I had another crash yesterday which was different. Appears to be an LDAP issue (I use was
using EUDC to look up an address in message-mode). I hadn't seen that before, or at
least I hadn't noticed it. C backtrace (I unfortunately forgot about the lbt command
yesterday) is in xemacs-ldap-crash.txt.
Today I had another one of the bytecode crashes, again while running under gdb (and I
didn't set gnus-use-bytecode to nil) the gdb backtrace is in
xemacs-bytecode-crash.txt.
Looking at the instructions below I see that the numbers have changed. Based on where
they are in the previous trace it looks like I wanted to try 140737322320744 and
0x7fffffffb738. Unfortunately I get a syntax error when I try to run ldp. Here's the
output of my attempts to ldp and print those addresses and the output of lbt.
(gdb) source /home/dsg/xemacs-21.4.22/src/.gdbinit
Really redefine built-in command "ptype"? (y or n) [answered Y; input not from
terminal]
(gdb) ldp (Lisp_Object) 140737322320744
Lisp => A syntax error in expression, near `)'.
(gdb) ldp * (Lisp_Object) 0x7fffffffb738
Lisp => A syntax error in expression, near `)'.
(gdb) print (Lisp_Object) 140737322320744
$1 = 140737322320744
(gdb) print * (Lisp_Object) 0x7fffffffb738
$2 = -166092088
(gdb) print * (Lisp_Object *) 0x7fffffffb738
$3 = 140737322263240
(gdb) lbt
# bind (codepoint)
mm-ucs-to-char(160)
mm-url-decode-entities()
# bind (mm-url-html-entities)
mm-url-decode-entities-nbsp()
# (unwind-protect ...)
# (unwind-protect ...)
# bind (temp-buffer string)
nnrss-decode-entities-string("On Wednesday evening, February 25th, from 08:00 PM
until Thursday morning, February 26th, at 06:00 AM (ET), maintenance will be performed on
MeetingPlace, affecting MeetingPlace Audio and Web Conferencing services.  If you
have questions or concerns, please contact Jason Rogers at 3-2477.")
# bind (--dolist-temp--19251 item file xml subject url extra changed author date
feed-subject enclosure comments rss-ns rdf-ns content-ns dc-ns hash-index server group)
nnrss-check-group("Infrastructure Status - Planned" "")
# bind (dont-check server group)
nnrss-request-group("Infrastructure Status - Planned" "" nil)
# bind (gnus-command-method group dont-check gnus-command-method)
byte-code("..." [method dont-check gname group gnus-command-method
gnus-find-method-for-group gnus-server-to-method gnus-get-function request-group
string-match "^[^:]+:" 0 nil 1] 5)
# (condition-case ... . ((quit (message "Quit activating %s" group) nil)))
# bind (method active method dont-check scan group)
gnus-activate-group("nnrss:Infrastructure Status - Planned" scan)
# bind (method-type cmethod retrieve-groups method active group info scanned-methods
type-cache methods-cache foreign-level alevel newsrc level)
gnus-get-unread-articles(nil)
# bind (gnus-read-active-file init nnmail-fetched-sources dont-connect level rawfile)
gnus-setup-news(nil nil nil)
byte-code("..." [gnus-slave gnus-current-startup-file gnus-startup-file
did-connect level dont-connect gnus-group-quit gnus-run-hooks gnus-startup-hook
gnus-make-newsrc-file gnus-dribble-read-file gnus-request-create-group "queue"
(nndraft "") "drafts" (nndraft "") gnus-setup-news nil
gnus-setup-news-hook gnus-start-draft-setup gnus-group-list-groups
gnus-group-first-unread-group gnus-configure-windows group gnus-group-set-mode-line
gnus-started-hook gnus-use-dribble-file gnus-agent] 4)
# bind (level did-connect slave dont-connect arg)
gnus-1(nil nil nil)
# bind (slave dont-connect arg)
gnus(nil)
# bind (command-debug-status)
call-interactively(gnus)
command-execute(gnus t)
# bind (_execute_command_keys_ _execute_command_name_ prefix-arg)
execute-extended-command(nil)
# bind (command-debug-status)
call-interactively(execute-extended-command)
# (condition-case ... . error)
# (catch top-level ...)
(gdb)
I still have this process open and can keep it so unless I have to reboot so I'll be
glad to try any other gdb checks you want.
Thanks,
--
Dave Goldberg
Associate Department Head, G06A: Advanced Technical Computing Center
The Mitre Corporation \ MS K331 \ 202 Burlington Rd. \ Bedford, MA 01730
dsg(a)mitre.org \ 781-271-3887
-----Original Message-----
From: Stephen J. Turnbull [mailto:stephen@xemacs.org]
Sent: Tuesday, February 17, 2009 11:04 PM
To: Goldberg, David S.
Cc: XEmacs Beta
Subject: Re: [Bug: 21.4.22] Sporadic crash entering a group with Gnus
David S. Goldberg writes:
> Could you get that code file under gdb again, do
> (gdb) source /path/to/source/of/xemacs/src/.gdbinit
> (gdb) ldp (Lisp_Object) 39408616
> (gdb) ldp * (Lisp_Object *) 0x7ffff8aec010
> and post the results here, please?
You can't do that without a process to debug.
Ah. That's right. Try the same thing with "print" instead of
"ldp".
Probably it should have been "print" in the first place.
I am happy to help track this down however I can.
Well, you could always run xemacs under gdb. For me there is no
noticable performance impact, but I often get much better stack
traces, and ldp (eval and print a Lisp expression) and lbt (Lisp back
trace) work because the process is still live, just suspended.
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta