action #122143
closed[qe-core][functional] test fails in bootloader because grub rescue mode entered due to network issue
0%
Description
Observation¶
we have this issue quite long time and there is no workaround or fix on it yet.
openQA test in scenario sle-15-SP5-Online-ppc64le-textmode+role_textmode@ppc64le-hmc fails in
bootloader
Test suite description¶
Maintainers: QE Core, mgriessmeier
Like default but explicitly select the system role "textmode".
Reproducible¶
Fails since (at least) Build 61.1 (current job)
Expected result¶
Last good: 58.1 (or more recent)
Further details¶
Always latest result in this scenario: latest
Updated by zluo over 1 year ago
- Related to action #120570: [qe-core][functional][tools] test fails in bootloader because root device is not ready and it leads to kernel panic size:M added
Updated by zluo over 1 year ago
grenache pvm has trouble, so after reboot it looks to be normal, however it has a new issue with reading file under /boot:
Updated by zluo over 1 year ago
- Subject changed from [qe-core][functional] test fails in bootloader because shutdown of lpar takes too long to [qe-core][functional] test fails in bootloader because grub rescue mode entered due to network issue
change the title now to work on a workaround for the issue with grub rescue mode.
Updated by zluo over 1 year ago
to catch up the case of grub rescue mode is not a problem:
https://github.com/os-autoinst/os-autoinst-distri-opensuse/pull/16120
but re-trying reset lpar netboot still not really working, so I think the root issue is at setup/allocation of pvm.
If I triggered 10 job, all worked fine without any issue, but if I triggered more than 20 jobs, the sporadic issue happens.
if issue like loading initrd gets error, this cannot be resolved by re-try.
Updated by rfan1 about 1 year ago
zluo wrote:
to catch up the case of grub rescue mode is not a problem:
https://github.com/os-autoinst/os-autoinst-distri-opensuse/pull/16120but re-trying reset lpar netboot still not really working, so I think the root issue is at setup/allocation of pvm.
If I triggered 10 job, all worked fine without any issue, but if I triggered more than 20 jobs, the sporadic issue happens.
if issue like loading initrd gets error, this cannot be resolved by re-try.
One minor comment, can we consider to renew the NBP file on dhcp/install server? [it is built on Sep/2020]
lrwxrwxrwx 1 root root 30 Sep 21 2020 grub2 -> boot/powerpc-ieee1275/core.elf
qanet:/srv/tftp/ppc64le/boot/grub2-ieee1275/powerpc-ieee1275 got updated now, thanks for report.
Updated by zluo about 1 year ago
- Status changed from In Progress to Resolved
found root cause, see PR https://github.com/os-autoinst/os-autoinst-distri-opensuse/pull/16711