The FDP team is no longer accepting new bugs in Bugzilla. Please report your issues under FDP project in Jira. Thanks.
Bug 2222252 - Investigate failure WARN in "load-balancer template IPv4/IPv6" system tests
Summary: Investigate failure WARN in "load-balancer template IPv4/IPv6" system tests
Keywords:
Status: CLOSED WONTFIX
Alias: None
Product: Red Hat Enterprise Linux Fast Datapath
Classification: Red Hat
Component: ovn23.09
Version: FDP 23.F
Hardware: Unspecified
OS: Unspecified
medium
low
Target Milestone: ---
: ---
Assignee: OVN Team
QA Contact: ying xu
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2023-07-12 12:19 UTC by Ales Musil
Modified: 2023-10-05 16:12 UTC (History)
5 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2023-10-05 16:12:19 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)
Logs from failed run (274.01 KB, application/zip)
2023-07-12 12:19 UTC, Ales Musil
no flags Details


Links
System ID Private Priority Status Summary Last Updated
Red Hat Issue Tracker FD-3007 0 None None None 2023-07-12 12:21:24 UTC

Description Ales Musil 2023-07-12 12:19:54 UTC
Created attachment 1975368 [details]
Logs from failed run

The test produce following warning in ovn-controller.log:

if_status|WARN|Trying to release unknown interface vm2

The warning is skipped for now to unblock the CI. We should investigate if it's produced by real bug in ovn-controller.

Comment 1 xsimonar 2023-07-13 14:12:43 UTC
It looks to me that there are three issues related to this:
1) When a logical port is added and deleted in 2 different IDL at ovn-sb, but handled within the same loop by ovn-controller, ovn-controller sees the port as deleted, and tries to remove it from if-status. The only issue here is that this situation should not cause a warning.
2) When conditional monitoring is used, and when the controller is registering the chassis, it uses the named-uuid of the chassis just created in a monitor condition. This is not the uuid of the chassis, and results in sb notifying us that all existing ports are deleted. As soon as ovn-controller got notified of the chassis update from sb, it will use the proper uuid, and situation will be back on track. This causes situation 1) to happen on test cases which do not delete ports.
3) System-tests are never executed with monitor-all=true despite what they tell (OVN_MONITOR_ALL is checked within ovn_attach, which is not called for system tests). So, even the monitor-all flavors of the test fail as... they use conditional monitoring.

Comment 4 OVN Bot 2023-07-26 04:09:13 UTC
ovn23.09 fast-datapath-rhel-9 clone created at https://bugzilla.redhat.com/show_bug.cgi?id=2226631
ovn23.06 fast-datapath-rhel-8 clone created at https://bugzilla.redhat.com/show_bug.cgi?id=2226632
ovn23.06 fast-datapath-rhel-9 clone created at https://bugzilla.redhat.com/show_bug.cgi?id=2226633
ovn23.03 fast-datapath-rhel-8 clone created at https://bugzilla.redhat.com/show_bug.cgi?id=2226634
ovn23.03 fast-datapath-rhel-9 clone created at https://bugzilla.redhat.com/show_bug.cgi?id=2226635
ovn22.12 fast-datapath-rhel-8 clone created at https://bugzilla.redhat.com/show_bug.cgi?id=2226636
ovn22.12 fast-datapath-rhel-9 clone created at https://bugzilla.redhat.com/show_bug.cgi?id=2226637
ovn22.09 fast-datapath-rhel-8 clone created at https://bugzilla.redhat.com/show_bug.cgi?id=2226638
ovn22.09 fast-datapath-rhel-9 clone created at https://bugzilla.redhat.com/show_bug.cgi?id=2226639

Comment 5 Mark Michelson 2023-10-05 16:12:19 UTC
I'm closing this. This issue got fixed during a time when there was a bug in ovn-sync that caused us to correctly create clones of issues for all versions of OVN in which a bug was fixed. However, we did not update the original issue. In this case, https://bugzilla.redhat.com/show_bug.cgi?id=2226631 took the place of this original issue and is in the proper state.


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