2

I have an Ubuntu Desktop system that is frequently crashing (once every 2-3 days maybe?). It was originally running 17.10 (and was crashing on this system) and when 18.04 came out I thought that upgrading might fix the frequent crashes which it didn't.

So I thought I would try to follow the instructions in this Ubuntu guide to enable kdump/linux-crashdump. Following the steps in the guide, everything seems to be correct (kdump-config show says ready to kdump, dmesg | grep -i crash shows reserving memory for crash dump, etc). However when it comes to the "testing crash dump mechanism" portion of the guide, the crash gets successfully triggered by echo c > /proc/sysrq-trigger but the line Saving vmcore from kernel crash never shows up and the system never reboots. Even if I manually reboot the machine, there is no dump file in /var/crash (which I confirmed is the location where the crash files are supposed to be).

I tried increasing the crashkernel size from 128M (the default) to 256M which I saw as a recommendation on some wikis/blog posts/whatever, but that didn't seem to do anything either.

Any thoughts on how to get some insight into these Kernel crashes?

For reference, here is the output from the three verification commands in the guide:

$ cat /proc/cmdline
BOOT_IMAGE=/boot/vmlinuz-4.15.0-20-generic root=UUID=20a96099-e38b-4dd7-b006-33c087b6f73b ro quiet splash crashkernel=384M-:256M vt.handoff=1

$ dmesg | grep -i crash
[    0.000000] Command line: BOOT_IMAGE=/boot/vmlinuz-4.15.0-20-generic root=UUID=20a96099-e38b-4dd7-b006-33c087b6f73b ro quiet splash crashkernel=384M-:256M vt.handoff=1
[    0.000000] Reserving 256MB of memory at 512MB for crashkernel (System RAM: 65468MB)
[    0.000000] Kernel command line: BOOT_IMAGE=/boot/vmlinuz-4.15.0-20-generic root=UUID=20a96099-e38b-4dd7-b006-33c087b6f73b ro quiet splash crashkernel=384M-:256M vt.handoff=1

$ kdump-config show
DUMP_MODE:        kdump
USE_KDUMP:        1
KDUMP_SYSCTL:     kernel.panic_on_oops=1
KDUMP_COREDIR:    /var/crash
crashkernel addr: 0x
   /var/lib/kdump/vmlinuz: symbolic link to /boot/vmlinuz-4.15.0-20-generic
kdump initrd: 
   /var/lib/kdump/initrd.img: symbolic link to /var/lib/kdump/initrd.img-4.15.0-20-generic
current state:    ready to kdump

kexec command:
  /sbin/kexec -p --command-line="BOOT_IMAGE=/boot/vmlinuz-4.15.0-20-generic root=UUID=20a96099-e38b-4dd7-b006-33c087b6f73b ro quiet splash vt.handoff=1 nr_cpus=1 systemd.unit=kdump-tools.service irqpoll nousb ata_piix.prefer_ms_hyperv=0" --initrd=/var/lib/kdump/initrd.img /var/lib/kdump/vmlinuz
3
  • You can use journalctl -b -1 -e to see the system log messages jsut before the crash. However, power fails, CPU overheating, CPU failures, etc can crash the system before anything gets logged. – waltinator May 18 '18 at 13:55
  • 1
    Did you ever figure this out? I want to update the Ubuntu serverguide, but am getting the same results as you. I also get crashkernel addr: 0x which I don't think is correct. – Doug Smythies Sep 23 '18 at 0:43
  • No sorry. I took another approach to fix my underlying problem (ended up being bad bios settings), so I never really followed up on this. – syazdani Sep 24 '18 at 2:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.