My server has 8GB of ram, around 4 GB is currently used by spamassassin (too
many process of /usr/bin/perl -T /usr/local/psa/admin/sbin/spammng -c -C
--max-children=1 start).
Is this normal? Can I somehow limit the process concurrency of spamassassin
or I could limit the amount of RAM it consumes?
Quoting Jai Gupta <[EMAIL PROTECTED]>:
My server has 8GB of ram, around 4 GB is currently used by spamassassin (too
many process of /usr/bin/perl -T /usr/local/psa/admin/sbin/spammng -c -C
--max-children=1 start).
Is this normal? Can I somehow limit the process concurrency of spamassassin
or I
A downgrade to SpamAssassin 3.2.3 returns functionality with per-user
settings immediately. Any notes on what handle_user does and whether this is
a new function of 3.2.4?
I'd like to find out more about handle_user as it seems to be specific to
3.2.4
Thanks
-Original Message-
From: Matt
Thanks Jeff,
Some stats could be seen on
http://stats.jaigupta.com/com/jaigupta.com-memory.html and
http://stats.jaigupta.com/com/jaigupta.com-processes.html
spamassassin is using RAM with an linearly increasing rate, if it is not
stopped soon my server will end with its all RAM used by spamassas
Have seen a bunch more Google "btnI" spam the past couple days, seem to be
triggering just fine on the rules posted elsewhere on this thread.
However, these last bunch seem to have a trick, the only other text in the
message aside from the URL seems to be a date string. Somehow that must
totall
Jai Gupta wrote:
> My server has 8GB of ram, around 4 GB is currently used by spamassassin (too
> many process of /usr/bin/perl -T /usr/local/psa/admin/sbin/spammng -c -C
> --max-children=1 start).
"spammng" is not SpamAssassin. Tell whatever is spawning this process
to limit the number of these
Matthew Goodman wrote:
A downgrade to SpamAssassin 3.2.3 returns functionality with per-user
settings immediately. Any notes on what handle_user does and whether this is
a new function of 3.2.4?
It's not new. In fact, it was introduced into spamd somewhere between SA
1.3 (October 2001) and SA
Well that certainly does address the origin of the handle_user functions. I
reviewed the link you sent regarding the previous bug but I have to be
honest when I say it doesn't shed any light on the situation whatsoever.
The last guy that had the problem on the mailing list stopped posting after
y
On Sun, 13 Jan 2008, Mike Cisar wrote:
>However, these last bunch seem to have a trick, the only other text in the
>message aside from the URL seems to be a date string. Somehow that must
>totally be screwing with Bayes since those messages are also triggering
>BAYES_00 or BAYES_02 and pretty mu
A bug report would seem like a reasonable thing to me.
Loren
- Original Message -
From: "Matthew Goodman" <[EMAIL PROTECTED]>
To:
Sent: Sunday, January 13, 2008 7:40 PM
Subject: RE: Problem with handle_user
Well that certainly does address the origin of the handle_user functi
10 matches
Mail list logo