[DRBD-user] Problem with drbd 8.4.1 on linux kernel 3.0.0

Kaloyan Kovachev kkovachev at varna.net
Fri Feb 3 12:44:13 CET 2012


On Fri, 3 Feb 2012 10:57:56 +0000, Owen Le Blanc <LeBlanc at man.ac.uk>
wrote:
> I have a test system using a linux kernel 3.0.0 with drbd 8.4.1 to try
> using the new feature with two volumes in one resource.  The
> difficulty I have is that the two nodes lose connection with each
> other for no apparent reason, and I am not able reliably to get them
> to reconnect.  In this respect, I might mention that the command
> 'syncer', which still appears in the 8.4 documentation for drbdadm, is
> no longer recognised.
> 
> On the primary node, /proc/drbd reads
> 
>  0: cs:StandAlone ro:Primary/Unknown ds:UpToDate/DUnknown   r-----
>     ns:0 nr:0 dw:116 dr:472 al:2 bm:0 lo:0 pe:0 ua:0 ap:0 ep:1 wo:f
oos:232
>  1: cs:StandAlone ro:Primary/Unknown ds:UpToDate/DUnknown   r-----
>     ns:0 nr:0 dw:116 dr:472 al:2 bm:0 lo:0 pe:0 ua:0 ap:0 ep:1 wo:f
oos:232
> 
> and on the secondary node
> 
>  0: cs:WFConnection ro:Secondary/Unknown ds:UpToDate/DUnknown C r-----
>     ns:0 nr:0 dw:0 dr:0 al:0 bm:0 lo:0 pe:0 ua:0 ap:0 ep:1 wo:f oos:1516
>  1: cs:WFConnection ro:Secondary/Unknown ds:UpToDate/DUnknown C r-----
>     ns:0 nr:0 dw:0 dr:0 al:0 bm:0 lo:0 pe:0 ua:0 ap:0 ep:1 wo:f oos:1516
> 

I have got a split brain. Outdate then stop drbd on the secondary -> do
(re)connect on the primary -> start secondary again

> Previously I have used drbdadm with attach/syncer/connect on both nodes,
> and sometimes I've had to use --overwrite-data-of-peer or
> --discard-data-of-peer to get out of this.  I'm aware of the new
> syntax using --force, but that doesn't seem to help either.
> 
>      -- Owen Le Blanc
>      University of Manchester
> _______________________________________________
> drbd-user mailing list
> drbd-user at lists.linbit.com
> http://lists.linbit.com/mailman/listinfo/drbd-user


More information about the drbd-user mailing list