hate to bring this up again.. but i tried the followin..

1. current cvs code + patch Harald fixed up
2. current cvs code - patch Harald fixed up

Both give the same problem I was having before.. Here's the rundown..

1. I goto the 'view other subscriptions' page with correct password an NO lock file (no lock/davis.lock).
2. the error comes right up with the following traceback.

Traceback (innermost last): File "/home/mailman/scripts/driver", line 89, in run_main main() File "/home/mailman/Mailman/Cgi/handle_opts.py", line 80, in main mlist.Save() File "/home/mailman/Mailman/MailList.py", line 867, in Save self.__lock.refresh() File "/home/mailman/Mailman/LockFile.py", line 204, in refresh raise NotLockedError NotLockedError:

&

<< -- start lock log -- >> Jun 01 00:03:58 2000 (6084) davis.lock laying claim Jun 01 00:03:58 2000 (6084) davis.lock got the lock Jun 01 00:03:58 2000 (6084) davis.lock laying claim Jun 01 00:03:58 2000 (6084) davis.lock already locked << -- stop lock log -- >>

3. Well.. I hit refresh and there is now a davis.lock file. And it waits (probably until

Jun 01 00:04:24 2000 (6095) davis.lock laying claim Jun 01 00:04:24 2000 (6095) davis.lock unexpected linkcount <> 2: 1 Jun 01 00:04:24 2000 (6095) davis.lock waiting for claim Jun 01 00:04:24 2000 (6095) davis.lock unexpected linkcount <> 2: 1 Jun 01 00:04:25 2000 (6095) davis.lock unexpected linkcount <> 2: 1

<snip> (you get the idea..)

Jun 01 00:04:51 2000 (6095) davis.lock unexpected linkcount <> 2: 1 Jun 01 00:04:53 2000 (6095) davis.lock unexpected linkcount <> 2: 1 Jun 01 00:06:05 2000 (6095) davis.lock unexpected linkcount <> 2: 1 Jun 01 00:06:05 2000 (6095) davis.lock waiting for claim Jun 01 00:06:06 2000 (6095) davis.lock unexpected linkcount <> 2: 1 Jun 01 00:06:06 2000 (6095) davis.lock unexpected linkcount <> 2: 1 Jun 01 00:06:08 2000 (6095) davis.lock unexpected linkcount <> 2: 1 Jun 01 00:06:08 2000 (6095) davis.lock lifetime has expired, breaking Jun 01 00:06:08 2000 (6095) davis.lock got the lock Jun 01 00:06:08 2000 (6095) davis.lock laying claim Jun 01 00:06:08 2000 (6095) davis.lock already locked

The other weird thing is i'm using the same Lockfile.py as the one you edited. at least diff didnt see any differences.

PLUS i can view other subscriptions on python.org. Is python.org using current CVS code from sourceforge?

PS. I'm willin to be a guinie pig again :)

-- Matt Davis - ICQ# 934680 http://dogpound.vnet.net/

The severity of the itch is proportional to the reach.