2.1.11.rc2 crash
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
GNU Mailman |
New
|
Medium
|
Unassigned |
Bug Description
While I know this *looks* like an error caused by having no space...
[root@lists /usr/home/
Filesystem Size Used Avail Capacity Mounted on
/dev/idad0s1a 14G 6.9G 5.6G 55% /
devfs 1.0K 1.0K 0B 100% /dev
[root@lists /var/log/mailman]# cat error
Aug 22 01:00:00 lists newsyslog[66511]: logfile turned over
Aug 22 09:54:05 2008 (72778) No such list "ccm-l<br><br>":
Aug 22 17:56:06 2008 (60050) Uncaught runner exception: [Errno 28] No
space left on device
Aug 22 17:56:06 2008 (60050) Traceback (most recent call last):
File "/usr/home/
_oneloop
self.
File "/usr/home/
_onefile
keepqueued = self._dispose(
File "/usr/home/
130, in _dispose
self.
File "/usr/home/
138, in enqueue
fp.
IOError: [Errno 28] No space left on device
Aug 22 17:56:07 2008 (60050) Uncaught runner exception: [Errno 28] No
space left on device
Aug 22 17:56:07 2008 (60050) Traceback (most recent call last):
File "/usr/home/
_oneloop
new_filebase = self._shunt.
File "/usr/home/
138, in enqueue
fp.
IOError: [Errno 28] No space left on device
Aug 22 17:56:07 2008 (60050) SHUNTING FAILED, preserving original entry:
1219445382.
Aug 22 18:01:07 2008 (60050) Uncaught runner exception: [Errno 6] Device
not configured
Aug 22 18:01:07 2008 (60050) Traceback (most recent call last):
File "/usr/home/
_oneloop
self.
File "/usr/home/
_onefile
keepqueued = self._dispose(
File "/usr/home/
74, in _dispose
self.
File "/usr/home/
160, in process
deliveryfun
File "/usr/home/
392, in bulkdeliver
syslog(
msgid)
File "/usr/home/
in write
self.
File "/usr/home/
in write_ex
logf.write(msg + '\n')
File "/usr/home/
line 73, in write
Logger.
File "/usr/home/
in write
_logexc(self, msg)
File "/usr/home/
_logexc
sys.
IOError: [Errno 6] Device not configured
Aug 22 18:01:08 2008 (60050) SHUNTING:
1219445385.
Aug 22 18:57:19 2008 (60049) Uncaught runner exception: [Errno 28] No
space left on device
Aug 22 18:57:19 2008 (60049) Traceback (most recent call last):
File "/usr/home/
_oneloop
self.
File "/usr/home/
_onefile
keepqueued = self._dispose(
File "/usr/home/
38, in _dispose
return IncomingRunner.
File "/usr/home/
130, in _dispose
more = self._dopipelin
File "/usr/home/
153, in _dopipeline
sys.
File "/usr/home/
55, in process
outq.
File "/usr/home/
138, in enqueue
fp.
admin(81551):
[http://
Originator: NO
df notwithstanding, how is this a Mailman issue? Is it possible you're
running into a user quota?
In most of the above tracebacks, Mailman has successfully opened a
tempfile to build a queue entry, and is denied by the OS with "[Errno 28]
No space left on device". There's not much Mailman can do about that.
In the case of the logging error, Mailman is attempting to write to the
smtp-failure log and hits an IOError exception (possibly error 28) and
tries to fall back to the original stderr which is probably the console
that started Mailman and is long since detached from the process and gets
the "[Errno 6] Device not configured" and logs that exception to the error
log.
The OS is not letting Mailman write to some files (but apparently the
error log is OK). Mailman can't do anything about that.