[Ocfs-users] Follow up on async I/O question

PENON Christian cpenon at ARGUSAUTO.com
Thu Mar 25 09:16:54 CST 2004

Hi everybody,

We have been using OCFS for one year now (many TARs opened, many patches
applied...) but we never worried about direct IO or async IO.
I assume that we use async IO (default behaviour for our system).
What should we look at ?

We use RedHat AS 2.1 (kernel errata 35) Ocfs 1.0.9-12 Oracle

Thank you for your help.


-----Message d'origine-----
De : Derek Suzuki [mailto:DSuzuki at ziprealty.com]
Envoyé : jeudi 25 mars 2004 02:54
À : 'Wim Coekaerts'
Cc : 'ocfs-users at oss.oracle.com'
Objet : RE: [Ocfs-users] Follow up on async I/O question

	Okay, thanks for the update.  We'll probably stay in sync mode for
now because the system is still way faster than our old platform.  I'm
setting up a test cluster that we can use to try out the newer OCFS releases
and perform some testing with async mode.

> -----Original Message-----
> From: Wim Coekaerts [mailto:wim.coekaerts at oracle.com]
> Sent: Wednesday, March 24, 2004 5:46 PM
> To: Derek Suzuki
> Cc: 'ocfs-users at oss.oracle.com'
> Subject: Re: [Ocfs-users] Follow up on async I/O question
> no that hasn't been solved yet.
> I am not sure why we ar still seeing some problems, actually, haven't
> had time to look at some of the later reported issues. normally the
> 1.0.9-x and 1.0.10/11 should work fine. as long as the 
> redologilfes are
> created in 1 chunk. there is absolutely no reason why anything else
> would fail. but I still have to double check.
Ocfs-users mailing list
Ocfs-users at oss.oracle.com

More information about the Ocfs-users mailing list