Project

General

Profile

Actions
Copy link

action #177138

closed

[tools]test fails in s390x zkvm bootloader_start tests, define a new vm fails due to "Cannot set interface flags on 'macvtapxx': Address already in use"

Added by rfan1 18 days ago. Updated 14 days ago.

Status:
Resolved
Priority:
High
Assignee:
mkittler
Category:
Bugs in existing tests
Target version:
openQA Project (public) - Ready
Start date:
2025-02-13
Due date:
% Done:

0%

Estimated time:
Difficulty:
Tags:
infra, s390x, reactive work

Description

Description

The issue is seen from ~1 hour ago, when openQA uses virsh command to define a vm, it reports the below error messages:

[2025-02-13T10:18:38.242390Z] [debug] [pid:98929] [run_ssh_cmd(virsh  start openQA-SUT-2 2> >(tee /tmp/os-autoinst-openQA-SUT-2-stderr.log >&2))] stderr:
  error: Failed to start domain 'openQA-SUT-2'
  error: Cannot set interface flags on 'macvtap148': Address already in use

I did some investigation on s390zl12, and found that there were some stale macvtap NICs even no running VM.

rfan@s390zl12:~> ip a|grep macvtap
28471: macvtap57@vlan2114: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 500
28472: macvtap58@vlan2114: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 500
28473: macvtap59@vlan2114: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 500
28474: macvtap60@vlan2114: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 500
28475: macvtap61@vlan2114: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 500
28476: macvtap62@vlan2114: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 500
28477: macvtap63@vlan2114: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 500
28478: macvtap64@vlan2114: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 500
28479: macvtap65@vlan2114: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 500
28480: macvtap66@vlan2114: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 500
28481: macvtap67@vlan2114: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 500
28482: macvtap68@vlan2114: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 500
28483: macvtap69@vlan2114: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 500

Observation

openQA test in scenario sle-micro-6.2-Base-qcow-s390x-ssh_mm_whitelist@s390x-kvm fails in
bootloader_start

Test suite description

SLE Micro image boot with ignition disk and default tests. Default tests are transactional-update, rebootmgr, health_check, cockpit service and some other checks specific to SLE Micro.

Reproducible

Fails since (at least) Build rfan0214

Expected result

Last good: (unknown) (or more recent)

Further details

Always latest result in this scenario: latest

Actions
Copy link
 #1

Updated by okurz 18 days ago

  • Tags set to infra, reactive work, s390x
  • Target version set to Ready
Actions
Copy link
 #2

Updated by mkittler 18 days ago

Also see https://suse.slack.com/archives/C02CANHLANP/p1739441650120979 and another thread linked from there for additional context.

Actions
Copy link
 #3

Updated by mkittler 18 days ago

  • Assignee set to mkittler

I restarted one of the affected test. The affected test failed 8 hours ago. The restarted test is now past the critical point, see https://openqa.suse.de/tests/16759998#details.

I have also seen other tests successfully running on previously affected worker slots. So perhaps this issue has already resolved itself.

Actions
Copy link
 #4

Updated by okurz 18 days ago

  • Priority changed from Normal to High
Actions
Copy link
 #5

Updated by mkittler 17 days ago ยท Edited

  • Status changed from New to In Progress

The restarted test worked. So I restarted other jobs that failed due to this and I'm currently monitoring them. (Unfortunately they are not picked up right now because of the max. running job limit.)

Actions
Copy link
 #6

Updated by mkittler 17 days ago

  • Status changed from In Progress to Feedback

The jobs I restarted were picked up and all are now passed the critical part. So I wouldn't know what else to do. We could just consider this resolved as well.

Actions
Copy link
 #7

Updated by rfan1 15 days ago

Thanks all for your help. now the issue is gone

Actions
Copy link
 #8

Updated by mkittler 14 days ago

  • Status changed from Feedback to Resolved
Actions
Copy link

Also available in: Atom PDF