On Thu, Nov 5, 2009 at 11:46 AM, Steve Basford
wrote:
> Freddie Cash wrote:
>>
>> Yes, I still have this directory. If anyone is interested in it, I
>> can tar it up and make it available. Can also tar up the working
>> directory is needed.
>
> Yep, I'll take a look and see if I can see anything
Freddie Cash wrote:
Yes, I still have this directory. If anyone is interested in it, I
can tar it up and make it available. Can also tar up the working
directory is needed.
Hi,
Yep, I'll take a look and see if I can see anything this end.
Cheers,
Steve
Sanesecurity
_
$$ uname -a
Linux 2.6.9-42.ELsmp #1 SMP Wed Jul 12 23:32:02 EDT 2006 x86_64 x86_64
x86_64 GNU/Linux
$$
Thanks,
Avinash
___
Help us build a comprehensive ClamAV guide: visit http://wiki.clamav.net
http://www.clamav.net/support/ml
Hi Tom,
We are using clamav version 0.95.2 with both official and unofficial
signatures.
Last week we observed clamd is taking more time for scanning mails due to
high load on the server. To fix the issue, installed older version 0.95.1,
but there was no use (later came back to 0.95.2). After rem
At 9:32 PM +0530 11/3/09, Avinash wrote:
Hi everyone,
Thanks for the quick response.
We are using the below 6 sanesecurity files.
junk.ndb
phish.ndb
scam.ndb
spear.ndb
lott.ndb
spam.ldb
Some more info:
I tried with adding these files one by one to clamd database, junk.ndb is
causing more loa
Hi everyone,
Thanks for the quick response.
We are using the below 6 sanesecurity files.
junk.ndb
phish.ndb
scam.ndb
spear.ndb
lott.ndb
spam.ldb
Some more info:
I tried with adding these files one by one to clamd database, junk.ndb is
causing more load among all. Phish.ndb, scam.ndb and spear.
> Last week I offered some help to early diagnose possible problems before
> they hit the end users and I was trying to establish some cooperation
> with you and the other db providers in order to improve your QA process.
Hi sorry for not replying earlier... I'll email off-list with a few
thou
Steve,
I see more and more custom db related issues on this list...
Last week I offered some help to early diagnose possible problems before
they hit the end users and I was trying to establish some cooperation
with you and the other db providers in order to improve your QA process.
Just in case
> Hi everyone,
>
> We are using Sanesecurity signatures in clamd for scanning mails. Recently
> we are seeing some load issues on clamd server due to sanesecurity
> signatures (load is automatically decreasing when the sanesecurity sigs
> are
> removed)
Hi Avinash,
I guess as others have already
On 11/03/2009 12:21 PM, Freddie Cash wrote:
>
> On a whim, I renamed the clamav database directory, ran freshclam to get
> just the basic signatures, and restarted clamd. Number of signatures went
> from 925,000+ to under 600,000, and CPU usage dropped to below 20%. Cleared
> out 1200 messages fr
On Mon, Nov 2, 2009 at 1:45 PM, Tom Shaw wrote:
> At 4:10 PM -0600 11/2/09, Noel Jones wrote:
>
>> On 11/2/2009 1:42 PM, Avinash wrote:
>>
>>> Hi everyone,
>>>
>>> We are using Sanesecurity signatures in clamd for scanning mails.
>>> Recently
>>> we are seeing some load issues on clamd server due
At 4:10 PM -0600 11/2/09, Noel Jones wrote:
On 11/2/2009 1:42 PM, Avinash wrote:
Hi everyone,
We are using Sanesecurity signatures in clamd for scanning mails. Recently
we are seeing some load issues on clamd server due to sanesecurity
signatures (load is automatically decreasing when the sanes
On 11/2/2009 1:42 PM, Avinash wrote:
Hi everyone,
We are using Sanesecurity signatures in clamd for scanning mails. Recently
we are seeing some load issues on clamd server due to sanesecurity
signatures (load is automatically decreasing when the sanesecurity sigs are
removed)
Does anyone face t
Trevor Dodds wrote:
I had the same problem running on a Sun Fire X2200 - Solaris 10.
88.7 I have a load avg of around 0.70 with .90 this shot up to 6.0
Cpu usage also on 88.7 is around 20% during large amounts of email with
.90 this shot up to 90%.
I've gone back to 88.7 as I can't afford the per
Hi,
High CPU usage and too many open files problems here with 0.90 on
OpenBSD 3.5. Have reversed to 0.88.7.
Abdul
___
Help us build a comprehensive ClamAV guide: visit http://wiki.clamav.net
http://lurker.clamav.net/list/clamav-users.html
nce hit.
I compared my conf files to 88.7 and they are pretty similar.
Trevor
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Pete 'Wolfy'
Hanson
Sent: 15 February 2007 12:52 AM
To: ClamAV users ML
Subject: Re: [Clamav-users] load under 0.90
On 2/14/
On 2/14/07, Rick Pim <[EMAIL PROTECTED]> wrote:
a followup to my last note.
using clamav-milter with --external seems to cause significant
load issues: the system load average seemed top climb without bound --
i shut things down when it hit around 40.
i shutdown clamd and clamav-milter and re
Tomasz Kojm wrote:
> On Thu, 28 Apr 2005 17:01:24 -0700
> Kurt Buff <[EMAIL PROTECTED]> wrote:
>
> > Tomasz Kojm wrote:
> > > ClamAV 0.90 will support the NodalCore hardware accelerator:
> > >
> > > http://sourceforge.net/forum/forum.php?forum_id=458047
> >
> > > With hw acceleration my old Athlo
Trog said this with great authority:
> If your users are also accessing the data via Samba, then use on-access
> scanning via Samba VFS. The vscan module does that
> http://www.openantivirus.org/projects.php#samba-vscan
Another good idea!
CD
Ever lied? You're a liar. Ever stolen? You're a thi
Lionel Bouton said this with great authority:
> You may try to wrap the scan in a Perl script which only scan files with
> modified mtimes.
Good idea!
CD
Ever lied? You're a liar. Ever stolen? You're a thief. Ever hated? The
bible equates hate with murder. Ever lusted? Jesus equated lust w
Jakub Suchy said this with great authority:
> Is it possible to access your data in differrent way than via Samba? Samba
> is very ineffective and slow and is causing overheads. Maybe you should
> also try 2.6.x kernel with more effective I/O scheduler
We're not going through Samba to clamscan the
Chris de Vidal wrote the following on 04/28/05 22:09 :
>Matt Fretwell said this while chewing gum:
>
>
>> Scan on an evening when your users are gone?
>>
>>
>
>Thanks for the fast reply!
>
>A 1.25TB Samba server takes several days to do a full scan with clamscan.
>
>
>
You may try to wrap
Chris de Vidal wrote:
> > <12 line signature snipped>
>
> You'll like my newer, shorter one :-)
>
> CD
>
> Ever lied? You're a liar. Ever stolen? You're a theif. Ever hated?
> The bible equates hate with murder. Ever lusted? Jesus equated lust with
> adultery. You've broken God's law. H
On Fri, 2005-04-29 at 10:26 +0200, Jakub Suchy wrote:
> > Rob MacGregor wrote:
> >> On 4/28/05, Chris de Vidal <[EMAIL PROTECTED]> wrote:
>
> >>>We have a 1.25TB Samba server that
> > Yes I/O bottleneck.
>
> Is it possible to access your data in differrent way than via Samba? Samba
> is very inef
> Rob MacGregor wrote:
>> On 4/28/05, Chris de Vidal <[EMAIL PROTECTED]> wrote:
>>>We have a 1.25TB Samba server that
> Yes I/O bottleneck.
Is it possible to access your data in differrent way than via Samba? Samba
is very ineffective and slow and is causing overheads. Maybe you should
also try
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Rob MacGregor wrote:
> On 4/28/05, Chris de Vidal <[EMAIL PROTECTED]> wrote:
>
>>ClamAV 0.83 from dag.wieers.com
>>CentOS 3.3 (A.K.A RedHat Enterprise Linux AS3)
>>Kernel 2.4.28
>>
>>I have a full system scan cron job. We have a 1.25TB Samba server t
On 4/28/05, Chris de Vidal <[EMAIL PROTECTED]> wrote:
> ClamAV 0.83 from dag.wieers.com
> CentOS 3.3 (A.K.A RedHat Enterprise Linux AS3)
> Kernel 2.4.28
>
> I have a full system scan cron job. We have a 1.25TB Samba server that
> causes the load average to go to 5+. When that happens the users n
Todd Lyons said this while chewing gum:
> If the scan is occurring during normal business hours, then yeah, you
> need to nice the heck out of it to keep it from sucking away
> performance. If you know a bit about c, you could add a commandline
> option --delay=x where x is the number of milliseco
On Thu, 28 Apr 2005 17:01:24 -0700
Kurt Buff <[EMAIL PROTECTED]> wrote:
> Tomasz Kojm wrote:
> > ClamAV 0.90 will support the NodalCore hardware accelerator:
> >
> > http://sourceforge.net/forum/forum.php?forum_id=458047
>
> > With hw acceleration my old Athlon 1200 was able to scan 2GB of
> > d
Tomasz Kojm wrote:
> ClamAV 0.90 will support the NodalCore hardware accelerator:
>
> http://sourceforge.net/forum/forum.php?forum_id=458047
> With hw acceleration my old Athlon 1200 was able to scan 2GB of
> data in one minute.
2gb in one minute, ceterus paribus, means roughly 8.5 hours per ter
On Thu, Apr 28, 2005 at 06:15:17PM -0300, René Bellora said:
> clamdscan is way faster
Not necessarily.
Single file:
[EMAIL PROTECTED]:~$ clamscan F5D5010.exe
F5D5010.exe: OK
--- SCAN SUMMARY ---
Known viruses: 33876
Engine version: 0.84rc2
Scanned directories: 0
Scanned files:
On Thu, Apr 28, 2005 at 04:18:41PM -0400, Chris de Vidal wrote:
> Have you ever lied, no matter the color? I have. Have you ever stolen
> anything, no matter the value? I have. Have you ever lusted? I have.
> Jesus said whoever looks at a woman with lustful intent has already
> committed adu
Tomasz Kojm wrote:
>With hw acceleration my old Athlon 1200 was able to scan 2GB of data in
>one minute.
>
>
2Gb in one minute!!!
That's pretty impressive - I wouldn't have thought you could do a
directory listing of that much data in one minute (assuming it's not a
single file of course!)
--
Chris de Vidal wanted us to know:
>Is our situation that foreign? I was half expecting someone to tell me
>"We have a 5TB FTP server and can scan it all under 40 minutes with .25
>extra load average! You just have to do this..."
Most people with stuff that large have a three teir system:
1) AV
Morgan Smith said this while chewing gum:
> Do compiler options help? YMMV but if I'm looking for speed I'll
> compile from source or create an RPM from the SRPM, and I'll add some
> optimizations:
> ./configure CFLAGS=-O3 -march=pentium4
>
> The man page for gcc may provide more insight.
OK I'll
Chris de Vidal wrote:
>ClamAV 0.83 from dag.wieers.com
>CentOS 3.3 (A.K.A RedHat Enterprise Linux AS3)
>Kernel 2.4.28
>
>I have a full system scan cron job. We have a 1.25TB Samba server that
>causes the load average to go to 5+. When that happens the users notice
>and I must kill the script.
On Thu, 28 Apr 2005 16:09:55 -0400 (EDT)
"Chris de Vidal" <[EMAIL PROTECTED]> wrote:
> Matt Fretwell said this while chewing gum:
> > Scan on an evening when your users are gone?
>
> Thanks for the fast reply!
>
> A 1.25TB Samba server takes several days to do a full scan with
> clamscan.
Clam
Matt Fretwell said this while chewing gum:
> Surely you're not doing a full scan each time, are you?
Yeah, just like the large Windows boxen we run. It's a corporate
requirement that we scan daily but clamscan is just too slow for a daily
scan so I run it weekly. It's a decent compromise; I'd co
René Bellora wrote:
> clamdscan is way faster
Only on a frequent stop/start type scenario. (Unless things have
changed). For an extended duration scan, any difference would be
negligible.
Matt
___
http://lurker.clamav.net/list/clamav-users.html
Chris de Vidal wrote:
> > Scan on an evening when your users are gone?
>
> Thanks for the fast reply!
>
> A 1.25TB Samba server takes several days to do a full scan with
> clamscan.
Surely you're not doing a full scan each time, are you? How often do you
scan the system?
I can't answer on
Chris de Vidal wrote:
ClamAV 0.83 from dag.wieers.com
CentOS 3.3 (A.K.A RedHat Enterprise Linux AS3)
Kernel 2.4.28
I have a full system scan cron job. We have a 1.25TB Samba server that
causes the load average to go to 5+. When that happens the users notice
and I must kill the script. I've even
Matt Fretwell said this while chewing gum:
> Scan on an evening when your users are gone?
Thanks for the fast reply!
A 1.25TB Samba server takes several days to do a full scan with clamscan.
CD
Have you ever lied, no matter the color? I have. Have you ever stolen
anything, no matter the valu
Chris de Vidal wrote:
> Ideas?
Scan on an evening when your users are gone?
Matt
___
http://lurker.clamav.net/list/clamav-users.html
Sorted the problem out - it appears that clamscan will fork new
processes everytime it is called by the qmail scanner - I switched to
using clamdscan which uses the clamd daemon.
It has halved the original load to average of 1-3 ...
On Fri, 2004-03-12 at 23:47, Jeremy Kitchen wrote:
> On Fri,
On Fri, 12 Mar 2004 12:18:13 +0200
Scott Ryan <[EMAIL PROTECTED]> wrote:
> Hello all, I have upgraded clamav from 0.55 to 0.67-1 so I can now
> catch the bagle-passwd worm. It works and it catches the virus, but I
> have noticed a significant increase in the load on the machine. I am
> running qma
On Fri, 2004-03-12 at 04:18, Scott Ryan wrote:
> Hello all, I have upgraded clamav from 0.55 to 0.67-1 so I can now catch
> the bagle-passwd worm. It works and it catches the virus, but I have
> noticed a significant increase in the load on the machine. I am running
> qmail on a 6cpu Xeon P3 700 wi
46 matches
Mail list logo
