|
Bugzilla – Full Text Bug Listing |
| Summary: | [kernel] openqa tests failing after update to 6.4.2 | ||
|---|---|---|---|
| Product: | [openSUSE] openSUSE Tumbleweed | Reporter: | Ana Guerrero <ana.guerrero> |
| Component: | Kernel | Assignee: | Santiago Zarate <santiago.zarate> |
| Status: | IN_PROGRESS --- | QA Contact: | E-mail List <qa-bugs> |
| Severity: | Normal | ||
| Priority: | P5 - None | CC: | acarvajal, ana.guerrero, dmueller, jslaby, lukas.kucharczyk, tiwai |
| Version: | Current | ||
| Target Milestone: | --- | ||
| Hardware: | Other | ||
| OS: | Other | ||
| Whiteboard: | |||
| Found By: | --- | Services Priority: | |
| Business Priority: | Blocker: | --- | |
| Marketing QA Status: | --- | IT Deployment: | --- |
|
Description
Ana Guerrero
2023-07-12 14:20:16 UTC
Some different values about time are likely the side-effect of switching CONFIG_HZ from 250 to 300 (bsc#1196438), not really due to the update to 6.4.x, I suppose. That said, as long as the values are close from the original values, they should be fine. Adding Dirk to Cc. The lack of /proc/fs/quota/warnings is intentional. It's been already an obsoleted interface, and it's finally disabled as CONFIG_BROKEN. And, some of errors about kernel modules are probably because of the lockdown enablement. (In reply to Takashi Iwai from comment #2) > And, some of errors about kernel modules are probably because of the > lockdown enablement. Note that lockdown was not enabled in 6.4.2 yet (by mistake -- a bad merge). (In reply to Jiri Slaby from comment #3) > (In reply to Takashi Iwai from comment #2) > > And, some of errors about kernel modules are probably because of the > > lockdown enablement. > > Note that lockdown was not enabled in 6.4.2 yet (by mistake -- a bad merge). Ah, interesting. The error shown there: tst_test.c:1558: TINFO: Timeout per run is 0h 00m 30s insmod: ERROR: could not insert module dummy_del_mod.ko: Key was rejected by service tst_cmd.c:121: TBROK: 'insmod' exited with a non-zero code 1 at tst_cmd.c:121 made me jumping to a wrong conclusion. The error implies that it's likely something to do with the signing, though. (In reply to Ana Guerrero from comment #0) > The first testsuite is sys-param-check, see > https://openqa.opensuse.org/tests/3421666 > There are a few tests with sysctl failing. > The code of the tests is from https://github.com/openSUSE/sys-param-check Yes, I think these are bad tests. Takashi correctly pointed out the CONFIG_HZ change. I am not sure if the HZ change brought anything else except random test failures :P. > The second testsuite is jeos-ltp-syscalls@uefi_virtio-2G see: > https://openqa.opensuse.org/tests/3421762 > There are 7 tests failings related to kernel modules. But yeah, I commented on those rejected keys of self-built modules in some bug already. Simply don't try to load unsigned modules with secureboot-enabled virtual machines. (The test is bad.) I'm not sure what sched_rr_get_interval01 tests. Is that new? (In reply to Jiri Slaby from comment #5) > I'm not sure what sched_rr_get_interval01 tests. Is that new? Ah, I see: TFAIL: /proc/sys/kernel/sched_rr_timeslice_ms != 100 got 90 I.e. HZ-related. (In reply to Jiri Slaby from comment #5) > But yeah, I commented on those rejected keys of self-built modules in some > bug already. bug 1211068 comment 1 that was. Thank you for all your replies! I have filed requests for the QA people to update the tests. I'm leaving this bug open for now in case more issues with tests appear when 6.4.3 with the lockdown patch arrives to openQA. Dirk, about the PR you made, there are more things to updated. I've filed https://progress.opensuse.org/issues/132677 |