[DRBD-user] `drbdadm primary all` fails but exits with status 0

Pavel Georgiev pavel at netclime.com
Thu Oct 18 16:33:25 CEST 2007

Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.


I`m using drbd 8.0pre5 that comes with ubuntu feisty and I have the following 
problem:

When I `reboot -f` the primary node in an active/passive mode, heartbeat fails 
to do the takeover. The reason in that the  /etc/ha.d/resource.d/drbddisk 
script tries to make the node primary before it knows that the other node is 
no longer available. The script does take of this by trying to run a few 
times `drbdadm primary all` but it evaluates the exit status. The real 
problem here is that `drbdadm primary all` exits with 0 even if it fails to 
make the current node primary because it thinks the other node is still 
alive:

root at client:~# drbdadm primary all
State change failed: (-1) Multiple primaries now allowed by config
root at client:~# echo $?
0

This makes takeover to fail. IS this a know issues, has anyone else had a 
problem with it? Is this a bug at all?



More information about the drbd-user mailing list