Tumbleweed - Review of the week 2021/18
Dear Tumbleweed users and hackers, Week 18 was a regular one with almost daily snapshots being published. A total of six snapshots went out to the users (0429, 0430, 0502, 0503, 0504, and 0506). The main changes in those snapshots included: * GTK+ 3.24.29 * TeXLive 2021 * PHP 7.4.18 * Python 3.8.9 * We switched from openMPI2 to openMPI4 Changes currently being prepared in the staging areas: * GNOME 40.1 * KDE Plasma 5.21.5 * icu 69.1: for the ring packages, nodejs15 seems to be the last blocker * Switch from go 1.15 to go 1.16: most packages fixed, the last failure in Staging:M is cilium * GCC 11 as default compiler: Move from special-purpose Staging:Gcc7 to Staging:O (incl. openQA test coverage). There are a few more build failures to be addressed. * UsrMerge: The current state is that we checked in all changes and are planning on actually doing the switch somewhen in the not too far future, likely together with the planned distro-rebuild for GCC 11 Cheers, Dominique
I have noticed that since 04/27/2021 the number of passed tests in the builds has decreased from around 185 to around 135-137 and the number of softfailed tests has increased from around 32 to around 90. Are those new builds stable enough for daily use or does the large increase in soft failures indicate a problem that has not been addressed? Thanks!
Hi, Am Freitag, 7. Mai 2021, 20:55:15 CEST schrieb Joe Salmeri:
I have noticed that since 04/27/2021 the number of passed tests in the builds has decreased from around 185 to around 135-137 and the number of softfailed tests has increased from around 32 to around 90.
Pretty much all of those softfails are because the system doesn't go fully idle in the time openQA expects. This is https://bugzilla.opensuse.org/show_bug.cgi?id=1185565.
Are those new builds stable enough for daily use or does the large increase in soft failures indicate a problem that has not been addressed?
Both. The bug causing this issue shouldn't cause any issues other than possibly higher power consumption. A likely fix is pending upstream and as a workaround kfence will be disabled with the next kernel update (5.12.2+). Cheers, Fabian
Thanks!
Thanks very much for those details! I tried to find more details on what those softfails were about but I'm still finding my way around the openqa site. Is there a link where I can go to see the additional details you were seeing regarding those softfails? That'd be real helpful in the future. Joe
On Friday, 7 May 2021 23.06.24 CEST Joe Salmeri wrote:
Thanks very much for those details!
I tried to find more details on what those softfails were about but I'm still finding my way around the openqa site.
Is there a link where I can go to see the additional details you were seeing regarding those softfails? That'd be real helpful in the future.
From openqa.opensuse.org/ you can click on the latest build of Tumbleweed showing a test result overview, e.g. https://openqa.opensuse.org/tests/overview? distri=microos&distri=opensuse&version=Tumbleweed&build=20210506&groupid=1 then you will see a page with colored circles representing the test job result for each, yellow meaning "softfailed". Each circle can be clicked to show details of each job, e.g. https://openqa.opensuse.org/tests/1730511 on that page one can scroll down until finding a test module result in the table view called "softfailed", again the same yellow color. In the row there are little thumbnail boxes, called "steps", one says "Soft Failed". When you click on it you will find a bug reference to https://bugzilla.suse.com/show_bug.cgi?id=1178761 so not the same bug as Fabian mentioned but an earlier one with similar symptoms. I just added a link between two bugs in bugzilla to provide that context. Find more details about the individual job results and their meanings on http://open.qa/docs/#_jobs Have fun, Oliver
Hi Oliver, Thanks for your detailed explanation. When I get to the step where I want to go to the bug reference (for example the one you provided): https://bugzilla.suse.com/show_bug.cgi?id=1178761 I get the following: You are not authorized to access bug #1178761. Originally I got a message about not being logged in but after logging in I get the "not authorized" error so I cannot read the details. I created my account the other day and verified the email so not sure why I am getting the not authorized. Thoughts? Thanks! Joe
Hello Joe, Am Sonntag, 9. Mai 2021, 01:22:48 CEST schrieb Joe Salmeri: ...
https://bugzilla.suse.com/show_bug.cgi?id=1178761
I get the following:
You are not authorized to access bug #1178761.
In some cases it helps to replace s/suse.com/opensuse.org/ but not in this one....
Originally I got a message about not being logged in but after logging in I get the "not authorized" error so I cannot read the details.
I created my account the other day and verified the email so not sure why I am getting the not authorized.
Thoughts?
ask SUSE to make the bug public, its on their internal bug tracker Cheers Axel
On 09.05.2021 11:41, Axel Braun wrote:
Hello Joe,
Am Sonntag, 9. Mai 2021, 01:22:48 CEST schrieb Joe Salmeri: ...
https://bugzilla.suse.com/show_bug.cgi?id=1178761
I get the following:
You are not authorized to access bug #1178761.
In some cases it helps to replace s/suse.com/opensuse.org/ but not in this one....
Both refer to exactly the same server.
Thanks Axel. Any idea what that bug is? It appears that many of the softfails (at least a large number of the ones I've tried to look at) are related to that bug. Joe
Hi, Am 2021-05-09 um 01:22 schrieb Joe Salmeri:
Thanks for your detailed explanation.
When I get to the step where I want to go to the bug reference (for example the one you provided):
https://bugzilla.suse.com/show_bug.cgi?id=1178761
I get the following:
You are not authorized to access bug #1178761.
I could make it public now.
Thanks Oliver, I appreciate it. I am most interested in how Fabian was able to determine that "Pretty much all of those softfails are because the system doesn't go fully idle in the time openQA expects" I'm looking for a way to know when the softfails can safely be ignored. Is there some place that summarizes the process which makes it easy to come to the conclusion that Fabian did?
On Tuesday, 11 May 2021 15.13.56 CEST Joe Salmeri wrote:
Thanks Oliver, I appreciate it.
I am most interested in how Fabian was able to determine that "Pretty much all of those softfails are because the system doesn't go fully idle in the time openQA expects"
I'm looking for a way to know when the softfails can safely be ignored.
Is there some place that summarizes the process which makes it easy to come to the conclusion that Fabian did?
I don't know for sure. openSUSE release managers would need to tell how they do that exactly right now. Likely the looked at the output of "totest-manager" from https://github.com/openSUSE/opensuse-release-tools to find out which tests referenced which bug. totest-manager looks at openQA test results equivalent to a view https://openqa.opensuse.org/tests/overview? result=softfailed&result=failed&result=incomplete&result=timeout_exceeded&arch=&machine=&modules=&distri=opensuse&groupid=1&version=Tumbleweed# and releases the product, i.e. a new Tumbleweed build, as soon as there are no more unreviewed, unignorable test failures left. A not-ok test is reviewed if there is a comment in openQA with a ticket reference. And a reviewed test will be seen as "ignorable" for totest-manager (ttm) when the release managers inform ttm by an according openQA comment with "@ttm ignore" next to the ticket to which this ignore-comment should be linked. Have fun, Oliver
participants (7)
-
Andrei Borzenkov -
Axel Braun -
Dominique Leuenberger / DimStar -
Fabian Vogt -
Joe Salmeri -
Joe Salmeri
-
Oliver Kurz