From: Lars Marowsky-Bree (lmb@suse.de)
Date: Thu Sep 18 2003 - 10:28:16 CEST
Date: Thu, 18 Sep 2003 10:28:16 +0200 From: Lars Marowsky-Bree <lmb@suse.de> Message-ID: <20030918082815.GA29799@marowsky-bree.de> Subject: Re: [suse-oracle] Linux-Clustering
On 2003-09-17T22:02:40,
Martin Konold <martin.konold@erfrakon.de> said:
> DRDB is a _very_ bad recommendation for a production system. DRDB is
> unreliable by any definition.
>
> Simply make up the potential fault matrix and you will notice that DRDB will
> not provide the safety expected from a failover cluster.
This is plain wrong. (Just my humble opinion of course ;-)
Look at the potential fault matrix and you will notice that DRBD
provides _higher_ reliability than a shared storage system without full
redundancy - ie, without RAID, redundant controllers with replicated
caches, independent channels to each node etc; to compare 1:1, you'd
need replicated external storage, ie EMC RDF.
DRBD has no single point of failure at all; though running it on local
RAID1 is recommended IMHO but not absolutely required. DRBD can have the
two nodes quite a bit apart, easily in separate rooms.
Now, do the cost comparison.
Sure, DRBD is not the perfect solution for all scenarios where an EMC
with RDF is appropriate, but for quite a few "lower end" systems, it is
adequate. ("lower end" being defined roughly as "80MB/s write
performance is acceptable, storage <= half a TB, two nodes only"; the
best appropriate solution for a given customer is of course to be
determined by a trustworthy consultant ;-)
Where exactly do you see drbd fail?
Sincerely,
Lars Marowsky-Brée <lmb@suse.de>
-- High Availability & Clustering ever tried. ever failed. no matter. SuSE Labs try again. fail again. fail better. Research & Development, SuSE Linux AG -- Samuel Beckett --------------------------------------------------------------------- To unsubscribe, e-mail: suse-oracle-unsubscribe@suse.com For additional commands, e-mail: suse-oracle-help@suse.com Please see http://www.suse.com/oracle/ before posting
This archive was generated by hypermail 2.1.7 : Thu Sep 18 2003 - 10:28:34 CEST