Package: enemies-of-carlotta
Version: 1.2.4-1
Severity: normal
This report is just a reminder of the problem I reported in summer at
http://liw.iki.fi/lists/[EMAIL PROTECTED]/msg00331.html .
There seem to be two race conditions here. In both cases, assume a
recipient that is permanently undeliverable.
- if the list gets at least one mail per week, then timestamp-bounced
will eternally be newer than a week, since it is updated at every
bounce
==> no probing for busy lists
- if the list gets a regular bounce between the sending of the probe
and the bouncing of the probe (it's not always even nearly
instantaneous), then the probe bounce gets ignored
==> probe gets ignored if it's a bad day
- I suppose that getting a regular bounce between having received the
probe bounce and the next cleaning woman run would also be
problematical
Perhaps ignoring regular bounces for recipients in the states bounced,
probed and probebounced would fix these?
-- System Information:
Debian Release: 4.0
APT prefers testing
APT policy: (500, 'testing')
Architecture: i386 (i686)
Shell: /bin/sh linked to /bin/bash
Kernel: Linux 2.4.26-bytemark-uml-20040706-1
Locale: LANG=fi_FI.UTF-8, LC_CTYPE=fi_FI.UTF-8 (charmap=UTF-8)
Versions of packages enemies-of-carlotta depends on:
ii procmail 3.22-16 Versatile e-mail processor
ii python 2.4.4-1 An interactive high-level object-o
ii python-support 0.5.6 automated rebuilding support for p
enemies-of-carlotta recommends no packages.
-- no debconf information
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
