Coda File System

Re: codasrv can't allocate RVM

From: Stephen J. Turnbull <turnbull_at_sk.tsukuba.ac.jp>
Date: Thu, 11 Oct 2001 11:15:59 +0900
>>>>> "me" == Stephen J Turnbull <turnbull_at_sk.tsukuba.ac.jp> writes:

>>>>> "Jan" == Jan Harkes <jaharkes_at_cs.cmu.edu> writes:

    Jan>     norton-reinit -rvm <rvmlog> <rvmdata> <newrvmdatalength> -load \
    Jan> 	ServerDump skip <volumeid> <volumeid>...

    me> Wups.  I got the following result with no skips, skip the
    me> first, skip the first and second.  Only the "skipping"
    me> messages and the identity of the volume needing salvage
    me> changes.  How do I salvage?

OK, I got salvaged.  Repeated cycles of codasrv stop and codasrv start
and redumping.  I captured the bash history, if you want I'll check
the exact sequence of what I did (I think I recall doing the
norton-reinit -dump directly after the server asserted, instead of
doing codasrv start ... stop ... norton-reinit -dump, and of course
the bogus state of the RVM log would be preserved in the dump until I
redumped, right?)

Now I've got a different problem.  The volume with the "unreachable
log entries" cannot be write-connected, there are CMLs from venus but
venus says:

09:54:19 volume xe:21-5 has unrepaired local subtree(s), skip checkpointing CML!

I cannot find any conflicts in that volume.  No symlinks, let alone
dangling ones.  What does this mean?

BTW, that volume contains only a "virgin" CVS checkout of XEmacs, so
purging the volume and recreating would be an acceptable method of
recovery.  It would be nice if this could be solved more directly,
though.


-- 
University of Tsukuba                Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Institute of Policy and Planning Sciences       Tel/fax: +81 (298) 53-5091
	  Don't ask how you can "do" free software business;
	  ask what your business can "do for" free software.
Received on 2001-10-10 22:16:40