On Tue, Jan 25, 2005 at 03:08:44PM -0500, Jim Maul wrote:
Martin Karol Zuziak wrote:
Hello list
Well, so far no replies to my post regarding "deep recursion" errors.
I have tested the message with 'spamassassin' (ie. not spamc/spamd) with the same result. The scanning took 21 CPU seconds on a Intel Pentium 3, 1133 MHz and consumed 180 MB memory.
Am I the only one seeing this problem? Could someone else test the message and report if you too are seeing excessive cpu and memory load.
The message is at http://www.math.ku.dk/~zuziak/sa/satrigger.txt .
[EMAIL PROTECTED] jmaul]$ time spamc -r < satrigger.txt
real 0m4.860s user 0m0.000s sys 0m0.000s
SA 2.64 on a P4 2.4ghz
Dont know how much memory it consumed during those 4.8 seconds but i seriously doubt it was 180mb. The machine only has 256mb and has about 50mb free after the scan which is about normal for this machine.
Interesting. Are you using razor, pyzor or dcc?
Maybe the problem only exists in 3.0. The errors I'm seeing seems to be generated by spamassassin itself, not an external module.
Martin Zuziak <[EMAIL PROTECTED]>
Im using razor, surbl and bayes. Pyzor and dcc are turned off.
Its also worth noting that i dont see any errors whatsoever during the scan.
-Jim
