[DRBD-user] DRBD replication ... does not make ext4 cluster aware

Lars Ellenberg lars.ellenberg at linbit.com
Tue Sep 6 11:08:23 CEST 2016

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


On Tue, Sep 06, 2016 at 03:58:09PM +0800, Chee Khoon Bong wrote:
> I have setup drbd on 2 node, both were ext4 fs, I have sync 2 node and it
> was both up to date. I mounted /dev/drbd0 on /mnt/ext4 on both node. When I
> create new file on /mnt/ext4/ on node 1 , i using "ls" to check for whether
> the file is replicated on /mnt/ext4 on node 2 but i found none. Once i
> restart node 2, the file created on node 1 is appearing on my node 2.

Concurrently using (dual primary, multi primary) DRBD on several
nodes requires the use of a (shared-disk-) cluster file system
such as GFS2 or OCFS2.  [or any other cluster aware thingy that
take care of locking and cache coherency issues]

"classic" (local-only, single node, exclusive access) file systems are
not cluster aware, and MUST NOT be concurrently mounted by multiple nodes.
Doing so anyways will trash your data.

-- 
: Lars Ellenberg
: LINBIT | Keeping the Digital World Running
: DRBD -- Heartbeat -- Corosync -- Pacemaker

DRBD® and LINBIT® are registered trademarks of LINBIT
__
please don't Cc me, but send to list -- I'm subscribed



More information about the drbd-user mailing list