[Linux-cluster] R: R: R: Gfs2 Problem

Marcello Percoco - Diennea marcello.percoco at diennea.com
Tue Sep 28 11:59:22 UTC 2010


Very wel, now i make a deep analysis, but i think thath the problemi s simply the conteention, so only our developers can solve the problem.
Thanks.

-- 
Marcello Percoco
IT Junior

Diennea
Viale G. Marconi, 30/14
48018 Faenza (RA) - Italy

E-Mail: marcello.percoco at diennea.com
Tel.: (+39) 0546 667432 - Int. 916
Fax:  (+39) 0546 399913

MagNews - E-Mail Marketing Solutions
http://www.magnews.it

Diennea - Technology for Marketing
http://www.diennea.com 

DISCLAIMER
Questo messaggio e i suoi allegati si rivolgono esclusivamente ai destinatari e possono contenere informazioni personali, confidenziali o protette da diritti. Se ha ricevuto questo messaggio per errore, l'utilizzo dei suoi contenuti è proibito e può esporre a conseguenze penali o civili. La invitiamo pertanto a rispedire immediatamente il messaggio al mittente e cancellarne gli allegati senza conservarne una copia. Per ulteriori informazioni, La preghiamo di contattarci all'indirizzo postmaster at diennea.com. Grazie 
This e-mail and any attachments may be confidential and the subject of legal professional privilege. Any disclosure, use, storage or copying of this e-mail without the consent of the sender is strictly prohibited. Please notify the sender immediately if you are not the intended recipient and then delete the e-mail from your inbox and do not disclose the contents to another person, use, copy or store the information in any medium. For further information write to postmaster at diennea.com. Thanks

-----Messaggio originale-----
Da: linux-cluster-bounces at redhat.com [mailto:linux-cluster-bounces at redhat.com] Per conto di Steven Whitehouse
Inviato: martedì 28 settembre 2010 13:35
A: linux clustering
Oggetto: Re: [Linux-cluster] R: R: Gfs2 Problem

Hi,

On Tue, 2010-09-28 at 13:15 +0200, Marcello Percoco - Diennea wrote:
> G:  s:UN n:2/2b1f92 f:l t:SH d:EX/0 l:0 a:0 r:4
>  H: s:SH f:aW e:0 p:10347 [java] gfs2_getattr+0x7d/0xc4 [gfs2] 

The entries you need to look for are those like the above where there is
a 'W' in the holder's flags field. If you do two dumps a few mins apart
and the lock state of the locks with waiters looks identical then there
is a real bug. If the lock state is different, or if different locks
have waiters, then that suggests that the issue is contention,

Steve.


--
Linux-cluster mailing list
Linux-cluster at redhat.com
https://www.redhat.com/mailman/listinfo/linux-cluster

Nessun virus nel messaggio in arrivo.
Controllato da AVG - www.avg.com 
Versione: 9.0.856 / Database dei virus: 271.1.1/3163 -  Data di rilascio: 09/27/10 19:56:00




More information about the Linux-cluster mailing list