Closed jbtrystram closed 2 days ago
Skipping CI for Draft Pull Request.
If you want CI signal for your change, please convert it to an actual PR.
You can still manually trigger a test run with /test all
is there some flag to say the machine is expected to reboot ?
There is one to say that the VM is expected to crash: https://github.com/coreos/fedora-coreos-config/blob/testing-devel/tests/kola/kdump/crash/test.sh#L7
tags: skip-base-checks
You can likely find the corresponding option in the code to set that for this test.
tags: skip-base-checks
yep that was it. Thanks @travier !
I think this is now ready for review I added a couple of retry loops to let some time for kdump to generate the initramfs and write the logs.
The test now pass :
[coreos-assembler]$ /mnt/bin/kola run kdump.crash.ssh
=== RUN kdump.crash.ssh
--- PASS: kdump.crash.ssh (68.66s)
PASS, output in tmp/kola/qemu-2024-07-08-2208-10301
the RHCOS test fails on the SSH command to crash the kernel. I suspect it's not the latest revision that was tested. Anyway, the kdump work:
[ 6.910879] kdump[559]: Kdump is using the default log level(3).
[ 7.136098] kdump[605]: saving to core@10.0.2.2:/home/core/crash/10.0.2.15-2024-07-08-23:59:44
[ 8.007655] kdump[605]: saving to core@10.0.2.2:/home/core/crash/10.0.2.15-2024-07-08-23:59:44
[ 7.317830] kdump[609]: saving vmcore-dmesg.txt to core@10.0.2.2:/home/core/crash/10.0.2.15-2024-07-08-23:59:44
[ 8.189387] kdump[609]: saving vmcore-dmesg.txt to core@10.0.2.2:/home/core/crash/10.0.2.15-2024-07-08-23:59:44
[ 7.486731] kdump.sh[611]: 159+1 records in
[ 7.490292] kdump.sh[611]: 159+1 records out
[ 7.494151] kdump.sh[611]: 81809 bytes (82 kB, 80 KiB) copied, 0.000764738 s, 107 MB/s[ 8.358288] kdump.sh[611]: 159+1 records in
[ 8.361851] kdump.sh[611]: 159+1 records out
[ 8.365709] kdump.sh[611]: 81809 bytes (82 kB, 80 KiB) copied, 0.000764738 s, 107 MB/s
[ 7.647105] kdump[614]: saving vmcore-dmesg.txt complete
[ 8.518662] kdump[614]: saving vmcore-dmesg.txt complete
[ 7.657912] kdump[616]: saving vmcore
[ 8.529470] kdump[616]: saving vmcore
[ 9.665279] kdump.sh[617]:
Checking for memory holes : [ 0.0 %] /
Checking for memory holes : [100.0 %] |
Excluding unnecessary pages : [100.0 %] \
Checking for memory holes : [100.0 %] -
Checking for memory holes : [100.0 %] /
Excluding unnecessary pages : [100.0 %] |
Copying data : [ 36.7 %] \ eta: 2s
Copying data : [ 96.5 %] - eta: 0s
Copying data : [100.0 %] / eta: 0s
Copying data : [100.0 %] | eta: 0s
[ 10.536834] kdump.sh[617]:
Checking for memory holes : [ 0.0 %] /
Checking for memory holes : [100.0 %] |
Excluding unnecessary pages : [100.0 %] \
Checking for memory holes : [100.0 %] -
Checking for memory holes : [100.0 %] /
Excluding unnecessary pages : [100.0 %] |
Copying data : [ 36.7 %] \ eta: 2s
Copying data : [ 96.5 %] - eta: 0s
Copying data : [100.0 %] / eta: 0s
Copying data : [100.0 %] | eta: 0s
[ 9.734010] kdump.sh[618]: 114846+1727 records in
[ 9.737953] kdump.sh[618]: 114846+1727 records out
[ 9.741836] kdump.sh[618]: 59162607 bytes (59 MB, 56 MiB) copied, 1.9093 s, 31.0 MB/s
[ 10.605567] kdump.sh[618]: 114846+1727 records in
[ 10.609512] kdump.sh[618]: 114846+1727 records out
[ 10.613395] kdump.sh[618]: 59162607 bytes (59 MB, 56 MiB) copied, 1.9093 s, 31.0 MB/s
[ 9.900371] kdump[621]: saving vmcore complete
[ 10.771928] kdump[621]: saving vmcore complete
[ 9.913585] kdump[623]: saving the /run/initramfs/kexec-dmesg.log to core@10.0.2.2:/home/core/crash/10.0.2.15-2024-07-08-23:59:44/
[ 10.785086] kdump[623]: saving the /run/initramfs/kexec-dmesg.log to core@10.0.2.2:/home/core/crash/10.0.2.15-2024-07-08-23:59:44/
[ 10.109477] kdump[630]: Executing final action systemctl reboot -f
[ 10.981034] kdump[630]: Executing final action systemctl reboot -f
[ 10.126948] systemd[1]: Shutting down.
[ 10.998506] systemd[1]: Shutting down.
This test setups two machines to test if kdump successfully exports vmcore to a SSH destination.
Fixes https://github.com/coreos/fedora-coreos-tracker/issues/1753
This is not fully functional yet but ready for review I think. The test setup works ( i can see the logs created on the remote machine) but kola fails the test somehow :
is there some flag to say the machine is expected to reboot ?