Monday, 2025-12-22

*** mhen_ is now known as mhen02:57
yasufumHi tacker team.08:02
shivamHello08:03
yasufumhi08:04
yasufum#startmeeting tacker08:04
opendevmeetMeeting started Mon Dec 22 08:04:55 2025 UTC and is due to finish in 60 minutes.  The chair is yasufum. Information about MeetBot at http://wiki.debian.org/MeetBot.08:04
opendevmeetUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.08:04
opendevmeetThe meeting name has been set to 'tacker'08:04
yasufum#link https://etherpad.opendev.org/p/tacker-meeting08:05
yasufumhi08:06
takahashi-tschi, sorry for the delay.08:06
yasufumWe,ve just started the meeting.08:06
yasufumme too actually08:06
yasufumOne topic on the etherpad from shivam08:07
yasufumshivam: Can you start the item?08:07
shivamyes08:07
shivamThe discussion topic is related to an error related to "DemoVirtualStorage" Images which caused Tacker Compliance Test Failures08:08
shivamSome Tacker compliance tests are failing during VNF instantiation because of duplicate images named DemoVirtualStorage already present.08:10
shivamAs per investigation, root cause seems that compliance testcases dynamically create images (e.g., DemoVirtualStorage) as part of test execution. These images are not getting cleaned up after the test execution completes.08:11
shivamAs a result, when a subsequent test run attempts to create an image with the same name, it finds multiple images with identical names, causing the instantiation to fail due to non-unique image matches.08:12
shivamError Message: "Error validating value 'DemoVirtualStorage': No images unique match found for {'name': 'DemoVirtualStorage'}"08:12
shivamOn local setups, this issue was temporarily resolved by manually deleting the already existing images before re-running the tests.08:13
shivamBut, since the same issue is also occurring in Zuul, a manual cleanup approach is not feasible, and  a permanent fix solution is required.08:13
shivamPossible fix approaches could be a post-test cleanup logic which should automatically delete the created DemoVirtualStorage images.08:14
shivamor Modifying the image creation configuration/parameter so that a unique image name is created for each test case.08:15
shivamA launchpad bug has been created for the same: https://bugs.launchpad.net/tacker/+bug/213698008:15
shivamSo, I would like to know the community’s thoughts on our bug, root cause analysis and fix suggestions.08:16
shivamAlso, if there are any additional fix suggestions/approach from community members?08:16
shivamThank you.08:16
YuyaKunoSorry for stupid question. In general, who should clean up after testing?08:17
yasufumthanks08:17
shivamkuno: As per my understanding, Ideally, it should be cleaned up by tacker test code itself because it is being created during test execution08:18
YuyaKunoOK, thank you. In NFV, the point was challenge of TST010. Currently no mechanism to clean up test environment by Robot framework.08:20
shivamok08:21
yasufumIMO, cleanup resources is not an essential issue without some specific usecases such as re-use a resource or so.08:23
yasufumIn this case from shivam, I agree to shivam's stragegy since it's simple and feasible for the issue.08:23
yasufumAny other comment?08:24
takahashi-tscI also agree, simple approach is better.08:25
yasufumthx08:25
yasufumgood08:26
shivamThank you. will proceed with the cleanup approach then. 08:27
yasufumthanks08:27
yasufumAll the topics done for today.08:27
yasufumBefore closing, I'd like to skip the next meeting, and also the first meeting next year.08:29
yasufumSo, it's the last one in this year.08:29
takahashi-tscSorry, just confirmation. Conclusion is that we adopt suggestion 2 "2. Modify the image creation configuration/parameter so that a unique image name is created for each test case.", right?08:29
yasufumyes08:30
takahashi-tscOK, thanks08:30
shivamOh.. ok, sorry I though it was first one08:31
shivamanyways, thanks for confirmation08:31
yasufumIn my understanding, deleting img is to release useless resource and renaming is to avoid duplicated name, so both are not contradictory.08:36
takahashi-tscYes, both apporach is possible, and one of them is enough. My suggestion is to adopt only suggestion 2 because it is easy approach.08:37
YuyaKunoIs there any capacity issue by suggestion 2?08:38
takahashi-tscShivam, do you have any views on it? It is an important point of views.08:41
shivamkuno-san: Hmm... I think yes there might be storage problem 08:41
shivamif the images are created for each testcase uniquely. Eventually, it will be required to delete those images to avoid any storage issue08:42
YuyaKunoI see.08:43
shivamBut since, we ar referring to only compliance testcase (on zuul) so, I am not sure whether it will cause the storage issue on zuul08:44
shivambecause for each build, a fresh environment will be created for execution08:45
yasufumThe amount of resources is not enough sometimes on zuul. So, we'd better to take care about.08:47
shivamok...08:50
yasufumI'm not sure it can be critical problem or not, it might be helpful to estimate the size of strage totally for the tests.08:57
shivamok.. In that case, i think we can check the resource utilization after applying second fix suggestion and then discuss the result with community.08:57
yasufumthanks08:57
shivamThank you.08:58
yasufumAny other comment?08:58
yasufumgood08:59
yasufumThank you for the discussion, guys.08:59
shivamthanks09:00
yasufumI've told that today is the last meeting in this year.09:00
yasufumI'll announce the first our meeting next year on the ML.09:01
yasufumThank you for joining.09:01
yasufumHave a great holiday, and happy new year!09:01
yasufumBye!09:02
shivamThank you. Bye09:02
takahashi-tscbye!09:02
yasufum#endmeeting09:03
opendevmeetMeeting ended Mon Dec 22 09:03:02 2025 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)09:03
opendevmeetMinutes:        https://meetings.opendev.org/meetings/tacker/2025/tacker.2025-12-22-08.04.html09:03
opendevmeetMinutes (text): https://meetings.opendev.org/meetings/tacker/2025/tacker.2025-12-22-08.04.txt09:03
opendevmeetLog:            https://meetings.opendev.org/meetings/tacker/2025/tacker.2025-12-22-08.04.log.html09:03

Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!