Coda File System

Venus crash

From: Stephen J. Turnbull <turnbull_at_sk.tsukuba.ac.jp>
Date: Mon, 24 Apr 2000 16:50:10 +0900 (JST)
On trying to du a moderately large hierarchy of files (vice and venus
on some machine), I got

venus: rvm_logflush.c:92: make_pad_buf: Assertion `(length >= 0) && (length < 512)' failed.
14:07:16 Fatal Signal (6); pid 20655 becoming a zombie...
14:07:16 You may use gdb to attach to 20655

Unfortunately I don't have the history of my actions there (I killed
the wrong xterm, it seems, and the shell history got overwritten), but
I believe that I successfully CVS checkout'd the module "xemacs" from
cvs.xemacs.org into /coda/Projects/XEmacs, then mv'd xemacs to
xemacs-21.2-HEAD (I plan to have several different XEmacs trees
there).  Then I tried to du xemacs-21.2-HEAD and it couldn't find it
or something (memory is fuzzy here), and then venus crashed.  Other
operations were interspersed.  /coda is the mount point for the root
volume, of course, and /coda/Projects/XEmacs is a separate volume.
The build configuration is default, and the server and venus are both
given default configurations with all data in ordinary Unix files for
testing.  /coda was mounted on at least one other machine at the
time.  All of the volumes were created with default args to
createvol_rep.

I also needed to get at some of the files on the Coda fs, so I killed
the process and restarted Venus.

This is 5.3.6, built on April 14, and probably checked out from CVS
April 13 at about 6:00 am EST (ie, Pittsburgh time) if I can trust my
directory listing.

-- 
University of Tsukuba                Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Institute of Policy and Planning Sciences       Tel/fax: +81 (298) 53-5091
_________________  _________________  _________________  _________________
What are those straight lines for?  "XEmacs rules."
Received on 2000-04-24 03:51:39