You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@spamassassin.apache.org by JOW <jg...@notes.wustl.edu> on 2007/09/28 23:06:52 UTC
Odd Memory problem: SA NOT using available memory
I'm out of my element on this, so please forgive me if I ask something silly.
We have two SpamAssassin (v3.002003) servers on Red Hat, each with 4gb RAM.
ServerA ROCKS while ServerB SUCKS.
So I've been comparing the two systems hoping to find some configuration
differences between them.
When I do a vmstat, I see a couple of things that are glaringly different:
ServerA
procs memory swap io system
cpu
r b swpd free buff cache si so bi bo in cs us sy id
wa
0 0 0 354012 154360 3395248 0 0 1 6 6 1 1 2 4
8
ServerB
procs memory swap io system
cpu
r b swpd free buff cache si so bi bo in cs us sy id
wa
1 0 0 3831400 73308 148168 0 0 3 184 108 325 18 3 68
11
ServerA (the good one) is using nearly all of the 4gb of RAM available to
it.
But ServerB is only using a fraction of available RAM and the # of context
switches is tons higher, too.
I hope this isn't a silly question, but how significant is this, and where
do I go from here?
Thanks for any help,
Joe
--
View this message in context: http://www.nabble.com/Odd-Memory-problem%3A-SA-NOT-using-available-memory-tf4536780.html#a12948517
Sent from the SpamAssassin - Users mailing list archive at Nabble.com.
Re: Odd Memory problem: SA NOT using available memory
Posted by Loren Wilton <lw...@earthlink.net>.
> So I shouldn't read anything into the disparity of memory usage between
> the
> two boxes?
Check the number of SA children configured. Maybe you have the bad box
limited to 1 or 2 children or some such.
Loren
>
> I'm at a loss as to why the one box is consistently in distress, while the
> other is consistently doing great.
>
> I should have mentioned we're using spamd (-m 50) and I've compared the
> main.cf's and found no differences. Seems like there must be something
> I'm
> missing though for these servers to behave so differently.
>
> Thanks again for the quick response, and any additional help,
> Joe
> --
> View this message in context:
> http://www.nabble.com/Odd-Memory-problem%3A-SA-NOT-using-available-memory-tf4536780.html#a12948922
> Sent from the SpamAssassin - Users mailing list archive at Nabble.com.
Re: Odd Memory problem: SA NOT using available memory
Posted by JOW <jg...@notes.wustl.edu>.
Thanks for the quick reply.
So I shouldn't read anything into the disparity of memory usage between the
two boxes?
I'm at a loss as to why the one box is consistently in distress, while the
other is consistently doing great.
I should have mentioned we're using spamd (-m 50) and I've compared the
main.cf's and found no differences. Seems like there must be something I'm
missing though for these servers to behave so differently.
Thanks again for the quick response, and any additional help,
Joe
--
View this message in context: http://www.nabble.com/Odd-Memory-problem%3A-SA-NOT-using-available-memory-tf4536780.html#a12948922
Sent from the SpamAssassin - Users mailing list archive at Nabble.com.
Re: Odd Memory problem: SA NOT using available memory
Posted by Matus UHLAR - fantomas <uh...@fantomas.sk>.
On 28.09.07 14:06, JOW wrote:
> I'm out of my element on this, so please forgive me if I ask something silly.
>
> We have two SpamAssassin (v3.002003) servers on Red Hat, each with 4gb RAM.
> ServerA ROCKS while ServerB SUCKS.
> ServerA (the good one) is using nearly all of the 4gb of RAM available to
> it. But ServerB is only using a fraction of available RAM and the # of
> context switches is tons higher, too.
that is not a thing SA controls. The usage of memory depends on OS.
SpamAssassin uses as much memory as it needs, unless there's memory leak in
it or in libraries SA uses. But in such case, that memory is wasted.
--
Matus UHLAR - fantomas, uhlar@fantomas.sk ; http://www.fantomas.sk/
Warning: I wish NOT to receive e-mail advertising to this address.
Varovanie: na tuto adresu chcem NEDOSTAVAT akukolvek reklamnu postu.
Depression is merely anger without enthusiasm.
Re: Odd Memory problem: SA NOT using available memory
Posted by JOW <jg...@notes.wustl.edu>.
>From a uname -v, they both appear to be #1 SMP Fri Oct 13 17:56:20 EDT 2006.
Here's a uname -a on each:
ServerB (B as in bad):
Linux expurgate2 2.4.21-47.0.1.ELsmp #1 SMP Fri Oct 13 17:56:20 EDT 2006
i686 i686 i386 GNU/Linux
ServerA (A as in awesome):
Linux expurgate1 2.4.21-47.0.1.ELsmp #1 SMP Fri Oct 13 17:56:20 EDT 2006
i686 i686 i386 GNU/Linux
I asked one of our Unix admins that very question, but he unfortunately is
not the one who built the servers, and didn't know. Is there any way to
tell now if they were compiled with different memory config options, or does
the fact that they have the same kernel version eliminate that possibility?
Thanks John
--
View this message in context: http://www.nabble.com/Odd-Memory-problem%3A-SA-NOT-using-available-memory-tf4536780.html#a12948990
Sent from the SpamAssassin - Users mailing list archive at Nabble.com.
Re: Odd Memory problem: SA NOT using available memory
Posted by "John D. Hardin" <jh...@impsec.org>.
On Fri, 28 Sep 2007, JOW wrote:
> ServerA (the good one) is using nearly all of the 4gb of RAM
> available to it. But ServerB is only using a fraction of available
> RAM and the # of context switches is tons higher, too.
>
> I hope this isn't a silly question, but how significant is this,
> and where do I go from here?
Are they both running the same kernel? The kernel on server B may have
been compiled with different memory config options thatn the one on
server A.
--
John Hardin KA7OHZ http://www.impsec.org/~jhardin/
jhardin@impsec.org FALaholic #11174 pgpk -a jhardin@impsec.org
key: 0xB8732E79 -- 2D8C 34F4 6411 F507 136C AF76 D822 E6E6 B873 2E79
-----------------------------------------------------------------------
Pelley: Will you pledge not to test a nuclear weapon?
Ahmadeinejad: CIA! Secret prison in Europe! Abu Ghraib!
-- Mahmoud Ahmadeinejad clumsily dodges a question
(60 minutes interview, 9/20/2007)
-----------------------------------------------------------------------
240 days until the Mars Phoenix lander arrives at Mars