On Mon, 11 Dec 2017 19:07:05 +0100 Karsten Heiken <
hei...@luis.uni-hannover.de> wrote:
> Bernhard Schmidt wrote:
> >Can you check whether numa=off on the kernel command line fixes this as
well?
>
> Indeed it does. Appending numa=off to the kernel command line fixes
> the bug in my KVM virtual
The provided work-around (setting numa=off) worked here.
Do you need further information about the system in use?
BTW: Thank you for the NUMA advice, it was about 21.00 CET when
I started a search engine and found the bug entry and the suggested
work-around, so, your advice came just in time.
Bernhard Schmidt wrote:
Can you check whether numa=off on the kernel command line fixes this as well?
Indeed it does. Appending numa=off to the kernel command line fixes
the bug in my KVM virtual machines as well as my physical servers (at
least the ones I've tested so far).
So this might be
Hi Karsten,
Thanks for the test. Can you check whether numa=off on the kernel command line
fixes this as well?
Bernhard
--
Diese Nachricht wurde von meinem Android-Mobiltelefon mit K-9 Mail gesendet.
I can reproduce the bug on multiple (physical) servers running in our
data center.
When booting my virtual servers with the new kernel, the issue did not
arise...
...until I enabled NUMA.
Booting a virtual machine with 2 sockets and 16 cores each works fine
when NUMA is disabled in KVM.
With
Just as a note:
Upgrading to jessie-backports version 4.9.51-1~bpo8+1 is also an option,
which works.
Christoph
<>
signature.asc
Description: OpenPGP digital signature
6 matches
Mail list logo