Index: [Article Count Order] [Thread]

Date:  Tue, 18 Apr 2006 18:44:32 -0600
From:  "Rodrigo Ordonez Licona" <rodrigo (at mark) xnet.com.mx>
Subject:  [coba-e:04755] Re: help
To:  <coba-e (at mark) bluequartz.org>
Message-Id:  <200604190044.k3J0ieLE019125 (at mark) admin.xnet.com.mx>
In-Reply-To:  <003401c662fa$4a2dcf20$0901a8c0 (at mark) alperin.calperin.com>
X-Mail-Count: 04755

This did not help us,

We are now seeing log messages that read 

Berkeley db /etc/sasldb2: No such file or directory Apr 18 16:50:46 admin
sendmail[2389]: unable to open Berkeley db /etc/sasldb2: No such file or
directory 

Not sure they are related did not appear before, they seem to be harmless,
we lowered again out pop limit to default 80 

And wait for tomorrow storm removed 100 accounts of email, however raq 550
could handle this without problems.

We did notice the Increase in TCP connections it went from 250 to 600 during
the stress time, 300 of which were suppoesed to be port 110 connections.

Hope this helps to give clues.

Regards

Rodrigo O
Xnet / Pwworks

-----Original Message-----
From: Carlos Alperin [mailto:calperin (at mark) senecacom.net] 
Sent: Martes, 18 de Abril de 2006 09:11 a.m.
To: coba-e (at mark) bluequartz.org
Subject: [coba-e:04748] Re: help
Importance: High

Ok,

This is the status:

As I couldn't recompile qpopper due to I don't know where is, I made the
modifications suggested by patricko (at mark) staff.singnet.com.sg, that is modify
/etc/xinetd.d/pop3 & /etc/pam.d/pop3.

In /etc/xinetd.d/pop3

# qpopper config file for xinetd
service pop-3
{
        disable = no
        flags       = REUSE NAMEINARGS
        socket_type = stream
        protocol    = tcp
        wait        = no
        user        = root
        server      = /usr/sbin/in.qpopper
        server_args = in.qpopper -R -S
        instances = 1024
}

And we increase the instances to 1024, just even if we go to 9999 as someone
said, when we restart the system, they put that back to 1024. Also, if you
go the the BlueQuartz GUI you cannot go above 1024. patricko proposs to
include also the -s on server_args but that one doesn't work.

Now, the one that really make the difference was the /etc/pam.d/pop3

As soon as we modify that one, everything went back to normal and fast.

#%PAM-1.0
auth       requisite    /lib/security/pam_nologin.so
auth       requisite    /lib/security/pam_shells.so
#auth       required    /lib/security/pam_pwdb.so shadow nullok
#account    required    /lib/security/pam_pwdb.so
auth       required     /lib/security/pam_stack.so service=system-auth
account    required     /lib/security/pam_stack.so service=system-auth

We're going to wait a couple of days to see how this works now.

Thanks, however any info about how to migrate to something else is going to
be appretiate, just it looks that for a lot of people this is not enough.

We have about 700 users at this time on e-mails account.

Thanks,

Carlos Alperin
Seneca Communications, LLC


-----Original Message-----
From: "Ing. Ernesto PñÓez Estñ×ez" [mailto:info (at mark) ecualinux.com]
Sent: Monday, April 17, 2006 12:46 PM
To: coba-e (at mark) bluequartz.org
Subject: [coba-e:04717] Re: help

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Carlos

this has been discussed several times on the list.

My guess (as genius) is that qpopper working via xinetd is too slow to
handle too much load (or too few).

I ended by compiling and installing dovecot as suggested on this list
previously and this nightmare went away finally.

It is really annoying indeed.

Is there anything else I can help you?

regards
- --
Ing. Ernesto PñÓez Estñ×ez
http://www.ecualinux.com
USA: + 1 404 795 0321
Ecuador: (02)3412402 - (09) 9246504


Carlos Alperin wrote:
> I'm not the one that originate the e-mail, but since you're the genius
let's ask:
> 
> I have CentOS 4.3 on a P4 3.0 GHz, 4 GB Ram and 250 GB HD on Raid 1 
> just
for /home.
> 
> We have problems with in.qpopper that gaves timeout all the time. We 
> tried
to install 
> Dovecot last week & didn't work. Then we change some parameters like 
> -s -R
-S on one 
> of the configuration files of the qpopper, and we have no problems 
> during
the last 4 
> days. 
> 
> Today, with the high traffic, the problems show up again.
> 
> Any clue how to fix this?
> 
> Any good procedure to move to Dovecot or some other POP handler?
> 
> Thanks,
> 
> Carlos Alperin
> Seneca Communications, LLC
> 
> On Mon, 17 Apr 2006 10:22:34 -0500, Ing. Ernesto PñÓez Estñ×ez wrote 
> Im the Genius of the lamp, how may I help you?
> 
> Diego Lo Giudice wrote:
>>>> Help
>>>>
>>>>
> --
> Ing. Ernesto PñÓez Estñ×ez
> http://www.ecualinux.com
> USA: + 1 404 795 0321
> Ecuador: (02)3412402 - (09) 9246504

> --
> Open WebMail Project (http://openwebmail.org)



-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFEQ8ZR8sSWs7RP4EMRAh1TAKDQ8bWGCh1xYVF4y4d91V7GRBbYwACeK3q+
WZ/js3xWzeWCoSP2KFBiDLc=
=iESI
-----END PGP SIGNATURE-----