[Pacemaker] pacemaker, heartbeat, cts and drbd

Raoul Bhatia [IPAX] r.bhatia at ipax.at
Fri Feb 6 07:50:43 EST 2009


hi,

i have a problem with cts (using pacemaker plus heartbeat) and drbd
devices, and do not know how to handle this. maybe you can enlighten
me:


i have a drbd device as a master-slave resource. my configuration
includes

* dopd configuration in ha.cf
> # enable dopd
> respawn hacluster /usr/lib/heartbeat/dopd
> apiauth dopd gid=haclient uid=hacluster

* outdate-peer configuration in drbd.conf
> outdate-peer "/usr/lib/heartbeat/drbd-peer-outdater -t 5";


the first thing cts does ist cycling all nodes to stop and then again to
start the clusterstack:

* stop node1
* stop node2
* start node1
* start node2

now, when node1 goes down, node2 will be the only up2date disk in the
cluster. then, node2 goes done and node1 is started. node1 will then
stonith node2 after some time when node2's clusterstack is not restarted
in time.

i have seen two things here


1) the outdating worked which means that node1 refuses to bring up the
drbd device in primary mode (as it has been marked as outdated).

this leaves the cluster in a semi-functional state and it needs node2
up and running to resolve this issue.

cts of course generates some errors (BadNews) for that.


is this working as intended? can i ignore those BadNews errors?

(because of some drbd.conf issues - having after="xxx" "messing" around
- i wasn't able to reproduce this issue and leave it running for e.g. 10
minutes.




2) pacemaker starts the drbd devices:
> Master/Slave Set: ms_drbd_mysql
>     drbd_mysql:0        (ocf::heartbeat:drbd):  Master wc01
>     drbd_mysql:1        (ocf::heartbeat:drbd):  Started wc02
..
> Master/Slave Set: ms_drbd_www
>     drbd_www:0  (ocf::heartbeat:drbd):  Slave wc01
>     drbd_www:1  (ocf::heartbeat:drbd):  Master wc02

but drbd is in standalone mode:

# cat /proc/drbd
version: 8.2.7 (api:88/proto:86-88)
GIT-hash: cd953e17c5eb47e7df835024fee4b13e5fcaa217 debian/rules build by
root at wc02, 2009-01-31 18:53:19
 0: cs:StandAlone st:Primary/Unknown ds:UpToDate/Outdated   r---
    ns:0 nr:0 dw:1776 dr:13913 al:56 bm:0 lo:0 pe:0 ua:0 ap:0 ep:1 wo:b
oos:1676
 1: cs:StandAlone st:Secondary/Unknown ds:UpToDate/Outdated   r---
    ns:0 nr:0 dw:0 dr:0 al:0 bm:0 lo:0 pe:0 ua:0 ap:0 ep:1 wo:b oos:224



for this, i have to review the exact things that happened there (when
has what been outdated, etc.)


cheers,
raoul
-- 
____________________________________________________________________
DI (FH) Raoul Bhatia M.Sc.          email.          r.bhatia at ipax.at
Technischer Leiter

IPAX - Aloy Bhatia Hava OEG         web.          http://www.ipax.at
Barawitzkagasse 10/2/2/11           email.            office at ipax.at
1190 Wien                           tel.               +43 1 3670030
FN 277995t HG Wien                  fax.            +43 1 3670030 15
____________________________________________________________________




More information about the Pacemaker mailing list