[ClusterLabs] Strange Corosync (TOTEM) logs, Pacemaker OK but DLM stuck

Ferenc Wágner wferi at niif.hu
Wed Aug 30 04:51:23 EDT 2017


"Ulrich Windl" <Ulrich.Windl at rz.uni-regensburg.de> writes:

>>>> Ferenc Wágner <wferi at niif.hu> schrieb am 28.08.2017 um 18:07 in Nachricht
> <87mv6jk75r.fsf at lant.ki.iif.hu>:
>
> cLVM under I/O load can be really slow (I'm talking about delays in the range
> of a few seconds).

Yes, I know, and it's really annoying.  I wonder why this needs to be
so...  There really should be a way to tune that, but I never digged
into the issue, just switched off the VG activation timeout instead.

> Be sure to have any timeouts adjusted accordingly. I wrote a tool that
> allows to monitor the read latency (as seen by applications), so I
> know these numbers.

Do you mean the LVM metadata read latency as seen by the LVM tools or
that of a mirrored data region?

> And things get significantly worse if you do cLVM mirroring with a
> mirrorlog replicated to each device.

I don't do any such things, so my data paths are independent of cLVM and
DLM.  And in this thread I'm trying to stress that my issue should be
independent of any disk related latency, because everything is in memory
already.

> Maybe the CLVM slows down at n^2, where n is the number of nodes; I
> don't know ;-)

It's a pig for sure.
-- 
Thanks,
Feri




More information about the Users mailing list