Bug 1867467 - multicluster-operators-standalone-subscription pod OOMkilled due to memory usage
Summary: multicluster-operators-standalone-subscription pod OOMkilled due to memory usage
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: Red Hat Advanced Cluster Management for Kubernetes
Classification: Red Hat
Component: Core Services / Observability
Version: rhacm-2.0
Hardware: x86_64
OS: Linux
unspecified
medium
Target Milestone: ---
: rhacm-2.2
Assignee: Chunlin Yang
QA Contact:
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2020-08-10 06:09 UTC by Chuan
Modified: 2021-04-09 16:36 UTC (History)
1 user (show)

Fixed In Version: rhacm-2.2
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2021-03-04 13:51:12 UTC
Target Upstream Version:
Embargoed:
gghezzo: rhacm-2.2+


Attachments (Terms of Use)
multicluster-operators-standalone-subscription OOM log file (18.50 KB, text/plain)
2020-08-10 06:09 UTC, Chuan
no flags Details


Links
System ID Private Priority Status Summary Last Updated
Github open-cluster-management backlog issues 4322 0 None None None 2020-09-08 02:36:02 UTC

Description Chuan 2020-08-10 06:09:47 UTC
Created attachment 1710919 [details]
multicluster-operators-standalone-subscription OOM log file

Description of problem:
in RHACM and CP4MCM integration env, the multicluster-operators-standalone-subscription pod restarting regularly due to OOMkilled for memory usage. 

increasing the memory limit to 2Gi solved the problem, monitoring shows its memory usage is around 650Mi, original memory limit is 512Mi.

Version-Release number of selected component (if applicable):
Build: 2.0.0-SNAPSHOT-2020-07-22-20-24-07

How reproducible:


Steps to Reproduce:
1.install RHACM 2.0
2.install CP4MCM

Actual results:


Expected results:


Additional info:

Comment 1 Ginny Ghezzo 2021-02-26 13:08:10 UTC
Root Cause: not enough memory allocated to the standalone subscription container.

The memory limit of the multicluster-operators-standalone-subscription pod has been increased to 2G in the multicluster subscription community operator CSV. But this resource limit setting is ignored by OLM. This happens when OLM deploys all operators, not only the multicluster subscription operator.

This will be documented in ACM 2.2

Comment 5 errata-xmlrpc 2021-03-04 13:51:12 UTC
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 (Red Hat Advanced Cluster Management for Kubernetes version 2.2 images), 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/RHEA-2021:0729


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