Coda File System

Re: Unable to write-reconnect

From: Johannes Martin <jmartin_at_notamusica.com>
Date: Tue, 1 Jun 2004 18:45:45 +0200 (CEST)
Hi,

thanks, Ivan, for your help:

On Mon, 31 May 2004, Ivan Popov wrote:

> > Right now, cfs lv does not report any conflicts and ctokens shows that I
> > do have a valid token. Still, I can't get out of the write-disconnected
> > mode.
>
> the reason can be a "hidden" conflict which can not be seen while
> the corresponding directory or volume is being used (like your shell having
> current directory on it).
>
> Try to remove all processes which could keep the volume referenced.
That helped partly - venus crashed...

--------------
06:19:37 volume coda:home:jmartin has unrepaired local subtree(s), skip checkpointing CML!

06:28:39 Propagating updates to volume coda:home:jmartin (|CML| = 43)
06:28:52 fatal error -- fsobj::dir_Create: (faviconrc, 5a728fc8.7f000001.fffffffe.801ff) Create failed!
06:28:54 RecovTerminate: dirty shutdown (1 uncommitted transactions)
Assertion failed: 0, file "fso_dir.cc", line 96
Sleeping forever.  You may use gdb to attach to process 895.06:30:54
RecovTerminate: dirty shutdown (1 uncommitted transactions)
--------------

I was able to restart, but that didn't get me write-reconnected either:
--------------
Date: Tue 06/01/2004

06:31:09 Coda Venus, version 6.0.6

06:31:09 /var/lib/coda/LOG size is 53252096 bytes
06:31:09 /var/lib/coda/DATA size is 213003776 bytes
06:31:09 Loading RVM data
06:31:10 Last init was Wed May 19 06:16:15 2004
06:31:10 Last shutdown was dirty
06:31:10 Starting RealmDB scan
06:31:10 	Found 2 realms
06:31:10 starting VDB scan
06:31:10 	4 volume replicas
06:31:10 	2 replicated volumes
06:31:10 	51 CML entries allocated
06:31:10 	0 CML entries on free-list
06:31:10 starting FSDB scan (83333, 2000000) (25, 75, 4)
06:31:10 	(.xsession-errors, 5a728fc8.7f000001.1c9c.1e96) found owrite object, discarding
06:31:11 	(lost file data backed up to /var/lib/coda/spool/.xsession-errors-6)
06:31:11 	640 cache files in table (55801 blocks)
06:31:24 	82693 cache files on free-list
06:31:25 starting HDB scan
06:31:25 	0 hdb entries in table
06:31:25 	0 hdb entries on free-list
06:31:25 Mounting root volume...
06:31:25 Venus starting...
06:31:25 /coda now mounted.
---------------

> (it can be as well something else or even a yet-to-uncover venus bug... :)
Maybe. Since I was pretty sure I hadn't modified anything important, I did
a venus -init...

	Johannes
Received on 2004-06-01 12:48:14