From: martin f krafft <[EMAIL PROTECTED]>
> Have they cut out some text on you?
Urks. That was my fault. For some reason, the way I do procmail
logging cut out the lines, due to the leading space. This is the
full message:
ERROR: maillib.crm broke. Here's the error\:=20
ERROR:=20
/usr/bin/crm: *WARNING*=20
Tried to fork your minion, but it failed. Your system may have run out of =
proce
ss slots
I'll try to keep working.
This happened at line 1123 of file /usr/share/crm114/mailreaver.crm
So even though I cannot imagine that the system has run out of
process slots, the problem seems to be not with crm114. Sorry for
the false alarm.
Anyway, is there something crm114 could do, like retry? As it
stands, it just returns UNKNOWN/0, which is not ideal.
Yes, it's possible to catch it with a TRAP statement and retry.
Point; this is a perfect object example of why crm114 C lib routines should
never VERIFY or ASSERT but rather check and then use *error routines, so
that TRAP can catch problems.
Now, assuming your friends didn't edit mailreaver too heavily, your
fault above is during the creation of the reavercache - the directory
of messages that mailreaver uses for training; that's done by
executing the bash "mkdir" command. So, it's actually possible
(although strange except in very wierd circumstances) that you actually
did run out of process slots.
Is this on a very heavily loaded mailserver being gang-banged by
a lot of spammers, and launching a CRM114 on each and every incoming
mail?
- Bill Yerazunis
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
