On Fri, 2011-11-18 at 19:36 +0100, Karsten Bräckelmann wrote:
> On Fri, 2011-11-18 at 08:16 +, Tom wrote:
> > (apologies if the html doesn't end up translating well!)
Damn, sorry. My attempt at pruning the large tables seriously fucked up
the formatting. :/
> > output from top, after running
On Fri, 2011-11-18 at 08:16 +, Tom wrote:
> Here's the stats from my cluster at the moment (8am) (these figures wll
> ramp up considerably!) (apologies if the html doesn't end up
> translating well!)
>
> Server
> Load Avg
> Processed/Min
> Busy Child Proc
> Proc Time
> 10.44.219.192
> 0.34
> 4
On Thu, 2011-11-17 at 15:55 +, Tom wrote:
> SPAMDOPTIONS="-d -L -i 10.44.219.208 -A 10.44.217.0/20 -m 40 -q -x -u
> spamd --min-children=40"
Do you really run a single spamd server, serving a /20 of potential SMTP
servers?
Also, you configured spamd to try hard and always keep exactly 40
chi
have you turned off RBLs and other network tests you dont need and disabled
any non-standard rules and plugins?
if you are using RBLS's have a a caching nameserver on the SA machine
itself (even if your 'local' DNS server is only a couple of milliseconds
away a caching namesserver on the box itsel
Greetings,
=== Information ==
Old Version: spamassassin-3.2.5-1.el5
New EL5 Version: spamassassin-3.3.1-2.el6.x86_64
New EL6 Version: spamassassin-3.3.1-2.el5.x86_64
SPAMDOPTIONS="-d -L -i 10.44.219.208 -A 10.44.217.0/20 -m 40 -q -x -u
spamd --min-children=40"
Other info Bayes
On Mon, 2006-10-09 at 11:43 -0400, Rob McEwen (PowerView Systems) wrote:
> > The last few weeks I have noted (angry users calling me by phone) that
> > the server is really slow.
>
> Don't know for sure, but I suspect slower than usual Razor and/or DCC servers?
>
> --Rob McEwen
>
I second that.
->
-> I have requested more memory and an additional processor for the server,
-> hope they give me something...
->
-> I will be working with the Greylisting and validrcptto, and let you know
-> how it goes.
->
-> Thanks again,
->
-> BR,
-> Matias.
Welcome. :-)
Remember validrcptto is a qmail
Matias Lopez Bergero wrote:
> Bowie Bailey wrote:
> > Matias Lopez Bergero wrote:
> > [...]
> > > Looks to me that SpamAssassin is taking to long to process the
> > > incoming emails, and as result, it is slowing down the server, and
> > > finally causing the DoS.
> > [...]
> >
> > The message abo
Chris St. Pierre wrote:
> On Mon, 9 Oct 2006, Craig Baird wrote:
>
>
>>I have an old Redhat box that started doing this a while back. After a lot of
>>hair pulling, I finally figured out that the problem was related to spam
>>floods, but seemed to be caused by the syslog daemon.
>
>
> That's a
Bowie Bailey wrote:
> Matias Lopez Bergero wrote:
>[...]
>>Looks to me that SpamAssassin is taking to long to process the
>>incoming emails, and as result, it is slowing down the server, and
>>finally causing the DoS.
> [...]
>
> The message about increasing the child processes is frequently
> mis
On Mon, 9 Oct 2006, Matias Lopez Bergero wrote:
>Thanks Robert,
>
>I have requested more memory and an additional processor for the server,
>hope they give me something...
>
>I will be working with the Greylisting and validrcptto, and let you know
>how it goes.
>
>Thanks again,
>
>BR,
>Matias.
Yo
Email Lists wrote:
> -> [...]
> -> Looks to me that SpamAssassin is taking to long to process the
> -> incoming emails, and as result, it is slowing down the server, and
> -> finally causing the DoS.
> [...]
>
> The first thing to do is relieve some of the load. To do this we need better
> mail
At 01:09 PM 10.9.2006 -0500, David B Funk wrote:
>On Mon, 9 Oct 2006, Matias Lopez Bergero wrote:
>
>> Hello!
>>
>> I was very happy using SpamAssassin at my email server (Xeon 2.8GHz, 1.5
>> GB memory, Dual Ultra SCSI HD 73.4GB in RAID 1, Linux 2.4.33)
>>
>> There are only 2500 email boxes at the
On Mon, 9 Oct 2006, Matias Lopez Bergero wrote:
> Hello!
>
> I was very happy using SpamAssassin at my email server (Xeon 2.8GHz, 1.5
> GB memory, Dual Ultra SCSI HD 73.4GB in RAID 1, Linux 2.4.33)
>
> There are only 2500 email boxes at the server. The server is running:
> Sendmail, SpamAssassin 3
On Mon, 9 Oct 2006, Craig Baird wrote:
> I have an old Redhat box that started doing this a while back. After a lot of
> hair pulling, I finally figured out that the problem was related to spam
> floods, but seemed to be caused by the syslog daemon.
That's actually not a bad suggestion; the OP s
I have an old Redhat box that started doing this a while back. After a lot of
hair pulling, I finally figured out that the problem was related to spam
floods, but seemed to be caused by the syslog daemon. I shut down syslogd
for a few days, and the problem went away completely. After those few
Matias Lopez Bergero wrote:
> Hello!
>
> I was very happy using SpamAssassin at my email server (Xeon 2.8GHz,
> 1.5 GB memory, Dual Ultra SCSI HD 73.4GB in RAID 1, Linux 2.4.33)
>
> The last few weeks I have noted (angry users calling me by phone)
> that the server is really slow. The loadav goes
->
-> I was very happy using SpamAssassin at my email server (Xeon 2.8GHz,
-> 1.5 GB memory, Dual Ultra SCSI HD 73.4GB in RAID 1, Linux 2.4.33)
->
-> The last few weeks I have noted (angry users calling me by phone)
-> that the server is really slow. The loadav goes from 1.5 to 12.5;
-> normal
Rob McEwen (PowerView Systems) wrote:
>>The last few weeks I have noted (angry users calling me by phone) that
>>the server is really slow.
>
>
> Don't know for sure, but I suspect slower than usual Razor and/or DCC servers?
I am just running Pyzor.
BR,
Matias.
> The last few weeks I have noted (angry users calling me by phone) that
> the server is really slow.
Don't know for sure, but I suspect slower than usual Razor and/or DCC servers?
--Rob McEwen
Hello!
I was very happy using SpamAssassin at my email server (Xeon 2.8GHz, 1.5
GB memory, Dual Ultra SCSI HD 73.4GB in RAID 1, Linux 2.4.33)
The last few weeks I have noted (angry users calling me by phone) that
the server is really slow. The loadav goes from 1.5 to 12.5; normally is
about 3.00.
On Wed, Oct 05, 2005 at 06:28:48AM +0100, John Hodson wrote:
> I have solved this problem with the help of suggestions from Rick
> Macdougall, Matthew Newton, and Bob Menchal. Thanks chaps!
Excellent!
> suggestions were using spamassassin -D to debug, corrupt Rule in .cf file,
> and corrupt bayes
I have solved this problem with the help of suggestions from Rick
Macdougall, Matthew Newton, and Bob Menchal. Thanks chaps!
suggestions were using spamassassin -D to debug, corrupt Rule in .cf file,
and corrupt bayes database.
It turned out to be a problem with one of the rules in one of the man
Hello John,
Monday, October 3, 2005, 1:11:23 PM, you wrote:
JH> Only clue I have is that if I run sa-learn, it takes minutes to respond at
JH> all, and in the output appears "Parsing of undecoded UTF-8 will give
JH> garbage when decoding entities at
JH> [...]/Mail/SpamAssassin/HTML.pm line
JH> 18
John Hodson wrote:
After having used Spamassassin with mime-defang for some considerable
time, with no problems, we are suddenly experiencing serious performance
problems.
Last thursday, after effectively being mail bombed, which caused max
mimedefang children to be spawned, and mail coming in
After having used Spamassassin with mime-defang for some considerable
time, with no problems, we are suddenly experiencing serious performance
problems.
Last thursday, after effectively being mail bombed, which caused max
mimedefang children to be spawned, and mail coming in at such a rate that
cally configured machines.
Thank you,
Tom
-Original Message-
From: Martin Hepworth [mailto:[EMAIL PROTECTED]
Sent: Friday, January 21, 2005 3:17 AM
To: J Thomas Hancock
Cc: users@spamassassin.apache.org
Subject: Re: Performance Problems
Tom
Known issue with spamdspawning too many children
Possibly you have a large auto-whitelist?
Loren
Tom
Known issue with spamdspawning too many children on SA 3.0.x.
You can
1)reduce the number of children with the -m parameter. Alot of people
have this to soemthing like 10 by default. If you reduce it to 5 or even
2 it should sort the problem
2) patch the source
http://bugzilla.spamassass
I'm having a weird memory issue with one of our mail proxies that I am
hoping to get some help with.
We currently have 2 mail proxies that are set up identically. They run
Fedora Core 3, postfix 2.1.4, and SA 3.0.X. SA was installed via perl's
MCPAN. We are using a Foundry ServerIron to load ba
into my MTA problem.
Dimitry
-Original Message-
From: Gavin Cato [mailto:[EMAIL PROTECTED]
Sent: Wednesday, 1 December 2004 7:27 PM
To: Dimitry Peisakhov; users@spamassassin.apache.org
Subject: Re: spamd performance problems - again
Could your MTA be the bottleneck?
On 1/12/04 5:14 PM
On Wed, Dec 01, 2004 at 11:06:12AM -0600, Smart,Dan wrote:
> I believe a SPAMD problem has be pinpointed with SMP kernels using
> hyperthreading. Look through the bug list.
>
Huh? I disagree. I push 4-5 msgs a sec through my hyperthreaded
machine without a problem. Please do not spread such FU
khov
> Cc: 'users@spamassassin.apache.org'
> Subject: Re: spamd performance problems - again
>
> While the queue is building up, run a top and see what's
> holding things up. dstat is another handy program for this
> sort of work.
>
> Since it's a dual
While the queue is building up, run a top and see what's holding things
up. dstat is another handy program for this sort of work.
Since it's a dual processor, I'd suggest building a 2.6 series kernel
for it, since that handles smp much nicer.
On Wed, Dec 01, 2004 at 05:14:02PM +1100, Dimitry Pe
Could your MTA be the bottleneck?
On 1/12/04 5:14 PM, "Dimitry Peisakhov" <[EMAIL PROTECTED]>
wrote:
> Hi Guys,
>
> I wrote to the list a few weeks ago asking for advice on spamd
> performance. I got some, and have implemented it, but dont know if i'm
> seeing a performance improvement. The p
Hi Guys,
I wrote to the list a few weeks ago asking for advice on spamd
performance. I got some, and have implemented it, but dont know if i'm
seeing a performance improvement. The performance i'm getting is far from
other people are reporting, it seems.
I'm running spamassassin 3.0.1 wit
36 matches
Mail list logo
