[Mailman-Users] flock errors via cron
scott cotton
scott at chronis.pobox.com
Thu Feb 4 19:32:00 CET 1999
The underlying problem is this: the flock module has a
default timeout that when passed causes a lock to be
stolen. That default timeout is a good value for most
operations, but not for long running operations like
run_queue.
So the change is simply to give it a long enought time out
value. This can be done by alterning the file
Mailman/OutgoingQueue.py like so:
look for the function "processQueue". in the first few
lines there is something like:
lock_file = flock.FileLock(somepath)
this should be changed to have an extra argument indicating
a reasonable timeout for running the queue:
lock_file = flock.FileLock(somepath, 14400)
is an example.
scott
On Thu, Feb 04, 1999 at 12:00:28PM -0600, Christopher Lindsey wrote:
| Howdy all,
|
| Every time that $prefix/cron/run_queue is executed I get a
| mail message with errors:
|
| Traceback (innermost last):
| File "/home/staff/mailman/cron/run_queue", line 31, in ?
| OutgoingQueue.processQueue()
| File "/home/staff/mailman/Mailman/OutgoingQueue.py", line 86, in processQueue
| lock_file.lock()
| File "/home/staff/mailman/Mailman/flock.py", line 119, in lock
| os.link(self.lockfile, self.tmpfname)
| os.error: (2, 'No such file or directory')
|
| This is on a Linux 2.0.30 system with Python 1.5.1. The file system
| is local, but exported via NFS with version 2.2beta37 of Olaf
| Kirch's "Universal NFS server".
|
| I've never done any programming in Python, so I'm hoping that
| someone will have an immediate answer before I'm forced to immerse
| myself in the task of learning yet *another* language. :)
|
| Thanks,
|
| Chris (who longs for the days of low-level programming again)
|
| ------------------------------------------------------
| Mailman-Users maillist - Mailman-Users at python.org
| http://www.python.org/mailman/listinfo/mailman-users
|
More information about the Mailman-Users
mailing list