[nfsv4] Re: stable storage for server restart

Next message: Spencer Shepler: "[nfsv4] Re: stable storage for server restart"
Previous message: Internet-Drafts@ietf.org: "[nfsv4] I-D ACTION:draft-ietf-nfsv4-acl-mapping-03.txt"
In reply to: rick@snowhite.cis.uoguelph.ca: "[nfsv4] stable storage for server restart"

spencer.shepler@sun.com

Date: Tue, 22 Feb 2005 00:28:15 -0800
From: Spencer Shepler <spencer.shepler@sun.com>
Message-ID: <20050222082815.GE130156@jurassic.eng.sun.com>
Subject: [nfsv4] Re: stable storage for server restart

On Sat, rick@snowhite.cis.uoguelph.ca wrote:
<...>
> I think the weakest part of this design is that, if the server crashes
> again while at "*", the append only log is not complete (possibly empty).
> This will result in clients not being allowed to reclaim, that otherwise
> should be able to (ie. no entry->no reclaim->NFS4ERR_NOGRACE for all reclaim
> requests).
> 
> The append log will grow, but I only see a problem if clients go hogwild
> with SetClientIDs. It does get truncated when the server restarts, so a
> sysadmin can just reboot when it gets too big:-)

The Solaris client will SETCLIENTID with a "fresh" server.
This will happen often if the client unmounts everything from
a server then remounts (new SETCLIENTID dance).  The use of the
automounter may trigger this behavior often; note that automounter
timeouts are on the order of 5/10 minutes.

Spencer

_______________________________________________
nfsv4 mailing list
nfsv4@ietf.org
https://www1.ietf.org/mailman/listinfo/nfsv4