memcg_stress from ubuntu_ltp_controllers fails with missing process for kill

Bug #1946201 reported by Krzysztof Kozlowski
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
ubuntu-kernel-tests
New
Undecided
Unassigned

Bug Description

memcg_stress from ubuntu_ltp_controllers:
---
15062 memcg_stress_test 1 TINFO: timeout per run is 0h 35m 0s
15063 memcg_stress_test 1 TINFO: Calculated available memory 191257 MB
15064 memcg_stress_test 1 TINFO: Testing 150 cgroups, using 1274 MB, interval 5
15065 memcg_stress_test 1 TINFO: Starting cgroups
15066 memcg_stress_test 1 TINFO: Testing cgroups for 900s
15067 memcg_stress_test 1 TINFO: Killing groups
15068 /opt/ltp/testcases/bin/memcg_stress_test.sh: 1: kill: No such process
15069
15070 INFO: ltp-pan reported some tests FAIL

---

Found on 2021.09.27/bionic/linux-azure-4.15/4.15.0-1125.138 instance Standard_D48_v3

tags: added: 4.15 azure bionic hinted sru-20210927 ubuntu-ltp
Revision history for this message
Po-Hsu Lin (cypressyew) wrote :

This is a peculiar one, I can see this failure on B-gcp-4.15 with instance n1-highmem-16 as well.

And I think it's causing the following 22 cgroup_fj_* test to fail (just like the memcg_test_3 causing these to fail in bug 1836694)

I tried to run this manually on n1-highmem-16 and it looks like:
  1. This test will pass if you run it as the first test, e.g.
     echo "memcg_stress memcg_stress_test.sh" > /tmp/target
     sudo /opt/ltp/runltp -f /tmp/target
  2. This issue can be reproduced by just running cgroup test first:
     echo "cgroup cgroup_regression_test.sh" > /tmp/targets
     echo "memcg_stress memcg_stress_test.sh" >> /tmp/targets
     sudo /opt/ltp/runltp -f /tmp/target

Po-Hsu Lin (cypressyew)
tags: added: ubuntu-ltp-controllers
removed: ubuntu-ltp
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.