Comment 11 for bug 1644056

Revision history for this message
kokushibyou (kokushibyou) wrote :

I am seeing this problem on heavily utilized HP G9 servers, specifically when the raid battery fails. The battery will fail, the chassis will print a similar error in dmesg, and the heavily utilized process (in this case Java/Elastic search) will stop responding. A reboot recovers the host, the HP utilities then report the raid battery is bad. We swap the battery, reboot the host, and it recovers.

WE've had this occurs 6 different times on 6 different boxes. The HP G9's are known to have troublesome raid batteries, I just didn't expect the kernel to have such an issue with it.

[Tue Oct 17 01:06:15 2017] /build/linux-EO9xOi/linux-4.4.0/mm/pgtable-generic.c:33: bad pmd ffff88156c4320b8(000000212e6009e2)

uname -a
Linux xx 4.4.0-59-generic #80-Ubuntu SMP Fri Jan 6 17:47:47 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux

cat /etc/issue
Ubuntu 16.04.1 LTS \n \l