gluster-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Gluster-devel] GlusterFS 1.3.0-pre2.2: AFR setup


From: Krishna Srinivas
Subject: Re: [Gluster-devel] GlusterFS 1.3.0-pre2.2: AFR setup
Date: Mon, 5 Mar 2007 01:32:51 +0530

Hi Gerry,
Race condition with write() is OK because that is allowed even
on the regular file systems (ext3 etc) which is why flock is used.
But if create() is called simultaneously from server1 and server2
it will be a problem. To fix it I need to dedicate a lock server in
AFR and use UNIFY over AFR. And we can use the same lock
server for AFR on all the clients. That will fix the problem, I will make
this change.
Krishna

On 3/5/07, Gerry Reno <address@hidden> wrote:
Krishna Srinivas wrote:
> But this setup will be vulnerable to race conditions, for example
> if server1 and server2 try to create file with same name in the
> same path.
>
Yes, does GlusterFS have a way to arbitrate that?  In other words it
would be ok as long as they both don't try to write the file at the same
time.  One write would have to complete before the other began.

Regards,
Gerry






reply via email to

[Prev in Thread] Current Thread [Next in Thread]