diff options
| author | bwarsaw | 2001-10-09 21:42:58 +0000 |
|---|---|---|
| committer | bwarsaw | 2001-10-09 21:42:58 +0000 |
| commit | f3c90d2f5e6346857ab76968d0cf18b6026ba8e9 (patch) | |
| tree | 1a4863d9a9040e85e2dabc3cb667ee77e1056a35 | |
| parent | f1c6bf4adc953e0355c7c8f0dc156f3368008a0a (diff) | |
| download | mailman-f3c90d2f5e6346857ab76968d0cf18b6026ba8e9.tar.gz mailman-f3c90d2f5e6346857ab76968d0cf18b6026ba8e9.tar.zst mailman-f3c90d2f5e6346857ab76968d0cf18b6026ba8e9.zip | |
process(): Added a big ol' comment describing a potential bogosity in
handling "delayed" DSN reports.
| -rw-r--r-- | Mailman/Bouncers/DSN.py | 7 |
1 files changed, 5 insertions, 2 deletions
diff --git a/Mailman/Bouncers/DSN.py b/Mailman/Bouncers/DSN.py index 186a94ae3..672b1a83f 100644 --- a/Mailman/Bouncers/DSN.py +++ b/Mailman/Bouncers/DSN.py @@ -39,8 +39,11 @@ def check(msg): # # Also grok out Action so we can do something with that too. action = msgblock.get('action', '') - # Should we treat delayed bounces the same? Yes, because if the - # transient problem clears up, they should get unbounced. + # BAW: Should we treat delayed bounces the same? Yes, because if + # the transient problem clears up, they should get unbounced. The + # other problem is what to do about a DSN that has both delayed + # and failed actions in multiple header blocks? We're not + # architected to handle that. ;/ if action.lower() not in ('failed', 'failure', 'delayed'): # Some non-permanent failure, so ignore this block continue |