unix.in is 100% and all email is stuck in Local Delivery

[aacservices]

HI
I refer to bellow forum...

http://www.scalix.com/forums/viewtopic.php?p=49797

unix.in utilizing 100% cpu ..
and any email comming to scalix is storing local Delivery queue and does not deliver into user inbox.

every time i restart Local delivery service. its delivers couple of 100.. then stuck again ..


i tryed to shutdown unix.in by using
omoff -d0 -w unix smtpd && omon unix smtpd

rebooted the whole server .

no luck

only error i am seeing in fatal.log

SERIOUS ERROR Browser (Mime_Browser ) Mon Sep 26 14:57:24 2011
[SYS 2] No such file or directory
File Name: /var/opt/scalix/sx/s/data/000007q/0249s3h
Pid of logging process: 10692
Current errno value: 2


ERROR Browser (Mime_Browser ) Mon Sep 26 14:57:24 2011
[OM.MIME 4000] Browser Args :mime.browse -l C -m -M -o /var/opt/scalix/sx/s/temp/mime_cache/mimebFFwtb 00043883089751c0
File Name: /var/opt/scalix/sx/s/data/000007q/0249s3h
Pid of logging process: 10692
Current errno value: 2

SERIOUS ERROR Browser (Mime_Browser ) Mon Sep 26 14:57:24 2011
[OM 9171] Converter failure:- %s -> %s

File Name: /var/opt/scalix/sx/s/data/000007q/0249s3h
Pid of logging process: 10692
Current errno value: 2

only error is : file system error to some directory ....

can any one please give me some advise .. how to solve this issue ??

thanks .

[mikethebike]

Hi,

you need to check how many and what the messages are in the queue.
omstat -q unix

if they all look the same subject etc, you may be getting spammed.

you may need to stop any more mails entering the system while you clear down that queue

Mick

[aacservices]

Hi,

you need to check how many and what the messages are in the queue.
omstat -q unix

Mick

Hi
thanks for response.

can you tell me is there any limitation how big email size sendmail can handle for unix.in (internet mail gateway)

because..

we have exaim Server as External Mailserver -> From here all email comes to Scalix server

now problem is :

when exim is trying to sent email to scalix which is 42 mb.. and unix.in goes 100%

after i delete those email from exim queue.. its seems there is not any unix.in process which is using 100%

thanks for your response.

[mikethebike]

there is no theoretical limit to what it can handle....although 42M is big in anyone's books for an external email
It may be worth increasing (or at least checking) how many queue runners you have configured (omstat -s -p ), generally its a balancing act with this one...increasing the service router or local delivery (or internet gateway) with more aux runners will speed things up, so in your instance may improve things as one runner will handle that big mail, whilst others will continue to process the rest of the queue.
It may just be that the queue builds up, and never gets a chance to recover (the bigger the queue gets, the longer it takes to process each iteration of the queue).

Mick