[Linux-cluster] Problem with more than one partition

[Robert Peterson]

aleava at libero.it wrote:
> First of all, thanks for your aswers,unfortunately, they don't solve our problem.
> I try to explain better: we have 1 diskarray with 4 LUNS mapped on 4 partitions (sda, sdb, sdc, sdd).
> I created 4 Volume group and 4 Logical Volume, one LV for every VG:
> /dev/VG01/LV01 (==>sda)
> /dev/VG02/LV02 (==>sdb)
> /dev/VG03/LV03 (==>sdc)
> /dev/VG04/LV04 (==>sdd)
> Then I make the gfs file system with the next command:
> gfs_mkfs -j 2 -t alpha_cluster:LV01 -p lock_dlm /dev/VG01/LV01
> gfs_mkfs -j 2 -t alpha_cluster:LV02 -p lock_dlm /dev/VG02/LV02
> gfs_mkfs -j 2 -t alpha_cluster:LV03 -p lock_dlm /dev/VG03/LV03
> gfs_mkfs -j 2 -t alpha_cluster:LV04 -p lock_dlm /dev/VG04/LV04
>
> I can mount all these 4 partition from one node without problems, but when I try to mount these same partitions from the other node, the mount command hangs and I must reset the machine, because also the "reboot" command doesn't work.
>
> I hope I provide all informations to make you understand my problem.
>
> Thanks in advance,
> Ale
>   
Hi Ale,

Your mkfs's and cluster.conf seem reasonable.  I'd also recommend bumping
your post_join_delay, though, because 3 seconds isn't much time.

It's hard to say what's going on.
This sounds like a bug, but since you said that you were running RHEL4U1, I
recommend upgrading both nodes to the latest U4 and see if it still happens.
There were lots of bugs fixed between U1 and U4.  BTW, you shouldn't mix
a U1 system with a cluster system at a higher level.

Regards,

Bob Peterson
Red Hat Cluster Suite