanaged" and manually editing the fluent.conf in the fluentd configmap


Note You need to log in before you can comment on or make changes to this bug.