Coda File System

Re: new coda issue: touch a file and coda dies

From: Jan Harkes <jaharkes_at_cs.cmu.edu>
Date: Thu, 7 Jul 2005 18:15:02 -0400
On Thu, Jul 07, 2005 at 01:47:09PM -0600, Patrick Walsh wrote:
> # cfs whereis /coda/director/snapin/pool_scm
>   dir224  dir225  dir225
> 
> 	A quick look at VRList on the server shows:
> 
> /snapin 7f000003 3 1000004 2000004 200000a 0 0 0 0 0 0

btw. recovering from this is actually surprisingly simple, edit the
VRList line to look like,

/snapin 7f000003 2 1000004 2000004 0 0 0 0 0 0 0

Then rebuild the VRDB (volutil makevrdb) and you'll end up with a doubly
replicated volume. Once you restart your client you can perform a
recursive ls and resolve/repair any differences between the replicas.

After that you could re-create a third replica on a new server, update
the VRList and populate it by re-resolving. The intermediate step isn't
really necessary, except that it is probably easier to first make sure
that all the data on the two replicas is consistent and accessible.

Jan
Received on 2005-07-07 18:17:53