Professional Software Engineering wrote
on Monday, February 09, 2009 18:34:
At 09:01 2009-02-09 -0700, LuKreme wrote:
You do not need a lock on a /dev/null delivery.
Agreed.
However, that is not what caused a lockfile warning message
for the OP.
To wit:
7:49pm [~/Mail] 676[0]> cat rc
:0:
/dev/null
7:23pm [~/Mail] 677[0]> procmail -m LOGABSTRACT=y rc < $SPAMPLE
From 8florenciae(_at_)cpdns(_dot_)net Sat Nov 15 13:38:07 2008
Subject: to dman
Folder: /dev/null
<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
Not only that, but procmail's code is smart enough to understand
that /dev/null doesn't need a lockfile. It simply operates
without applying one in this case. Using an iteration of 1000 or
so and timing the operation, one can see that the lockfile colon
on a recipe whose action is to send the message to /dev/null does
not slow things down one whit.
Lockfile warning messages in the log likewise don't even show up
for directory saves, though they are also unnecessary.
7:51pm [~/Mail] 679[0]> cat rc
:0:
/var/tmp/
7:51pm [~/Mail] 680[0]> procmail -m LOGABSTRACT=y rc < $SPAMPLE
From 8florenciae(_at_)cpdns(_dot_)net Sat Nov 15 13:38:07 2008
Subject: to dman
Folder: /var/tmp/new/1234205477.9701_1.panix5.panix.com 1690
<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
You get the warning messages by using a lockfile
on a nested recipe, though:
7:52pm [~/Mail] 685[0]> cat rc
:0:
{ KILROY = "was here" }
HOST
7:52pm [~/Mail] 686[0]> procmail -m LOGABSTRACT=y rc < $SPAMPLE
procmail: Extraneous locallockfile ignored
[^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^]
From 8florenciae(_at_)cpdns(_dot_)net Sat Nov 15 13:38:07 2008
Subject: to dman
Folder: 0
<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
Dallman
____________________________________________________________
procmail mailing list Procmail homepage: http://www.procmail.org/
procmail(_at_)lists(_dot_)RWTH-Aachen(_dot_)DE
http://MailMan.RWTH-Aachen.DE/mailman/listinfo/procmail