Hi Steve, thanks for your answer
but I have not put the "localflocks" mount parameter anywhere. Look at
"gfs_tool df" output:
# gfs_tool df /mnt/gfs
/mnt/gfs:
SB lock proto = "lock_dlm"
SB lock table = "H-N:gfs01"
SB ondisk format = 1309
SB multihost format = 1401
Block size = 4096
Journals = 2
Resource Groups = 200
Mounted lock proto = "lock_dlm"
Mounted lock table = "H-N:gfs01"
Mounted host data = "jid=0:id=196610:first=1"
Journal number = 0
Lock module flags = 0
Local flocks = TRUE
Local caching = FALSE
Oopses OK = FALSE
it says 'Mounted lock proto = "lock_dlm" ' because that is what I did.
So why is it using "local flocks"?
Frank
Date: Wed, 23 Dec 2009 13:15:46 +0000 From: Steven Whitehouse
<swhiteho@xxxxxxxxxx> To: linux clustering <linux-cluster@xxxxxxxxxx>
Subject: Re: lock_dlm but local flocks = true?
Message-ID: <1261574146.2219.4.camel@localhost> Content-Type:
text/plain; charset="UTF-8" Hi, On Wed, 2009-12-23 at 13:24 +0100,
frank wrote:
> Hi,
> I'm running RHEL 5.4 with a GFS1 filesystem (over a SAN)
> After checking the "ping_pong" test on the gfs filesystem I noticed
> counters was too much high, like doing the test with a local filesystem.
> But gfs filesystem is mounted with "lock_dlm" option. The only strange
> thing I see is in "gfs_tool df" output because it shows "Local flocks =
> TRUE" instead of "Local flocks = FALSE"
>
> How is this possible? Can I change this in some way?
> Thanks for your help.
>
>
You need to avoid specifying the localflocks mount parameter if you want
the flocks to be true cluster locks,
Steve.
--
Aquest missatge ha estat analitzat per MailScanner
a la cerca de virus i d'altres continguts perillosos,
i es considera que està net.
For all your IT requirements visit: http://www.transtec.co.uk
--
Linux-cluster mailing list
Linux-cluster@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/linux-cluster
