[ZODB-Dev] Horizon for highly-available ZODB storage?

[Fabiano Weimar dos Santos]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Sean,

Em Qui 03 Jan 2002 19:46, sean.upton@uniontrib.com escreveu:
> This is an interesting idea; the nice thing about a replicated block device
> is that it should replicate data rather quickly, though, if I understand
> correctly, the FS has to be unmounted until the backup node takes over.

Yes and No. The master file system is mounted and the slave is umounted. If 
the master fail, the HEARTBEAT and IPMON will mount the slave as a master via 
nfs until the master come back to run.
 
> Are there any issues with corruption of a FileStorage Data.fs that have to
> be dealt with as the result of an incomplete replication of data blocks via
> DRBD, if for example, there is a sudden power outage on the primary node?

Because of this kind of problem that we have to use a journaled file system. 
Thus, data loss is very hard.

> My main concern is the ability to have reliability without the need for
> human intervention, until, perhaps after the fact - guaranteeing
> site/service uptime/integrity.

Good intensions. In my project i allways have the same preocupations.  

The main funtion od DRDB is to suply the replication with fault tolerance in 
the file system. You will replicate the ZODB in the 2 server for, under a 
fail situation of a server, the IPMON will detect the fail, stop the services 
restart with a new configuration (with the help of some shell scripts).

Take a look at the howto. The idea is very flexible and i think that is a 
immediate alternative (but it's not easy to configurate).

- -- 
Fabiano Weimar dos Santos
X3ng Web Technology
Consultoria e Desenvolvimento
Celular: 54 9114-7230
ICQ: 40187432
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iD8DBQE8NNWKTqbDpNXhEbERAmwKAKCihU/r//IPXLpDi8mL0PLX7MteAgCg7apW
CW8anye/U+NXHnYiQDXHyjI=
=aWkB
-----END PGP SIGNATURE-----