[ZODB-Dev] Clustering ZEO Server with drbd
Jens Vagelpohl
jens at dataflake.org
Mon May 29 12:09:50 EDT 2006
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 29 May 2006, at 16:59, garthpl at periweb.com wrote:
> If you setup drbd and heartbeat , the spare will take over the
> floating ip from the zeo server and restart the service. drbd is
> simply doing a block level replication of the zeo server file
> system where the zeo data is stored. The question is what files
> needs to be replicated so that when the the service is restarted on
> the spare, all the files needed by the zeo server to operate are
> there? I'm trying to ensure from a system architecture standpoint
> I'm not missing anything.
>
I would replicate any file starting with "Data.fs" in the file name.
They are all in the "var" subdirectory of the ZEO server instance.
To be on the safe side I would not run the ZEO server process on the
standby machine until you actually switch to it since the replicating
file system will keep modifying the database file and the standby ZEO
server process will not know about it.
I'd be interested in test results and how much of a I/O-slowdown you
see on the primary ZEO server compared to a standard non-replicating
setup. ZEO servers are completely I/O-bound and if you have a large
setup it is important to ensure disk I/O stays as fast as possible.
jens
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (Darwin)
iD8DBQFEexzORAx5nvEhZLIRAk6GAKC4pX7O8+76PSpDC6THqFcC7nlzLwCgjAlK
j+LSOg7W0tq9iMqRFLezAaY=
=ctLf
-----END PGP SIGNATURE-----
More information about the ZODB-Dev
mailing list