+++ This bug was initially created as a clone of Bug #2269991 +++
kdump-capture.service
- Kdump Vmcore Save Service...
[ 20.180771] systemd[1]: Starting kdump-capture.service - Kdump Vmcore Save Service...
[ 20.241361] kdump[457]:
Kdump is using the default log level(3).
[ 20.361272] kdump.sh[444]: [debug_mem] kdump saving vmcore
[ 20.361620] kdump.sh[444]: MemFree: 202688 kB
[ 20.366827] kdump.sh[444]: Cached: 101056 kB
[ 20.369049] kdump.sh[444]: Slab: 50368 kB
[ 20.398942] kdump[493]: saving to /sysroot/var/crash/127.0.0.1-2024-03-17-11:33:11/
[ 20.420740] kdump[498]: saving vmcore-dmesg.txt to /sysroot/var/crash/127.0.0.1-2024-03-17-11:33:11/
[ 20.477668] kdump[504]: saving vmcore-dmesg.txt complete
[ 20.480576] kdump[506]: saving vmcore
[ 20.529767] kdump.sh[507]: Checking for memory holes : [ 0.0 %] / Checking for memory holes : [100.0 %] | readpage_elf: Attempt to read non-existent page at 0xc000000000000.
[ 20.530079] kdump.sh[507]: readmem: type_addr: 0, addr:c00c000000000000, size:16384
[ 20.530291] kdump.sh[507]: __exclude_unnecessary_pages: Can't read the buffer of struct page.
[ 20.530475] kdump.sh[507]: create_2nd_bitmap: Can't exclude unnecessary pages.
[ 20.531713] kdump.sh[507]: The kernel version is not supported.
[ 20.536928] kdump.sh[507]: The makedumpfile operation may be incomplete.
[ 20.537135] kdump.sh[507]: makedumpfile Failed.
[ 20.540278] kdump[509]:
saving vmcore failed, exitcode:1
Reproducible: Always
--- Additional comment from Baoquan He on 2024-03-18 02:30:49 UTC ---
This is reported by CKI:
https://datawarehouse.cki-project.org/kcidb/tests/11702842
The details can be found in test_console.log:
https://s3.amazonaws.com/arr-cki-prod-trusted-artifacts/trusted-artifacts/1216602630/test_ppc64le/6411623983/artifacts/run.done.02/job.01/recipes/15750794/tasks/8/logs/test_console.log
--- Additional comment from Coiby on 2024-03-18 02:41:56 UTC ---
This issue has been fixed in kexec-tools-2.0.28-6.fc41.x86_64 which now depends on makedumpfile.
--- Additional comment from Coiby on 2024-03-18 02:49:27 UTC ---
And I can confirm this issue has been gone with Fedora-Rawhide-20240317.n.0.
--- Additional comment from Baoquan He on 2024-03-18 06:49:09 UTC ---
Hi Coiby,
This is a different issue related to ppc64le makedumpfile filtering error, you are supposed to close below makedumpfile package issue:
Bug 2269640 - makedumpfile command not found when building kdump initramfs
So I reopen this bug.
Thanks
Baoquan
--- Additional comment from Baoquan He on 2024-03-19 07:32:54 UTC ---
Add IBM power experts into this bz in case they happen to know the reason.
--- Additional comment from IBM Bug Proxy on 2024-03-19 08:41:03 UTC ---
------- Comment From mainamdar.com 2024-03-19 04:10 EDT-------
This is reverse Mirrored Bug 2269991 - makedumpfile failed when dumping vmcore
--- Additional comment from IBM Bug Proxy on 2024-03-19 09:22:13 UTC ---
------- Comment From hbathini.com 2024-03-19 05:17 EDT-------
The below makedumpfile fix:
https://github.com/makedumpfile/makedumpfile/commit/94241fd2feed0
and crash-utility fix:
https://github.com/crash-utility/crash/commit/5b24e363a8980
are needed since kernel commit:
https://github.com/torvalds/linux/commit/55c49fee57af9
--- Additional comment from Baoquan He on 2024-03-25 06:10:31 UTC ---
Thanks a lot for your quick response and the helpful link, Hari.
--- Additional comment from Dusty Mabe on 2024-03-26 13:45:06 UTC ---
We are hitting this in the CoreOS pipelines too. CoreOS tracker issue in https://github.com/coreos/fedora-coreos-tracker/issues/1698