Re: Interesting RMAN induced "hang" behavior
From: Robert Klemme <shortcutter_at_googlemail.com>
Date: Sat, 28 Nov 2009 17:22:08 +0100
Message-ID: <7nd11gF3lg6d5U1_at_mid.individual.net>
On 11/27/2009 05:59 PM, Pat wrote:
> The question I've got for the group though is, has anybody seen
> anything like this before? I feel like I've got to be missing
> something here; I mean I used to use RMAN to backup to *tape* for
> heaven's sake which is slower than even a highly loaded XIV, and I
> never saw this problem.
>
> What am I missing here? Does anybody have a good theory on why a slow
> backup target would cause these sorts of symptoms?
Date: Sat, 28 Nov 2009 17:22:08 +0100
Message-ID: <7nd11gF3lg6d5U1_at_mid.individual.net>
On 11/27/2009 05:59 PM, Pat wrote:
> The question I've got for the group though is, has anybody seen
> anything like this before? I feel like I've got to be missing
> something here; I mean I used to use RMAN to backup to *tape* for
> heaven's sake which is slower than even a highly loaded XIV, and I
> never saw this problem.
>
> What am I missing here? Does anybody have a good theory on why a slow
> backup target would cause these sorts of symptoms?
RMAN likely deals with IO to tapes differently than to disks. When you write to a tape you can be pretty sure that it will be slower than disks. Hence RMAN might take precautions and not read DB blocks as fast while in the case of disks the assumption might be that an idle backup disk device will be faster than the main disks and RMAN just reads DB blocks as fast as possible. If the backup disk is in fact slower and there is significant traffic on the DB the effect you are describing might show up. This is just a guess of course.
Kind regards
robert
-- remember.guy do |as, often| as.you_can - without end http://blog.rubybestpractices.com/Received on Sat Nov 28 2009 - 10:22:08 CST