Re: Fwd: Re: [qmailtoaster] qq soft reject errors on high load

[Rajesh M]

eric

i did not yet simulate it using your script

these are errors in my live production server during peak hours today and there 
were mails with and without attachments.

the max processes i got was 47 only.

rajesh


- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To: qmailtoaster-list@qmailtoaster.com
Sent: Mon, 4 Sep 2017 11:19:35 -0600
Subject:

Did you hit it with email containing attachments as well?


On 9/4/2017 10:58 AM, Rajesh M wrote:
> eric
>
> the max i saw was 47 connections using the shell script you gave me.
>
> there were a few errors related to qq soft reject but they got delivered.
>
> rajesh
>
>
>
> - Original Message -
> From: Eric Broch [mailto:ebr...@whitehorsetc.com]
> To: qmailtoaster-list@qmailtoaster.com
> Sent: Sat, 2 Sep 2017 12:35:14 -0600
> Subject:
>
> Rajesh,
>
> I'm still not sure the process limit is being reached.
>
> I'm testing a COS6/QMT server now and haven't been able to bring about
> 'qq soft reject' or failure of any sort. I'm hitting it with email using
> a delivery script from two servers with attachment size of 320KB.
>
> try the script below (psmem) to monitor:
>
> 
>
> #!/bin/bash
> ps -C $1 -O rss | gawk '{ count ++; sum += $2 }; END {count --; print
> "Number of processes =",count; print "Memory usage per process
> =",sum/1024/count, "MB"; print "Total memory usage =", sum/1024, "MB" ;};'
>
> 
>
> # chmod 755 psmem
>
> # watch -n 1 ./psmem simscan
>
>
> Not sure I should change anything until we figure out what's going on.
>
> Eric
>
>
>
> On 9/2/2017 12:03 PM, Rajesh M wrote:
>> eric
>>
>> i could not get the bash script working with either of the two options. 
>> still gives error
>>line 23: `return-limits': not a valid identifier.
>>
>>
>>
>> can the limits issue in simscan be fixed ?
>>
>> thanks,
>> rajesh
>>
>> - Original Message -
>> From: Eric Broch [mailto:ebr...@whitehorsetc.com]
>> To:qmailtoaster-list@qmailtoaster.com
>> Sent: Fri, 1 Sep 2017 23:46:15 -0600
>> Subject:
>>
>> Rajesh,
>>
>> It has something to do with #!/bin/bash or #!/bin/sh.
>>
>> At top of script, when I use #!/bin/sh I get the same error you do, but
>> when I use #!/bin/bash script works fine.
>>
>>
>> Eric
>>
>>
>> On 9/1/2017 11:07 PM, Rajesh M wrote:
>>> eric
>>>
>>> 1) is their any resolution for this issue ?
>>>
>>> 2) if i run the script it throws error as such
>>> .
>>> /limits.sh: line 23: `return-limits': not a valid identifier
>>>
>>> #!/bin/bash
>>> if [ "$#" -ne "1" ]; then
>>>   echo ""
>>>   echo -e "\033[01;32mLimit checker\033[00m"
>>>   echo -e "\033[01;37mUsage:\033[01;33m $0 process_name\033[00m"
>>>   echo ""
>>>   exit 0
>>> fi
>>>
>>> return-limits(){
>>>   for process in $@; do
>>>  process_pids=`ps -C $process -o pid --no-headers | cut -d " " -f 2`
>>>
>>>  if [ -z $@ ]; then
>>> echo "[no $process running]"
>>>  else
>>> for pid in $process_pids; do
>>>echo "[$process #$pid -- limits]"
>>>cat /proc/$pid/limits
>>>  done
>>>  fi
>>>   done
>>> }
>>> return-limits $1
>>>
>>> rajesh
>>>
>>>
>>>
>>> - Original Message -
>>> From: Eric Broch [mailto:ebr...@whitehorsetc.com]
>>> To:qmailtoaster-list@qmailtoaster.com
>>> Sent: Fri, 1 Sep 2017 17:28:40 -0600
>>> Subject:
>>>
>>> I sent myself a large file so that I could examine simscan with the
>>> below script (limits.sh). Here's the output:
>>>
>>> [root@pet105 simscan-1.4.0]# ./limits simscan
>>> [simscan #31535 -- limits]
>>> Limit   
>>>    
>>>    
>>>    
>>>    
>>>    
>>>    Soft Limit 
>>>    
>>>    
>>>     Hard 
>>> Limit   
>>>    
>>>    
>>>   Units
>>> Max cpu time   
>>>    
>>>    
>>>    
>>>   unlimited unlimited 
>>>    
>>>    
>>>   

Re: Fwd: Re: [qmailtoaster] qq soft reject errors on high load

[Rajesh M]

eric

i got 3 types of errors like earlier.

failure to fork, ripmime
and  /var/qmail/simscan/1simscan: check_spam had an error ret: -1

what exactly would be  /var/qmail/simscan/1simscan ... there is no such file

do want me to send you a few of my log files directly to your email id ?

rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To: qmailtoaster-list@qmailtoaster.com
Sent: Mon, 4 Sep 2017 11:19:04 -0600
Subject:

What kind of qq soft reject errors were there? Failure to fork or ripmime?


On 9/4/2017 10:58 AM, Rajesh M wrote:
> eric
>
> the max i saw was 47 connections using the shell script you gave me.
>
> there were a few errors related to qq soft reject but they got delivered.
>
> rajesh
>
>
>
> - Original Message -
> From: Eric Broch [mailto:ebr...@whitehorsetc.com]
> To: qmailtoaster-list@qmailtoaster.com
> Sent: Sat, 2 Sep 2017 12:35:14 -0600
> Subject:
>
> Rajesh,
>
> I'm still not sure the process limit is being reached.
>
> I'm testing a COS6/QMT server now and haven't been able to bring about
> 'qq soft reject' or failure of any sort. I'm hitting it with email using
> a delivery script from two servers with attachment size of 320KB.
>
> try the script below (psmem) to monitor:
>
> 
>
> #!/bin/bash
> ps -C $1 -O rss | gawk '{ count ++; sum += $2 }; END {count --; print
> "Number of processes =",count; print "Memory usage per process
> =",sum/1024/count, "MB"; print "Total memory usage =", sum/1024, "MB" ;};'
>
> 
>
> # chmod 755 psmem
>
> # watch -n 1 ./psmem simscan
>
>
> Not sure I should change anything until we figure out what's going on.
>
> Eric
>
>
>
> On 9/2/2017 12:03 PM, Rajesh M wrote:
>> eric
>>
>> i could not get the bash script working with either of the two options. 
>> still gives error
>>line 23: `return-limits': not a valid identifier.
>>
>>
>>
>> can the limits issue in simscan be fixed ?
>>
>> thanks,
>> rajesh
>>
>> - Original Message -
>> From: Eric Broch [mailto:ebr...@whitehorsetc.com]
>> To:qmailtoaster-list@qmailtoaster.com
>> Sent: Fri, 1 Sep 2017 23:46:15 -0600
>> Subject:
>>
>> Rajesh,
>>
>> It has something to do with #!/bin/bash or #!/bin/sh.
>>
>> At top of script, when I use #!/bin/sh I get the same error you do, but
>> when I use #!/bin/bash script works fine.
>>
>>
>> Eric
>>
>>
>> On 9/1/2017 11:07 PM, Rajesh M wrote:
>>> eric
>>>
>>> 1) is their any resolution for this issue ?
>>>
>>> 2) if i run the script it throws error as such
>>> .
>>> /limits.sh: line 23: `return-limits': not a valid identifier
>>>
>>> #!/bin/bash
>>> if [ "$#" -ne "1" ]; then
>>>   echo ""
>>>   echo -e "\033[01;32mLimit checker\033[00m"
>>>   echo -e "\033[01;37mUsage:\033[01;33m $0 process_name\033[00m"
>>>   echo ""
>>>   exit 0
>>> fi
>>>
>>> return-limits(){
>>>   for process in $@; do
>>>  process_pids=`ps -C $process -o pid --no-headers | cut -d " " -f 2`
>>>
>>>  if [ -z $@ ]; then
>>> echo "[no $process running]"
>>>  else
>>> for pid in $process_pids; do
>>>echo "[$process #$pid -- limits]"
>>>cat /proc/$pid/limits
>>>  done
>>>  fi
>>>   done
>>> }
>>> return-limits $1
>>>
>>> rajesh
>>>
>>>
>>>
>>> - Original Message -
>>> From: Eric Broch [mailto:ebr...@whitehorsetc.com]
>>> To:qmailtoaster-list@qmailtoaster.com
>>> Sent: Fri, 1 Sep 2017 17:28:40 -0600
>>> Subject:
>>>
>>> I sent myself a large file so that I could examine simscan with the
>>> below script (limits.sh). Here's the output:
>>>
>>> [root@pet105 simscan-1.4.0]# ./limits simscan
>>> [simscan #31535 -- limits]
>>> Limit   
>>>    
>>>    
>>>    
>>>    
>>>    
>>>    Soft Limit 
>>>    
>>>    
>>>     Hard 
>>> Limit   
>>>    
>>>    
>>>   Units
>>> Max cpu time   
>>>    
>>>    
>>>    
>>>   unlimited unlimited 
>>>    
>>>  

RE: Fwd: Re: [qmailtoaster] qq soft reject errors on high load

[Rajesh M]

eric

the max i saw was 47 connections using the shell script you gave me.

there were a few errors related to qq soft reject but they got delivered.

rajesh



- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To: qmailtoaster-list@qmailtoaster.com
Sent: Sat, 2 Sep 2017 12:35:14 -0600
Subject:

Rajesh,

I'm still not sure the process limit is being reached.

I'm testing a COS6/QMT server now and haven't been able to bring about
'qq soft reject' or failure of any sort. I'm hitting it with email using
a delivery script from two servers with attachment size of 320KB.

try the script below (psmem) to monitor:



#!/bin/bash
ps -C $1 -O rss | gawk '{ count ++; sum += $2 }; END {count --; print
"Number of processes =",count; print "Memory usage per process
=",sum/1024/count, "MB"; print "Total memory usage =", sum/1024, "MB" ;};'



# chmod 755 psmem

# watch -n 1 ./psmem simscan


Not sure I should change anything until we figure out what's going on.

Eric



On 9/2/2017 12:03 PM, Rajesh M wrote:
> eric
>
> i could not get the bash script working with either of the two options. still 
> gives error
>   line 23: `return-limits': not a valid identifier.
>
>
>
> can the limits issue in simscan be fixed ?
>
> thanks,
> rajesh
>
> - Original Message -
> From: Eric Broch [mailto:ebr...@whitehorsetc.com]
> To:qmailtoaster-list@qmailtoaster.com
> Sent: Fri, 1 Sep 2017 23:46:15 -0600
> Subject:
>
> Rajesh,
>
> It has something to do with #!/bin/bash or #!/bin/sh.
>
> At top of script, when I use #!/bin/sh I get the same error you do, but
> when I use #!/bin/bash script works fine.
>
>
> Eric
>
>
> On 9/1/2017 11:07 PM, Rajesh M wrote:
>> eric
>>
>> 1) is their any resolution for this issue ?
>>
>> 2) if i run the script it throws error as such
>> .
>> /limits.sh: line 23: `return-limits': not a valid identifier
>>
>> #!/bin/bash
>> if [ "$#" -ne "1" ]; then
>>  echo ""
>>  echo -e "\033[01;32mLimit checker\033[00m"
>>  echo -e "\033[01;37mUsage:\033[01;33m $0 process_name\033[00m"
>>  echo ""
>>  exit 0
>> fi
>>
>> return-limits(){
>>  for process in $@; do
>> process_pids=`ps -C $process -o pid --no-headers | cut -d " " -f 2`
>>
>> if [ -z $@ ]; then
>>echo "[no $process running]"
>> else
>>for pid in $process_pids; do
>>   echo "[$process #$pid -- limits]"
>>   cat /proc/$pid/limits
>> done
>> fi
>>  done
>> }
>> return-limits $1
>>
>> rajesh
>>
>>
>>
>> - Original Message -
>> From: Eric Broch [mailto:ebr...@whitehorsetc.com]
>> To:qmailtoaster-list@qmailtoaster.com
>> Sent: Fri, 1 Sep 2017 17:28:40 -0600
>> Subject:
>>
>> I sent myself a large file so that I could examine simscan with the
>> below script (limits.sh). Here's the output:
>>
>> [root@pet105 simscan-1.4.0]# ./limits simscan
>> [simscan #31535 -- limits]
>> Limit        
>>         
>>      Soft Limit   
>>         Hard 
>> Limit        
>>    Units
>> Max cpu time       
>>        unlimited 
>> unlimited       
>>      seconds
>> Max file size       
>>       unlimited unlimited 
>>         
>>    bytes
>> Max data size       
>>       unlimited unlimited 
>>         
>>    bytes
>> Max stack size 10485760 unlimited    
>>         bytes
>> Max core file size      
>>   0 unlimited      
>>       bytes
>> Max resident set      
>>     unlimited unlimited   
>>          
>> bytes
>> Max processes 1024 31121     
>>         
>>    processes
>> Max open files       
>>  

Fwd: Re: [qmailtoaster] qq soft reject errors on high load

[Eric Broch]

They're all temporary, for scanning purposes. Once the scanning is done, 
they all are deleted.



On 9/3/2017 3:57 PM, Remo Mattei wrote:

I do not see anything on my  :)  /var/qmail/simscan


Remo
On 9/3/17 12:13 PM, Eric Broch wrote:


Rajesh,


Those shouldn't be log files in that directory, but directories of 
this form...


[root@mail ~]# ls -l /var/qmail/simscan
total 7
drwxr-x--- 2 clamav vchkpw 4096 Apr 12  2016 1460525628.43125.19212
drwxr-x--- 2 clamav vchkpw 4096 Apr 12  2016 1460525917.767516.19248
drwxr-x--- 2 clamav vchkpw  131 Apr 12  2016 1460525934.725762.19262
drwxr-x--- 2 clamav vchkpw  131 Apr 12  2016 1460525934.728000.19263
drwxr-x--- 2 clamav vchkpw  131 Apr 12  2016 1460525934.730011.19264
drwxr-x--- 2 clamav vchkpw 4096 Apr 12  2016 1460525959.679164.19279
drwxr-x--- 2 clamav vchkpw  131 Apr 12  2016 1460525972.624759.19287

in which messages are broken into their respective parts (below)...

/var/qmail/simscan/1493074576.527224.5829:
total 1472

-rw-r- 1 clamav vchkpw 37 Apr 24 16:56 
addr.1493074576.527224.5829

-rw-r- 1 clamav vchkpw  20719 Apr 24 16:56 __If_
-rw-r- 1 clamav vchkpw 846597 Apr 24 16:56 msg.1493074576.527224.5829
-rw-r- 1 clamav vchkpw  0 Apr 24 16:56 textfile0
-rw--- 1 clamav vchkpw   9090 Apr 24 16:56 textfile1
-rw--- 1 clamav vchkpw  47495 Apr 24 16:56 textfile2
-rw--- 1 clamav vchkpw 362496 Apr 24 16:56 WGS-VRV Indoor Units.doc
-rw--- 1 clamav vchkpw 203264 Apr 24 16:56 WGS-VRV IV 208-230V 
Heat Pump.doc


for examination.

There should be no log files in /var/qmail/simscan and their should 
be no old files there either.


Eric


On 9/3/2017 12:54 PM, Rajesh M wrote:

eric

around 1 gb spare.

there are a lot of log files.

rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To:qmailtoaster-list@qmailtoaster.com
Sent: Sun, 3 Sep 2017 11:21:24 -0600
Subject:

Rajesh,

How much  available space is on the /var/qmail/simscan directory?

Eric
On 9/3/2017 12:19 AM, Rajesh M wrote:

eric

i would like to conduct more load intensive tests.

is it possible to provide me the shell script that you use for this purpose.

rajesh

-
To unsubscribe, e-mail:qmailtoaster-list-unsubscr...@qmailtoaster.com
For additional commands, e-mail:qmailtoaster-list-h...@qmailtoaster.com




-
To unsubscribe, e-mail:qmailtoaster-list-unsubscr...@qmailtoaster.com
For additional commands, e-mail:qmailtoaster-list-h...@qmailtoaster.com






--
Eric Broch
White Horse Technical Consulting (WHTC)

Re: [qmailtoaster] qq soft reject errors on high load

[Rajesh M]

eric

the logs are not in that simscan directory..

they re in the /var/log which is also in the same partition.

the files in the ls -l /var/qmail/simscan are temp ones which keep coming and 
going


rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To: qmailtoaster-list@qmailtoaster.com
Sent: Sun, 3 Sep 2017 13:13:47 -0600
Subject:

Rajesh,


Those shouldn't be log files in that directory, but directories of this
form...

[root@mail ~]# ls -l /var/qmail/simscan
total 7
drwxr-x--- 2 clamav vchkpw 4096 Apr 12  2016 1460525628.43125.19212
drwxr-x--- 2 clamav vchkpw 4096 Apr 12  2016 1460525917.767516.19248
drwxr-x--- 2 clamav vchkpw  131 Apr 12  2016 1460525934.725762.19262
drwxr-x--- 2 clamav vchkpw  131 Apr 12  2016 1460525934.728000.19263
drwxr-x--- 2 clamav vchkpw  131 Apr 12  2016 1460525934.730011.19264
drwxr-x--- 2 clamav vchkpw 4096 Apr 12  2016 1460525959.679164.19279
drwxr-x--- 2 clamav vchkpw  131 Apr 12  2016 1460525972.624759.19287

in which messages are broken into their respective parts (below)...

/var/qmail/simscan/1493074576.527224.5829:
total 1472

-rw-r- 1 clamav vchkpw     37 Apr 24 16:56 addr.1493074576.527224.5829
-rw-r- 1 clamav vchkpw  20719 Apr 24 16:56 __If_
-rw-r- 1 clamav vchkpw 846597 Apr 24 16:56 msg.1493074576.527224.5829
-rw-r- 1 clamav vchkpw      0 Apr 24 16:56 textfile0
-rw--- 1 clamav vchkpw   9090 Apr 24 16:56 textfile1
-rw--- 1 clamav vchkpw  47495 Apr 24 16:56 textfile2
-rw--- 1 clamav vchkpw 362496 Apr 24 16:56 WGS-VRV Indoor Units.doc
-rw--- 1 clamav vchkpw 203264 Apr 24 16:56 WGS-VRV IV 208-230V Heat
Pump.doc

for examination.

There should be no log files in /var/qmail/simscan and their should be
no old files there either.

Eric


On 9/3/2017 12:54 PM, Rajesh M wrote:
> eric
>
> around 1 gb spare.
>
> there are a lot of log files.
>
> rajesh
>
> - Original Message -
> From: Eric Broch [mailto:ebr...@whitehorsetc.com]
> To: qmailtoaster-list@qmailtoaster.com
> Sent: Sun, 3 Sep 2017 11:21:24 -0600
> Subject:
>
> Rajesh,
>
> How much  available space is on the /var/qmail/simscan directory?
>
> Eric
> On 9/3/2017 12:19 AM, Rajesh M wrote:
>> eric
>>
>> i would like to conduct more load intensive tests.
>>
>> is it possible to provide me the shell script that you use for this purpose.
>>
>> rajesh
>
> -
> To unsubscribe, e-mail: qmailtoaster-list-unsubscr...@qmailtoaster.com
> For additional commands, e-mail: qmailtoaster-list-h...@qmailtoaster.com
>
>
>
>
> -
> To unsubscribe, e-mail: qmailtoaster-list-unsubscr...@qmailtoaster.com
> For additional commands, e-mail: qmailtoaster-list-h...@qmailtoaster.com


-
To unsubscribe, e-mail: qmailtoaster-list-unsubscr...@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-h...@qmailtoaster.com

Re: [qmailtoaster] qq soft reject errors on high load

[Eric Broch]

Are you sure they're log files in /var/qmail/simscan? They shouldn't be 
log files.



On 9/3/2017 1:17 PM, Rajesh M wrote:

i have over 50 gb of log files which can be deleted

rajesh

- Original Message -
From: Remo Mattei [mailto:r...@mattei.org]
To: qmailtoaster-list@qmailtoaster.com
Sent: Sun, 3 Sep 2017 12:13:28 -0700
Subject:

1gb is not very much!

Il giorno 03 set 2017, alle ore 11:54, Rajesh M <24x7ser...@24x7server.net> ha 
scritto:

eric

around 1 gb spare.

there are a lot of log files.

rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To: qmailtoaster-list@qmailtoaster.com
Sent: Sun, 3 Sep 2017 11:21:24 -0600
Subject:

Rajesh,

How much  available space is on the /var/qmail/simscan directory?

Eric

On 9/3/2017 12:19 AM, Rajesh M wrote:
eric

i would like to conduct more load intensive tests.

is it possible to provide me the shell script that you use for this purpose.

rajesh


-
To unsubscribe, e-mail: qmailtoaster-list-unsubscr...@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-h...@qmailtoaster.com


-
To unsubscribe, e-mail: qmailtoaster-list-unsubscr...@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-h...@qmailtoaster.com

-
To unsubscribe, e-mail: qmailtoaster-list-unsubscr...@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-h...@qmailtoaster.com




-
To unsubscribe, e-mail: qmailtoaster-list-unsubscr...@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-h...@qmailtoaster.com


--
Eric Broch
White Horse Technical Consulting (WHTC)

Re: [qmailtoaster] qq soft reject errors on high load

[Rajesh M]

i have over 50 gb of log files which can be deleted

rajesh

- Original Message -
From: Remo Mattei [mailto:r...@mattei.org]
To: qmailtoaster-list@qmailtoaster.com
Sent: Sun, 3 Sep 2017 12:13:28 -0700
Subject:

1gb is not very much!

Il giorno 03 set 2017, alle ore 11:54, Rajesh M <24x7ser...@24x7server.net> ha 
scritto:

eric

around 1 gb spare.

there are a lot of log files.

rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To: qmailtoaster-list@qmailtoaster.com
Sent: Sun, 3 Sep 2017 11:21:24 -0600
Subject:

Rajesh,

How much  available space is on the /var/qmail/simscan directory?

Eric
> On 9/3/2017 12:19 AM, Rajesh M wrote:
> eric
>
> i would like to conduct more load intensive tests.
>
> is it possible to provide me the shell script that you use for this purpose.
>
> rajesh


-
To unsubscribe, e-mail: qmailtoaster-list-unsubscr...@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-h...@qmailtoaster.com


-
To unsubscribe, e-mail: qmailtoaster-list-unsubscr...@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-h...@qmailtoaster.com

-
To unsubscribe, e-mail: qmailtoaster-list-unsubscr...@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-h...@qmailtoaster.com


-
To unsubscribe, e-mail: qmailtoaster-list-unsubscr...@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-h...@qmailtoaster.com

Re: [qmailtoaster] qq soft reject errors on high load

[Eric Broch]

Rajesh,


Those shouldn't be log files in that directory, but directories of this 
form...


[root@mail ~]# ls -l /var/qmail/simscan
total 7
drwxr-x--- 2 clamav vchkpw 4096 Apr 12  2016 1460525628.43125.19212
drwxr-x--- 2 clamav vchkpw 4096 Apr 12  2016 1460525917.767516.19248
drwxr-x--- 2 clamav vchkpw  131 Apr 12  2016 1460525934.725762.19262
drwxr-x--- 2 clamav vchkpw  131 Apr 12  2016 1460525934.728000.19263
drwxr-x--- 2 clamav vchkpw  131 Apr 12  2016 1460525934.730011.19264
drwxr-x--- 2 clamav vchkpw 4096 Apr 12  2016 1460525959.679164.19279
drwxr-x--- 2 clamav vchkpw  131 Apr 12  2016 1460525972.624759.19287

in which messages are broken into their respective parts (below)...

/var/qmail/simscan/1493074576.527224.5829:
total 1472

-rw-r- 1 clamav vchkpw 37 Apr 24 16:56 addr.1493074576.527224.5829
-rw-r- 1 clamav vchkpw  20719 Apr 24 16:56 __If_
-rw-r- 1 clamav vchkpw 846597 Apr 24 16:56 msg.1493074576.527224.5829
-rw-r- 1 clamav vchkpw  0 Apr 24 16:56 textfile0
-rw--- 1 clamav vchkpw   9090 Apr 24 16:56 textfile1
-rw--- 1 clamav vchkpw  47495 Apr 24 16:56 textfile2
-rw--- 1 clamav vchkpw 362496 Apr 24 16:56 WGS-VRV Indoor Units.doc
-rw--- 1 clamav vchkpw 203264 Apr 24 16:56 WGS-VRV IV 208-230V Heat 
Pump.doc


for examination.

There should be no log files in /var/qmail/simscan and their should be 
no old files there either.


Eric


On 9/3/2017 12:54 PM, Rajesh M wrote:

eric

around 1 gb spare.

there are a lot of log files.

rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To: qmailtoaster-list@qmailtoaster.com
Sent: Sun, 3 Sep 2017 11:21:24 -0600
Subject:

Rajesh,

How much  available space is on the /var/qmail/simscan directory?

Eric
On 9/3/2017 12:19 AM, Rajesh M wrote:

eric

i would like to conduct more load intensive tests.

is it possible to provide me the shell script that you use for this purpose.

rajesh


-
To unsubscribe, e-mail: qmailtoaster-list-unsubscr...@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-h...@qmailtoaster.com




-
To unsubscribe, e-mail: qmailtoaster-list-unsubscr...@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-h...@qmailtoaster.com

Re: [qmailtoaster] qq soft reject errors on high load

[Remo Mattei]

1gb is not very much! 

Il giorno 03 set 2017, alle ore 11:54, Rajesh M <24x7ser...@24x7server.net> ha 
scritto:

eric

around 1 gb spare.

there are a lot of log files.

rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To: qmailtoaster-list@qmailtoaster.com
Sent: Sun, 3 Sep 2017 11:21:24 -0600
Subject: 

Rajesh,

How much  available space is on the /var/qmail/simscan directory?

Eric
> On 9/3/2017 12:19 AM, Rajesh M wrote:
> eric
> 
> i would like to conduct more load intensive tests.
> 
> is it possible to provide me the shell script that you use for this purpose.
> 
> rajesh


-
To unsubscribe, e-mail: qmailtoaster-list-unsubscr...@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-h...@qmailtoaster.com


-
To unsubscribe, e-mail: qmailtoaster-list-unsubscr...@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-h...@qmailtoaster.com

-
To unsubscribe, e-mail: qmailtoaster-list-unsubscr...@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-h...@qmailtoaster.com

Re: [qmailtoaster] qq soft reject errors on high load

[Eric Broch]

You got the script?


On 9/3/2017 12:57 PM, Rajesh M wrote:

eric

yes. the idea is to use my backup server to run the script and send a large 
volume of email to the production server in the night to a single email account.

rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To: 24x7ser...@24x7server.net
Sent: Sun, 3 Sep 2017 08:25:03 -0600
Subject:

Hi Rajesh,

I may have misunderstood your question. Are you requiring the script I
was using to load test my server? The one I was using to send myself email?

Or was it a script to monitor performance you were looking for?

Eric


On 9/3/2017 12:23 AM, Eric Broch wrote:

Sorry, Rajesh, I don't have one...I usually depend on the folk on line.


On 9/3/2017 12:19 AM, Rajesh M wrote:

eric

i would like to conduct more load intensive tests.

is it possible to provide me the shell script that you use for this purpose.

rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To:qmailtoaster-list@qmailtoaster.com
Sent: Sat, 2 Sep 2017 14:04:36 -0600
Subject:

My testing was at about 1/2 that rate with 320K attachments, almost
double that rate with 1/2 email 320K attachments, and 1/2 email with no
attachments. No soft rejects. Ugh!...difficult to troubleshoot if not
repeatable.


On 9/2/2017 1:36 PM, Rajesh M wrote:

eric

thanks for the info. the script you sent works

the server processes roughly around 8 emails per day during week days and 
around 80 percent of it comes during around 12 hours. So during peak hours it 
is around 8000 emails per hour.

i will test this out monday morning peak hours and revert.

thanks
rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To:qmailtoaster-list@qmailtoaster.com
Sent: Sat, 2 Sep 2017 12:35:14 -0600
Subject:

Rajesh,

I'm still not sure the process limit is being reached.

I'm testing a COS6/QMT server now and haven't been able to bring about
'qq soft reject' or failure of any sort. I'm hitting it with email using
a delivery script from two servers with attachment size of 320KB.

try the script below (psmem) to monitor:



#!/bin/bash
ps -C $1 -O rss | gawk '{ count ++; sum += $2 }; END {count --; print
"Number of processes =",count; print "Memory usage per process
=",sum/1024/count, "MB"; print "Total memory usage =", sum/1024, "MB" ;};'



# chmod 755 psmem

# watch -n 1 ./psmem simscan


Not sure I should change anything until we figure out what's going on.

Eric



On 9/2/2017 12:03 PM, Rajesh M wrote:

eric

i could not get the bash script working with either of the two options. still 
gives error
 line 23: `return-limits': not a valid identifier.



can the limits issue in simscan be fixed ?

thanks,
rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To:qmailtoaster-list@qmailtoaster.com
Sent: Fri, 1 Sep 2017 23:46:15 -0600
Subject:

Rajesh,

It has something to do with #!/bin/bash or #!/bin/sh.

At top of script, when I use #!/bin/sh I get the same error you do, but
when I use #!/bin/bash script works fine.


Eric


On 9/1/2017 11:07 PM, Rajesh M wrote:

eric

1) is their any resolution for this issue ?

2) if i run the script it throws error as such
.
/limits.sh: line 23: `return-limits': not a valid identifier

#!/bin/bash
if [ "$#" -ne "1" ]; then
echo ""
echo -e "\033[01;32mLimit checker\033[00m"
echo -e "\033[01;37mUsage:\033[01;33m $0 process_name\033[00m"
echo ""
exit 0
fi

return-limits(){
for process in $@; do
   process_pids=`ps -C $process -o pid --no-headers | cut -d " " -f 2`

   if [ -z $@ ]; then
  echo "[no $process running]"
   else
  for pid in $process_pids; do
 echo "[$process #$pid -- limits]"
 cat /proc/$pid/limits
   done
   fi
done
}
return-limits $1

rajesh



- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To:qmailtoaster-list@qmailtoaster.com
Sent: Fri, 1 Sep 2017 17:28:40 -0600
Subject:

I sent myself a large file so that I could examine simscan with the
below script (limits.sh). Here's the output:

[root@pet105 simscan-1.4.0]# ./limits simscan
[simscan #31535 -- limits]
Limit   
    
    
    
    
  Soft Limit  
    
    
 Hard Limit   

Re: [qmailtoaster] qq soft reject errors on high load

[Rajesh M]

eric

around 1 gb spare.

there are a lot of log files.

rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To: qmailtoaster-list@qmailtoaster.com
Sent: Sun, 3 Sep 2017 11:21:24 -0600
Subject:

Rajesh,

How much  available space is on the /var/qmail/simscan directory?

Eric
On 9/3/2017 12:19 AM, Rajesh M wrote:
> eric
>
> i would like to conduct more load intensive tests.
>
> is it possible to provide me the shell script that you use for this purpose.
>
> rajesh


-
To unsubscribe, e-mail: qmailtoaster-list-unsubscr...@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-h...@qmailtoaster.com


-
To unsubscribe, e-mail: qmailtoaster-list-unsubscr...@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-h...@qmailtoaster.com

Re: [qmailtoaster] qq soft reject errors on high load

[Eric Broch]

Rajesh,

How much  available space is on the /var/qmail/simscan directory?

Eric
On 9/3/2017 12:19 AM, Rajesh M wrote:

eric

i would like to conduct more load intensive tests.

is it possible to provide me the shell script that you use for this purpose.

rajesh



-
To unsubscribe, e-mail: qmailtoaster-list-unsubscr...@qmailtoaster.com
For additional commands, e-mail: qmailtoaster-list-h...@qmailtoaster.com

Re: [qmailtoaster] qq soft reject errors on high load

[Rajesh M]

eric

i would like to conduct more load intensive tests.

is it possible to provide me the shell script that you use for this purpose.

rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To: qmailtoaster-list@qmailtoaster.com
Sent: Sat, 2 Sep 2017 14:04:36 -0600
Subject:

My testing was at about 1/2 that rate with 320K attachments, almost
double that rate with 1/2 email 320K attachments, and 1/2 email with no
attachments. No soft rejects. Ugh!...difficult to troubleshoot if not
repeatable.


On 9/2/2017 1:36 PM, Rajesh M wrote:
> eric
>
> thanks for the info. the script you sent works
>
> the server processes roughly around 8 emails per day during week days and 
> around 80 percent of it comes during around 12 hours. So during peak hours it 
> is around 8000 emails per hour.
>
> i will test this out monday morning peak hours and revert.
>
> thanks
> rajesh
>
> - Original Message -
> From: Eric Broch [mailto:ebr...@whitehorsetc.com]
> To: qmailtoaster-list@qmailtoaster.com
> Sent: Sat, 2 Sep 2017 12:35:14 -0600
> Subject:
>
> Rajesh,
>
> I'm still not sure the process limit is being reached.
>
> I'm testing a COS6/QMT server now and haven't been able to bring about
> 'qq soft reject' or failure of any sort. I'm hitting it with email using
> a delivery script from two servers with attachment size of 320KB.
>
> try the script below (psmem) to monitor:
>
> 
>
> #!/bin/bash
> ps -C $1 -O rss | gawk '{ count ++; sum += $2 }; END {count --; print
> "Number of processes =",count; print "Memory usage per process
> =",sum/1024/count, "MB"; print "Total memory usage =", sum/1024, "MB" ;};'
>
> 
>
> # chmod 755 psmem
>
> # watch -n 1 ./psmem simscan
>
>
> Not sure I should change anything until we figure out what's going on.
>
> Eric
>
>
>
> On 9/2/2017 12:03 PM, Rajesh M wrote:
>> eric
>>
>> i could not get the bash script working with either of the two options. 
>> still gives error
>>line 23: `return-limits': not a valid identifier.
>>
>>
>>
>> can the limits issue in simscan be fixed ?
>>
>> thanks,
>> rajesh
>>
>> - Original Message -
>> From: Eric Broch [mailto:ebr...@whitehorsetc.com]
>> To:qmailtoaster-list@qmailtoaster.com
>> Sent: Fri, 1 Sep 2017 23:46:15 -0600
>> Subject:
>>
>> Rajesh,
>>
>> It has something to do with #!/bin/bash or #!/bin/sh.
>>
>> At top of script, when I use #!/bin/sh I get the same error you do, but
>> when I use #!/bin/bash script works fine.
>>
>>
>> Eric
>>
>>
>> On 9/1/2017 11:07 PM, Rajesh M wrote:
>>> eric
>>>
>>> 1) is their any resolution for this issue ?
>>>
>>> 2) if i run the script it throws error as such
>>> .
>>> /limits.sh: line 23: `return-limits': not a valid identifier
>>>
>>> #!/bin/bash
>>> if [ "$#" -ne "1" ]; then
>>>   echo ""
>>>   echo -e "\033[01;32mLimit checker\033[00m"
>>>   echo -e "\033[01;37mUsage:\033[01;33m $0 process_name\033[00m"
>>>   echo ""
>>>   exit 0
>>> fi
>>>
>>> return-limits(){
>>>   for process in $@; do
>>>  process_pids=`ps -C $process -o pid --no-headers | cut -d " " -f 2`
>>>
>>>  if [ -z $@ ]; then
>>> echo "[no $process running]"
>>>  else
>>> for pid in $process_pids; do
>>>echo "[$process #$pid -- limits]"
>>>cat /proc/$pid/limits
>>>  done
>>>  fi
>>>   done
>>> }
>>> return-limits $1
>>>
>>> rajesh
>>>
>>>
>>>
>>> - Original Message -
>>> From: Eric Broch [mailto:ebr...@whitehorsetc.com]
>>> To:qmailtoaster-list@qmailtoaster.com
>>> Sent: Fri, 1 Sep 2017 17:28:40 -0600
>>> Subject:
>>>
>>> I sent myself a large file so that I could examine simscan with the
>>> below script (limits.sh). Here's the output:
>>>
>>> [root@pet105 simscan-1.4.0]# ./limits simscan
>>> [simscan #31535 -- limits]
>>> Limit   
>>>    
>>>    
>>>    
>>>    
>>>    
>>>    Soft Limit 
>>>    
>>>    
>>>     Hard 
>>> Limit   
>>>    
>>>    
>>>   Units
>>> Max cpu time   
>>>    
>>>    
>>>    

RE: Fwd: Re: [qmailtoaster] qq soft reject errors on high load

[Rajesh M]

eric

thanks for the info. the script you sent works

the server processes roughly around 8 emails per day during week days and 
around 80 percent of it comes during around 12 hours. So during peak hours it 
is around 8000 emails per hour.

i will test this out monday morning peak hours and revert.

thanks
rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To: qmailtoaster-list@qmailtoaster.com
Sent: Sat, 2 Sep 2017 12:35:14 -0600
Subject:

Rajesh,

I'm still not sure the process limit is being reached.

I'm testing a COS6/QMT server now and haven't been able to bring about
'qq soft reject' or failure of any sort. I'm hitting it with email using
a delivery script from two servers with attachment size of 320KB.

try the script below (psmem) to monitor:



#!/bin/bash
ps -C $1 -O rss | gawk '{ count ++; sum += $2 }; END {count --; print
"Number of processes =",count; print "Memory usage per process
=",sum/1024/count, "MB"; print "Total memory usage =", sum/1024, "MB" ;};'



# chmod 755 psmem

# watch -n 1 ./psmem simscan


Not sure I should change anything until we figure out what's going on.

Eric



On 9/2/2017 12:03 PM, Rajesh M wrote:
> eric
>
> i could not get the bash script working with either of the two options. still 
> gives error
>   line 23: `return-limits': not a valid identifier.
>
>
>
> can the limits issue in simscan be fixed ?
>
> thanks,
> rajesh
>
> - Original Message -
> From: Eric Broch [mailto:ebr...@whitehorsetc.com]
> To:qmailtoaster-list@qmailtoaster.com
> Sent: Fri, 1 Sep 2017 23:46:15 -0600
> Subject:
>
> Rajesh,
>
> It has something to do with #!/bin/bash or #!/bin/sh.
>
> At top of script, when I use #!/bin/sh I get the same error you do, but
> when I use #!/bin/bash script works fine.
>
>
> Eric
>
>
> On 9/1/2017 11:07 PM, Rajesh M wrote:
>> eric
>>
>> 1) is their any resolution for this issue ?
>>
>> 2) if i run the script it throws error as such
>> .
>> /limits.sh: line 23: `return-limits': not a valid identifier
>>
>> #!/bin/bash
>> if [ "$#" -ne "1" ]; then
>>  echo ""
>>  echo -e "\033[01;32mLimit checker\033[00m"
>>  echo -e "\033[01;37mUsage:\033[01;33m $0 process_name\033[00m"
>>  echo ""
>>  exit 0
>> fi
>>
>> return-limits(){
>>  for process in $@; do
>> process_pids=`ps -C $process -o pid --no-headers | cut -d " " -f 2`
>>
>> if [ -z $@ ]; then
>>echo "[no $process running]"
>> else
>>for pid in $process_pids; do
>>   echo "[$process #$pid -- limits]"
>>   cat /proc/$pid/limits
>> done
>> fi
>>  done
>> }
>> return-limits $1
>>
>> rajesh
>>
>>
>>
>> - Original Message -
>> From: Eric Broch [mailto:ebr...@whitehorsetc.com]
>> To:qmailtoaster-list@qmailtoaster.com
>> Sent: Fri, 1 Sep 2017 17:28:40 -0600
>> Subject:
>>
>> I sent myself a large file so that I could examine simscan with the
>> below script (limits.sh). Here's the output:
>>
>> [root@pet105 simscan-1.4.0]# ./limits simscan
>> [simscan #31535 -- limits]
>> Limit        
>>         
>>      Soft Limit   
>>         Hard 
>> Limit        
>>    Units
>> Max cpu time       
>>        unlimited 
>> unlimited       
>>      seconds
>> Max file size       
>>       unlimited unlimited 
>>         
>>    bytes
>> Max data size       
>>       unlimited unlimited 
>>         
>>    bytes
>> Max stack size 10485760 unlimited    
>>         bytes
>> Max core file size      
>>   0 unlimited      
>>       bytes
>> Max resident set      
>>     unlimited unlimited   
>>          
>> bytes
>> Max processes 1024 31121     
>>     

Fwd: Re: [qmailtoaster] qq soft reject errors on high load

[Eric Broch]

Rajesh,

I'm still not sure the process limit is being reached.

I'm testing a COS6/QMT server now and haven't been able to bring about 
'qq soft reject' or failure of any sort. I'm hitting it with email using 
a delivery script from two servers with attachment size of 320KB.


try the script below (psmem) to monitor:



#!/bin/bash
ps -C $1 -O rss | gawk '{ count ++; sum += $2 }; END {count --; print 
"Number of processes =",count; print "Memory usage per process 
=",sum/1024/count, "MB"; print "Total memory usage =", sum/1024, "MB" ;};'




# chmod 755 psmem

# watch -n 1 ./psmem simscan


Not sure I should change anything until we figure out what's going on.

Eric



On 9/2/2017 12:03 PM, Rajesh M wrote:

eric

i could not get the bash script working with either of the two options. still 
gives error
  line 23: `return-limits': not a valid identifier.



can the limits issue in simscan be fixed ?

thanks,
rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To:qmailtoaster-list@qmailtoaster.com
Sent: Fri, 1 Sep 2017 23:46:15 -0600
Subject:

Rajesh,

It has something to do with #!/bin/bash or #!/bin/sh.

At top of script, when I use #!/bin/sh I get the same error you do, but
when I use #!/bin/bash script works fine.


Eric


On 9/1/2017 11:07 PM, Rajesh M wrote:

eric

1) is their any resolution for this issue ?

2) if i run the script it throws error as such
.
/limits.sh: line 23: `return-limits': not a valid identifier

#!/bin/bash
if [ "$#" -ne "1" ]; then
 echo ""
 echo -e "\033[01;32mLimit checker\033[00m"
 echo -e "\033[01;37mUsage:\033[01;33m $0 process_name\033[00m"
 echo ""
 exit 0
fi

return-limits(){
 for process in $@; do
process_pids=`ps -C $process -o pid --no-headers | cut -d " " -f 2`

if [ -z $@ ]; then
   echo "[no $process running]"
else
   for pid in $process_pids; do
  echo "[$process #$pid -- limits]"
  cat /proc/$pid/limits
done
fi
 done
}
return-limits $1

rajesh



- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To:qmailtoaster-list@qmailtoaster.com
Sent: Fri, 1 Sep 2017 17:28:40 -0600
Subject:

I sent myself a large file so that I could examine simscan with the
below script (limits.sh). Here's the output:

[root@pet105 simscan-1.4.0]# ./limits simscan
[simscan #31535 -- limits]
Limit                  
   Soft Limit           Hard Limit  
         Units
Max cpu time              unlimited 
unlimited            seconds
Max file size             unlimited 
unlimited            bytes
Max data size             unlimited 
unlimited            bytes
Max stack size 10485760 unlimited            
bytes
Max core file size        0 unlimited     
       bytes
Max resident set          unlimited unlimited  
          bytes
Max processes 1024 31121              
  processes
Max open files            1024 4096   
              files
Max locked memory         65536 65536    
            bytes
Max address space         unlimited unlimited  
          bytes
Max file locks            unlimited 
unlimited            locks
Max pending signals       31121 31121      
          signals
Max msgqueue size         819200 819200    
           bytes
Max nice priority         0       
             0
Max realtime priority     0          
          0
Max realtime timeout      unlimited unlimited     
       us

Notice the 'Max processes' and 'Max stack size', this could very well be
the issue for our simscan failures.

Eric

On 9/1/2017 5:20 PM, Eric Broch wrote:

This looks like a decent script:


#!/bin/bash
if [ "$#" -ne "1" ]; then
echo ""
echo -e "\033[01;32mLimit checker\033[00m"
echo -e "\033[01;37mUsage:\033[01;33m $0 process_name\033[00m"
echo ""
exit 0
fi

return-limits(){
for process in $@; do
      

Re: [qmailtoaster] qq soft reject errors on high load

[Rajesh M]

eric

i could not get the bash script working with either of the two options. still 
gives error
 line 23: `return-limits': not a valid identifier.



can the limits issue in simscan be fixed ?

thanks,
rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To: qmailtoaster-list@qmailtoaster.com
Sent: Fri, 1 Sep 2017 23:46:15 -0600
Subject:

Rajesh,

It has something to do with #!/bin/bash or #!/bin/sh.

At top of script, when I use #!/bin/sh I get the same error you do, but
when I use #!/bin/bash script works fine.


Eric


On 9/1/2017 11:07 PM, Rajesh M wrote:
> eric
>
> 1) is their any resolution for this issue ?
>
> 2) if i run the script it throws error as such
> .
> /limits.sh: line 23: `return-limits': not a valid identifier
>
> #!/bin/bash
> if [ "$#" -ne "1" ]; then
> echo ""
> echo -e "\033[01;32mLimit checker\033[00m"
> echo -e "\033[01;37mUsage:\033[01;33m $0 process_name\033[00m"
> echo ""
> exit 0
> fi
>
> return-limits(){
> for process in $@; do
>process_pids=`ps -C $process -o pid --no-headers | cut -d " " -f 2`
>
>if [ -z $@ ]; then
>   echo "[no $process running]"
>else
>   for pid in $process_pids; do
>  echo "[$process #$pid -- limits]"
>  cat /proc/$pid/limits
>done
>fi
> done
> }
> return-limits $1
>
> rajesh
>
>
>
> - Original Message -
> From: Eric Broch [mailto:ebr...@whitehorsetc.com]
> To: qmailtoaster-list@qmailtoaster.com
> Sent: Fri, 1 Sep 2017 17:28:40 -0600
> Subject:
>
> I sent myself a large file so that I could examine simscan with the
> below script (limits.sh). Here's the output:
>
> [root@pet105 simscan-1.4.0]# ./limits simscan
> [simscan #31535 -- limits]
> Limit                  
>    Soft Limit           Hard Limit  
>          Units
> Max cpu time              unlimited 
> unlimited            seconds
> Max file size             unlimited 
> unlimited            bytes
> Max data size             unlimited 
> unlimited            bytes
> Max stack size 10485760 unlimited            
> bytes
> Max core file size        0 unlimited     
>        bytes
> Max resident set          unlimited unlimited 
>            bytes
> Max processes 1024 31121             
>    processes
> Max open files            1024 4096  
>                files
> Max locked memory         65536 65536    
>             bytes
> Max address space         unlimited unlimited  
>           bytes
> Max file locks            unlimited 
> unlimited            locks
> Max pending signals       31121 31121     
>            signals
> Max msgqueue size         819200 819200   
>             bytes
> Max nice priority         0      
>               0
> Max realtime priority     0         
>            0
> Max realtime timeout      unlimited unlimited    
>         us
>
> Notice the 'Max processes' and 'Max stack size', this could very well be
> the issue for our simscan failures.
>
> Eric
>
> On 9/1/2017 5:20 PM, Eric Broch wrote:
>> This looks like a decent script:
>>
>> 
>> #!/bin/bash
>> if [ "$#" -ne "1" ]; then
>> echo ""
>> echo -e "\033[01;32mLimit checker\033[00m"
>> echo -e "\033[01;37mUsage:\033[01;33m $0 process_name\033[00m"
>> echo ""
>> exit 0
>> fi
>>
>> return-limits(){
>> for process in $@; do
>>       process_pids=`ps -C $process -o pid --no-headers | cut -d " " 
>> -f 2`
>>
>> if [ -z $@ ]; then
>> echo "[no $process running]"
>> else
>> for pid in $process_pids; do
>> echo "[$process #$pid -- limits]"
>>             cat /proc/$pid/limits
>>       done
>>       fi
>>    done
>> }
>>
>> return-limits $1
>> 
>>
>> Example:
>> # ./limits.sh tcpserver
>> or
>> # ./limits.sh simscan
>>
>> Eric
>>
>>
>> On 9/1/2017 2:51 PM, Rajesh M wrote:
>>> eric
>>>
>>> how do i check the number of user 

Re: [qmailtoaster] qq soft reject errors on high load

[Eric Broch]

Rajesh,

If you've got the script working you can add a line above the call to 
'return_limits':  echo "Count: $1 (`pgrep $1  | wc -l`)"


This line will give you the number of the argument's ($1) processes as 
well as the rest of the output.




#!/bin/bash
if [ "$#" -ne "1" ]; then
   echo ""
   echo -e "\033[01;32mLimit checker\033[00m"
   echo -e "\033[01;37mUsage:\033[01;33m $0 process_name\033[00m"
   echo ""
   exit 0
fi

return-limits(){
   for process in $@; do
  process_pids=`ps -C $process -o pid --no-headers | cut -d " " -f 2`

  if [ -z $@ ]; then
 echo "[no $process running]"
  else
 for pid in $process_pids; do
    echo "[$process #$pid -- limits]"
    cat /proc/$pid/limits
  done
  fi
   done
}

echo "Count: $1 (`pgrep $1  | wc -l`)"
return-limits $1



Run the script with the 'watch' command (below) and you can examine the 
number of simscan processes on you server.


# watch -n 1 ./limits.sh simscan

Eric


On 9/1/2017 11:46 PM, Eric Broch wrote:


Rajesh,

It has something to do with #!/bin/bash or #!/bin/sh.

At top of script, when I use #!/bin/sh I get the same error you do, 
but when I use #!/bin/bash script works fine.



Eric


On 9/1/2017 11:07 PM, Rajesh M wrote:

eric

1) is their any resolution for this issue ?

2) if i run the script it throws error as such
.
/limits.sh: line 23: `return-limits': not a valid identifier

#!/bin/bash
if [ "$#" -ne "1" ]; then
echo ""
echo -e "\033[01;32mLimit checker\033[00m"
echo -e "\033[01;37mUsage:\033[01;33m $0 process_name\033[00m"
echo ""
exit 0
fi

return-limits(){
for process in $@; do
   process_pids=`ps -C $process -o pid --no-headers | cut -d " " -f 2`

   if [ -z $@ ]; then
  echo "[no $process running]"
   else
  for pid in $process_pids; do
 echo "[$process #$pid -- limits]"
 cat /proc/$pid/limits
   done
   fi
done
}
return-limits $1

rajesh



- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To:qmailtoaster-list@qmailtoaster.com
Sent: Fri, 1 Sep 2017 17:28:40 -0600
Subject:

I sent myself a large file so that I could examine simscan with the
below script (limits.sh). Here's the output:

[root@pet105 simscan-1.4.0]# ./limits simscan
[simscan #31535 -- limits]
Limit                     Soft Limit           
Hard Limit           Units
Max cpu time              unlimited unlimited           
 seconds
Max file size             unlimited unlimited            
bytes
Max data size             unlimited unlimited            
bytes
Max stack size 10485760 unlimited            bytes
Max core file size        0 unlimited            bytes
Max resident set          unlimited unlimited            
bytes
Max processes 1024 31121                processes
Max open files            1024 4096                 
files
Max locked memory         65536 65536                
bytes
Max address space         unlimited unlimited            
bytes
Max file locks            unlimited unlimited            
locks
Max pending signals       31121 31121                
signals
Max msgqueue size         819200 819200               
bytes
Max nice priority         0                    0
Max realtime priority     0                    0
Max realtime timeout      unlimited unlimited            us

Notice the 'Max processes' and 'Max stack size', this could very well be
the issue for our simscan failures.

Eric

On 9/1/2017 5:20 PM, Eric Broch wrote:

This looks like a decent script:


#!/bin/bash
if [ "$#" -ne "1" ]; then
echo ""
echo -e "\033[01;32mLimit checker\033[00m"
echo -e "\033[01;37mUsage:\033[01;33m $0 process_name\033[00m"
echo ""
exit 0
fi

return-limits(){
for process in $@; do
      process_pids=`ps -C $process -o pid --no-headers | cut -d " " -f 2`

if [ -z $@ ]; then
echo "[no $process running]"
else
for pid in $process_pids; do
echo "[$process #$pid -- limits]"
            cat /proc/$pid/limits
      done
      fi
   done
}

return-limits $1


Example:
# ./limits.sh tcpserver
or
# ./limits.sh simscan

Eric


On 9/1/2017 2:51 PM, Rajesh M wrote:

eric

how do i check the number of user processes being consumed and by which user

rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To:qmailtoaster-list@qmailtoaster.com
Sent: Fri, 1 Sep 2017 14:25:36 -0600
Subject:

Number of user processes.


On 9/1/2017 1:57 PM, Rajesh M wrote:

eric

what is RLIMIT ? is it related to memory ?

rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]

Re: [qmailtoaster] qq soft reject errors on high load

[Eric Broch]

Rajesh,

It has something to do with #!/bin/bash or #!/bin/sh.

At top of script, when I use #!/bin/sh I get the same error you do, but 
when I use #!/bin/bash script works fine.



Eric


On 9/1/2017 11:07 PM, Rajesh M wrote:

eric

1) is their any resolution for this issue ?

2) if i run the script it throws error as such
.
/limits.sh: line 23: `return-limits': not a valid identifier

#!/bin/bash
if [ "$#" -ne "1" ]; then
echo ""
echo -e "\033[01;32mLimit checker\033[00m"
echo -e "\033[01;37mUsage:\033[01;33m $0 process_name\033[00m"
echo ""
exit 0
fi

return-limits(){
for process in $@; do
   process_pids=`ps -C $process -o pid --no-headers | cut -d " " -f 2`

   if [ -z $@ ]; then
  echo "[no $process running]"
   else
  for pid in $process_pids; do
 echo "[$process #$pid -- limits]"
 cat /proc/$pid/limits
   done
   fi
done
}
return-limits $1

rajesh



- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To: qmailtoaster-list@qmailtoaster.com
Sent: Fri, 1 Sep 2017 17:28:40 -0600
Subject:

I sent myself a large file so that I could examine simscan with the
below script (limits.sh). Here's the output:

[root@pet105 simscan-1.4.0]# ./limits simscan
[simscan #31535 -- limits]
Limit                     Soft Limit           
Hard Limit           Units
Max cpu time              unlimited unlimited           
 seconds
Max file size             unlimited unlimited            
bytes
Max data size             unlimited unlimited            
bytes
Max stack size 10485760 unlimited            bytes
Max core file size        0 unlimited            bytes
Max resident set          unlimited unlimited            
bytes
Max processes 1024 31121                processes
Max open files            1024 4096                 
files
Max locked memory         65536 65536                
bytes
Max address space         unlimited unlimited            
bytes
Max file locks            unlimited unlimited            
locks
Max pending signals       31121 31121                
signals
Max msgqueue size         819200 819200               
bytes
Max nice priority         0                    0
Max realtime priority     0                    0
Max realtime timeout      unlimited unlimited            us

Notice the 'Max processes' and 'Max stack size', this could very well be
the issue for our simscan failures.

Eric

On 9/1/2017 5:20 PM, Eric Broch wrote:

This looks like a decent script:


#!/bin/bash
if [ "$#" -ne "1" ]; then
echo ""
echo -e "\033[01;32mLimit checker\033[00m"
echo -e "\033[01;37mUsage:\033[01;33m $0 process_name\033[00m"
echo ""
exit 0
fi

return-limits(){
for process in $@; do
      process_pids=`ps -C $process -o pid --no-headers | cut -d " " -f 2`

if [ -z $@ ]; then
echo "[no $process running]"
else
for pid in $process_pids; do
echo "[$process #$pid -- limits]"
            cat /proc/$pid/limits
      done
      fi
   done
}

return-limits $1


Example:
# ./limits.sh tcpserver
or
# ./limits.sh simscan

Eric


On 9/1/2017 2:51 PM, Rajesh M wrote:

eric

how do i check the number of user processes being consumed and by which user

rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To:qmailtoaster-list@qmailtoaster.com
Sent: Fri, 1 Sep 2017 14:25:36 -0600
Subject:

Number of user processes.


On 9/1/2017 1:57 PM, Rajesh M wrote:

eric

what is RLIMIT ? is it related to memory ?

rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To:qmailtoaster-list@qmailtoaster.com
Sent: Fri, 1 Sep 2017 13:25:49 -0600
Subject:

Here's a section of code in simscan.c that controls the number of processes.

#ifdef HAS_ULIMIT_NPROC
  /* Set ulimits to prevent hangs if it forks too many processes */
  getrlimit(RLIMIT_NPROC, );
  limits.rlim_cur = 1024;
  setrlimit(RLIMIT_NPROC, );
#endif

The RLIMIT is set to 1024.

I think this may be the issue and would explain why there would be
forking issues with qmail-queue, ripmime, and others. This is in place
to prevent hangs...could be a good thing.


On 9/1/2017 1:04 PM, Remo Mattei wrote:

interesting...

mine was an easy fix by changing the owner and permissions so
underline has some other triggering to the kernel and how it reads
those layers.

On 9/1/17 11:18 AM, Eric Broch wrote:

1)  Problem forking

vfork() is used to execute qmail-queue

The error 'simscan: error forking qmail-queue' will happen for either
of the following reasons:

a) [EAGAIN]
   The 

Re: [qmailtoaster] qq soft reject errors on high load

[Rajesh M]

eric

1) is their any resolution for this issue ?

2) if i run the script it throws error as such
.
/limits.sh: line 23: `return-limits': not a valid identifier

#!/bin/bash
if [ "$#" -ne "1" ]; then
   echo ""
   echo -e "\033[01;32mLimit checker\033[00m"
   echo -e "\033[01;37mUsage:\033[01;33m $0 process_name\033[00m"
   echo ""
   exit 0
fi

return-limits(){
   for process in $@; do
  process_pids=`ps -C $process -o pid --no-headers | cut -d " " -f 2`

  if [ -z $@ ]; then
 echo "[no $process running]"
  else
 for pid in $process_pids; do
echo "[$process #$pid -- limits]"
cat /proc/$pid/limits
  done
  fi
   done
}
return-limits $1

rajesh



- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To: qmailtoaster-list@qmailtoaster.com
Sent: Fri, 1 Sep 2017 17:28:40 -0600
Subject:

I sent myself a large file so that I could examine simscan with the
below script (limits.sh). Here's the output:

[root@pet105 simscan-1.4.0]# ./limits simscan
[simscan #31535 -- limits]
Limit                     Soft Limit           
Hard Limit           Units
Max cpu time              unlimited unlimited           
 seconds
Max file size             unlimited unlimited            
bytes
Max data size             unlimited unlimited            
bytes
Max stack size 10485760 unlimited            bytes
Max core file size        0 unlimited            bytes
Max resident set          unlimited unlimited            
bytes
Max processes 1024 31121                processes
Max open files            1024 4096                 
files
Max locked memory         65536 65536                
bytes
Max address space         unlimited unlimited            
bytes
Max file locks            unlimited unlimited            
locks
Max pending signals       31121 31121                
signals
Max msgqueue size         819200 819200               
bytes
Max nice priority         0                    0
Max realtime priority     0                    0
Max realtime timeout      unlimited unlimited            us

Notice the 'Max processes' and 'Max stack size', this could very well be
the issue for our simscan failures.

Eric

On 9/1/2017 5:20 PM, Eric Broch wrote:
>
> This looks like a decent script:
>
> 
> #!/bin/bash
> if [ "$#" -ne "1" ]; then
> echo ""
> echo -e "\033[01;32mLimit checker\033[00m"
> echo -e "\033[01;37mUsage:\033[01;33m $0 process_name\033[00m"
> echo ""
> exit 0
> fi
>
> return-limits(){
> for process in $@; do
>       process_pids=`ps -C $process -o pid --no-headers | cut -d " " -f 2`
>
> if [ -z $@ ]; then
> echo "[no $process running]"
> else
> for pid in $process_pids; do
> echo "[$process #$pid -- limits]"
>             cat /proc/$pid/limits
>       done
>       fi
>    done
> }
>
> return-limits $1
> 
>
> Example:
> # ./limits.sh tcpserver
> or
> # ./limits.sh simscan
>
> Eric
>
>
> On 9/1/2017 2:51 PM, Rajesh M wrote:
>> eric
>>
>> how do i check the number of user processes being consumed and by which user
>>
>> rajesh
>>
>> - Original Message -
>> From: Eric Broch [mailto:ebr...@whitehorsetc.com]
>> To:qmailtoaster-list@qmailtoaster.com
>> Sent: Fri, 1 Sep 2017 14:25:36 -0600
>> Subject:
>>
>> Number of user processes.
>>
>>
>> On 9/1/2017 1:57 PM, Rajesh M wrote:
>>> eric
>>>
>>> what is RLIMIT ? is it related to memory ?
>>>
>>> rajesh
>>>
>>> - Original Message -
>>> From: Eric Broch [mailto:ebr...@whitehorsetc.com]
>>> To:qmailtoaster-list@qmailtoaster.com
>>> Sent: Fri, 1 Sep 2017 13:25:49 -0600
>>> Subject:
>>>
>>> Here's a section of code in simscan.c that controls the number of processes.
>>>
>>> #ifdef HAS_ULIMIT_NPROC
>>>  /* Set ulimits to prevent hangs if it forks too many processes 
>>> */
>>>  getrlimit(RLIMIT_NPROC, );
>>>  limits.rlim_cur = 1024;
>>>  setrlimit(RLIMIT_NPROC, );
>>> #endif
>>>
>>> The RLIMIT is set to 1024.
>>>
>>> I think this may be the issue and would explain why there would be
>>> forking issues with qmail-queue, ripmime, and others. This is in place
>>> to prevent hangs...could be a good thing.
>>>
>>>
>>> On 9/1/2017 1:04 PM, Remo Mattei wrote:
 interesting...

 mine was an easy fix by changing the owner and permissions so
 underline has some other triggering to the kernel and how it reads
 those layers.

 On 9/1/17 11:18 AM, Eric Broch wrote:
> 1)  Problem forking
>
> vfork() is used to execute qmail-queue
>
> The error 'simscan: error forking qmail-queue' will happen for either
> of the 

Re: [qmailtoaster] qq soft reject errors on high load

[Jeff Koch]

Hi Rajesh:

Intel(R) Xeon(R) CPU E3-1230 V2

ASUS P8B-M,  16GB RAM



On 9/1/2017 3:53 PM, Rajesh M wrote:

jeff

could you please let me know the cpu details of your machine

is it a dell machine or some other with  intel E5 processor ?

rajesh


- Original Message -
From: Jeff Koch [mailto:jeffk...@intersessions.com]
To: qmailtoaster-list@qmailtoaster.com
Sent: Fri, 1 Sep 2017 13:39:37 -0400
Subject:

Eric / Remo / Rajesh - we reported this same problem a number of months
ago and thought we had fixed it. But alas, it's still with us. The good
news is that we haven't had any customer complaints. I'll be very
interested if it can be resolved.

Jeff Koch


On 9/1/2017 12:01 PM, Eric Broch wrote:

I'm not sure what's going on here. Is this a relatively new phenomenon?

I wonder if it's a memory, or even a disk speed, issue since it only
happens at peak hours?

I think I'll appy Johannes Weberhofer's patch and put it out there for
you...and cross or fingers.


On 9/1/2017 6:28 AM, Rajesh M wrote:

remo / eric

i have still not being able to resolve the qq soft reject error.

these are my findings

1) the errors i see are "error forking qmail-queue" and "ripmime error" which 
causes the qq soft reject.

2) the max concurrent connections in the logs is around 35.

3)  These errors come up during peak working hours when the server is under a  
load of 4 - 8, and they increase the load even more to over 10-15.

4) i came across this link (not sure if this is related)
https://github.com/qmail/simscan/blob/master/simscan.c

5) i can share with you my live smtp logs with simscan debug.

have extracted some lines below


Error forking qmail-queue

@400059a8fa7b0a2ed1b4 tcpserver: status: 31/200
@400059a8fa7b13162584 simscan: cdb looking up version spam
@400059a8fa7b13166bd4 simscan: runned_scanners is  attach: 1.4.0 clamav: 
0.98.6/m: spam: 3.3.2
@400059a8fa7b13166fbc simscan: found 3.3.2
@400059a8fa7b13168efc simscan:[10757]:CLEAN (5.00/30.00):9.7712s:-Possible 
Spam- RE_ REQUIRE BOOKING // 1X20  //
ICD TKD TO 
BANDARABASS:103.241.181.228:cs@atlasdecargo.com:rathe...@radiant-india.net
@400059a8fa7b1316cd7c simscan: done, execing qmail-queue
@400059a8fa7b1316fc5c simscan: error forking qmail-queue
@400059a8fa7b13199854 simscan: exit error code: 71
@400059a8fa7b131c4004 qmail-smtpd: qq soft reject (mail server temporarily 
rejected message (#4.3.0)): MAILFROM:
RCPTTO:rathe...@y.net


Error in ripmime

@400059a8fa98045a4bc4 simscan: pelookup: domain is aissamaritime.in
@400059a8fa98045a4bc4 simscan: cdb looking up aissamaritime.in
@400059a8fa98045a4fac simscan: pelookup: local part is shailesh_k_bom
@400059a8fa98045a4fac simscan: lpart: local part is **
@400059a8fa98045a5394 simscan: cdb looking upshailesh_k_...@aissamaritime.in
@400059a8fa98045a5394 simscan: ripmime error
@400059a8fa98045a6334 simscan: exit error code: 71
@400059a8fa98045a95fc qmail-smtpd: qq soft reject (mail server temporarily 
rejected message (#4.3.0)):
MAILFROM:  RCPTTO:shailesh_k_...@yy.in


Wierd error logs

@400059a9032f3aa79a24 simscan: clamdscan: --- SCAN SUMMARY 
---
@400059a9032f3aa7b964 simscan: clamdscan: 
/var/qmail/simscan/1504248613.321653.5221: OK
@400059a9032f3aa86d14 simscan: clamdscan:
@400059a9032f3aa870fc simscan: clamdscan: ---simscan: cdb looking up 
version clamav
@400059a9032f3aa8a3c4  SCAN simscan: clamdscan: SIUnMfMeAcRtYe d- 
-fsimscan: runned_scanners is  attach: 1.4.0 clamav: 0.98.6/m:
@400059a9032f3aa8c304 -isimscan: found 0.98.6/m:
@400059a9032f3aa8c6ec -l-e-s-:- -0
@400059a9032f3aa8f1e4 --simscan: normal clamdscan return code: 0
@400059a9032f3aa8f1e4
@400059a9032f3aa8f5cc simscan: clamdscan: Infected fsimscan: clamdscan: 
iTliemes::  00
@400059a9032f3aa93834 .simscan: clamdscan: 1T1i8m es:e c0 .1(002  ms 
e0simscan: calling spamc
@400059a9032f3aa96ee4 c  s(0)simscan: calling /usr/bin/spamc simscan: 
clamdscan:
@400059a9032f3aa999dc  spamcm
@400059a9032f3aa999dc 0 s)
@400059a9032f3aa9ad64 /var/qmail/simscan/1simscan: check_spam had an error 
ret: -1
@400059a9032f3aa9e02c 504248613.307311.5215: OK
@400059a9032f337c simscan: clamdscan:
@400059a9032f3764 simscan: clamdscan: --- SCAN SUMMARY 
---
@400059a9032f3aab3fbc simscan: clamdscan: Infected files: 0
@400059a9032f3aaba164 simscan: clamdscan: Time: 0.135 sec (0 m 0 s)
@400059a9032f3aac39bc simscan: clamdscan: /var/qmail/simscan/15simscan: 
exit error code: 71
@400059a9032f3aaca334 04248613.308469.5216: OK
@400059a9032f3aacddcc simscan: clamdscan:
@400059a9032f3aace984 simscan: clamdscan: --- SCAN SUMMARY 
---
@400059a9




@400059a9032f3aa86d14 simscan: clamdscan:
@400059a9032f3aa870fc simscan: clamdscan: ---simscan: cdb looking up 
version clamav

Re: [qmailtoaster] qq soft reject errors on high load

[Eric Broch]

I sent myself a large file so that I could examine simscan with the 
below script (limits.sh). Here's the output:


[root@pet105 simscan-1.4.0]# ./limits simscan
[simscan #31535 -- limits]
Limit Soft Limit   Hard Limit   Units
Max cpu time  unlimited unlimited    seconds
Max file size unlimited unlimited    bytes
Max data size unlimited unlimited    bytes
Max stack size 10485760 unlimited    bytes
Max core file size    0 unlimited    bytes
Max resident set  unlimited unlimited    bytes
Max processes 1024 31121    processes
Max open files    1024 4096 files
Max locked memory 65536 65536    bytes
Max address space unlimited unlimited    bytes
Max file locks    unlimited unlimited    locks
Max pending signals   31121 31121    signals
Max msgqueue size 819200 819200   bytes
Max nice priority 0    0
Max realtime priority 0    0
Max realtime timeout  unlimited unlimited    us

Notice the 'Max processes' and 'Max stack size', this could very well be 
the issue for our simscan failures.


Eric

On 9/1/2017 5:20 PM, Eric Broch wrote:


This looks like a decent script:


#!/bin/bash
if [ "$#" -ne "1" ]; then
echo ""
echo -e "\033[01;32mLimit checker\033[00m"
echo -e "\033[01;37mUsage:\033[01;33m $0 process_name\033[00m"
echo ""
exit 0
fi

return-limits(){
for process in $@; do
      process_pids=`ps -C $process -o pid --no-headers | cut -d " " -f 2`

if [ -z $@ ]; then
echo "[no $process running]"
else
for pid in $process_pids; do
echo "[$process #$pid -- limits]"
            cat /proc/$pid/limits
      done
      fi
   done
}

return-limits $1


Example:
# ./limits.sh tcpserver
or
# ./limits.sh simscan

Eric


On 9/1/2017 2:51 PM, Rajesh M wrote:

eric

how do i check the number of user processes being consumed and by which user

rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To:qmailtoaster-list@qmailtoaster.com
Sent: Fri, 1 Sep 2017 14:25:36 -0600
Subject:

Number of user processes.


On 9/1/2017 1:57 PM, Rajesh M wrote:

eric

what is RLIMIT ? is it related to memory ?

rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To:qmailtoaster-list@qmailtoaster.com
Sent: Fri, 1 Sep 2017 13:25:49 -0600
Subject:

Here's a section of code in simscan.c that controls the number of processes.

#ifdef HAS_ULIMIT_NPROC
     /* Set ulimits to prevent hangs if it forks too many processes */
     getrlimit(RLIMIT_NPROC, );
     limits.rlim_cur = 1024;
     setrlimit(RLIMIT_NPROC, );
#endif

The RLIMIT is set to 1024.

I think this may be the issue and would explain why there would be
forking issues with qmail-queue, ripmime, and others. This is in place
to prevent hangs...could be a good thing.


On 9/1/2017 1:04 PM, Remo Mattei wrote:

interesting...

mine was an easy fix by changing the owner and permissions so
underline has some other triggering to the kernel and how it reads
those layers.

On 9/1/17 11:18 AM, Eric Broch wrote:

1)  Problem forking

vfork() is used to execute qmail-queue

The error 'simscan: error forking qmail-queue' will happen for either
of the following reasons:

a) [EAGAIN]
  The system-wide limit on the total number of processes under
  execution would be exceeded, or the system-imposed limit on the
  total number of processes under execution by a single user would
  be exceeded.
b) [ENOMEM]
  There is insufficient swap space for the new process.


On 9/1/2017 11:53 AM, Eric Broch wrote:

Are there any errors in /var/log/messages that correspond?


On 9/1/2017 11:39 AM, Jeff Koch wrote:

Eric / Remo / Rajesh - we reported this same problem a number of
months ago and thought we had fixed it. But alas, it's still with
us. The good news is that we haven't had any customer complaints.
I'll be very interested if it can be resolved.

Jeff Koch


On 9/1/2017 12:01 PM, Eric Broch wrote:

I'm not sure what's going on here. Is this a relatively new
phenomenon?

I wonder if it's a memory, or even a disk speed, issue since it
only happens at peak hours?

I think I'll appy Johannes Weberhofer's patch and put it out there
for you...and cross or fingers.


On 9/1/2017 6:28 AM, Rajesh M wrote:

remo / eric

i have still not being able to resolve the qq soft reject error.

these are my findings

1) the errors i see are "error forking qmail-queue" and "ripmime error" which 
causes the qq soft reject.

2) the max concurrent connections in the logs is around 35.

3)  These errors come up during peak working hours when the server is under a  
load of 4 - 8, and they increase the load even more to over 10-15.

4) i came across this link (not sure if this is related)

Re: [qmailtoaster] qq soft reject errors on high load

[Eric Broch]

This looks like a decent script:


#!/bin/bash
if [ "$#" -ne "1" ]; then
echo ""
echo -e "\033[01;32mLimit checker\033[00m"
echo -e "\033[01;37mUsage:\033[01;33m $0 process_name\033[00m"
echo ""
exit 0
fi

return-limits(){
for process in $@; do
      process_pids=`ps -C $process -o pid --no-headers | cut -d " " -f 2`

if [ -z $@ ]; then
echo "[no $process running]"
else
for pid in $process_pids; do
echo "[$process #$pid -- limits]"
            cat /proc/$pid/limits
      done
      fi
   done
}

return-limits $1


Example:
# ./limits.sh tcpserver
or
# ./limits.sh simscan

Eric


On 9/1/2017 2:51 PM, Rajesh M wrote:

eric

how do i check the number of user processes being consumed and by which user

rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To: qmailtoaster-list@qmailtoaster.com
Sent: Fri, 1 Sep 2017 14:25:36 -0600
Subject:

Number of user processes.


On 9/1/2017 1:57 PM, Rajesh M wrote:

eric

what is RLIMIT ? is it related to memory ?

rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To: qmailtoaster-list@qmailtoaster.com
Sent: Fri, 1 Sep 2017 13:25:49 -0600
Subject:

Here's a section of code in simscan.c that controls the number of processes.

#ifdef HAS_ULIMIT_NPROC
     /* Set ulimits to prevent hangs if it forks too many processes */
     getrlimit(RLIMIT_NPROC, );
     limits.rlim_cur = 1024;
     setrlimit(RLIMIT_NPROC, );
#endif

The RLIMIT is set to 1024.

I think this may be the issue and would explain why there would be
forking issues with qmail-queue, ripmime, and others. This is in place
to prevent hangs...could be a good thing.


On 9/1/2017 1:04 PM, Remo Mattei wrote:

interesting...

mine was an easy fix by changing the owner and permissions so
underline has some other triggering to the kernel and how it reads
those layers.

On 9/1/17 11:18 AM, Eric Broch wrote:

1)  Problem forking

vfork() is used to execute qmail-queue

The error 'simscan: error forking qmail-queue' will happen for either
of the following reasons:

a) [EAGAIN]
  The system-wide limit on the total number of processes under
  execution would be exceeded, or the system-imposed limit on the
  total number of processes under execution by a single user would
  be exceeded.
b) [ENOMEM]
  There is insufficient swap space for the new process.


On 9/1/2017 11:53 AM, Eric Broch wrote:

Are there any errors in /var/log/messages that correspond?


On 9/1/2017 11:39 AM, Jeff Koch wrote:

Eric / Remo / Rajesh - we reported this same problem a number of
months ago and thought we had fixed it. But alas, it's still with
us. The good news is that we haven't had any customer complaints.
I'll be very interested if it can be resolved.

Jeff Koch


On 9/1/2017 12:01 PM, Eric Broch wrote:

I'm not sure what's going on here. Is this a relatively new
phenomenon?

I wonder if it's a memory, or even a disk speed, issue since it
only happens at peak hours?

I think I'll appy Johannes Weberhofer's patch and put it out there
for you...and cross or fingers.


On 9/1/2017 6:28 AM, Rajesh M wrote:

remo / eric

i have still not being able to resolve the qq soft reject error.

these are my findings

1) the errors i see are "error forking qmail-queue" and "ripmime error" which 
causes the qq soft reject.

2) the max concurrent connections in the logs is around 35.

3)  These errors come up during peak working hours when the server is under a  
load of 4 - 8, and they increase the load even more to over 10-15.

4) i came across this link (not sure if this is related)
https://github.com/qmail/simscan/blob/master/simscan.c

5) i can share with you my live smtp logs with simscan debug.

have extracted some lines below


Error forking qmail-queue

@400059a8fa7b0a2ed1b4 tcpserver: status: 31/200
@400059a8fa7b13162584 simscan: cdb looking up version spam
@400059a8fa7b13166bd4 simscan: runned_scanners is  attach: 1.4.0 clamav: 
0.98.6/m: spam: 3.3.2
@400059a8fa7b13166fbc simscan: found 3.3.2
@400059a8fa7b13168efc simscan:[10757]:CLEAN (5.00/30.00):9.7712s:-Possible 
Spam- RE_ REQUIRE BOOKING // 1X20  //
ICD TKD TO 
BANDARABASS:103.241.181.228:cs@atlasdecargo.com:rathe...@radiant-india.net
@400059a8fa7b1316cd7c simscan: done, execing qmail-queue
@400059a8fa7b1316fc5c simscan: error forking qmail-queue
@400059a8fa7b13199854 simscan: exit error code: 71
@400059a8fa7b131c4004 qmail-smtpd: qq soft reject (mail server temporarily 
rejected message (#4.3.0)): MAILFROM:
RCPTTO:rathe...@y.net


Error in ripmime

@400059a8fa98045a4bc4 simscan: pelookup: domain is aissamaritime.in
@400059a8fa98045a4bc4 simscan: cdb looking up aissamaritime.in
@400059a8fa98045a4fac simscan: pelookup: local part is shailesh_k_bom
@400059a8fa98045a4fac simscan: lpart: local part is **
@400059a8fa98045a5394 simscan: cdb looking upshailesh_k_...@aissamaritime.in
@400059a8fa98045a5394 simscan: 

Re: [qmailtoaster] qq soft reject errors on high load

[Rajesh M]

eric

how do i check the number of user processes being consumed and by which user

rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To: qmailtoaster-list@qmailtoaster.com
Sent: Fri, 1 Sep 2017 14:25:36 -0600
Subject:

Number of user processes.


On 9/1/2017 1:57 PM, Rajesh M wrote:
> eric
>
> what is RLIMIT ? is it related to memory ?
>
> rajesh
>
> - Original Message -
> From: Eric Broch [mailto:ebr...@whitehorsetc.com]
> To: qmailtoaster-list@qmailtoaster.com
> Sent: Fri, 1 Sep 2017 13:25:49 -0600
> Subject:
>
> Here's a section of code in simscan.c that controls the number of processes.
>
> #ifdef HAS_ULIMIT_NPROC
>     /* Set ulimits to prevent hangs if it forks too many processes */
>     getrlimit(RLIMIT_NPROC, );
>     limits.rlim_cur = 1024;
>     setrlimit(RLIMIT_NPROC, );
> #endif
>
> The RLIMIT is set to 1024.
>
> I think this may be the issue and would explain why there would be
> forking issues with qmail-queue, ripmime, and others. This is in place
> to prevent hangs...could be a good thing.
>
>
> On 9/1/2017 1:04 PM, Remo Mattei wrote:
>> interesting...
>>
>> mine was an easy fix by changing the owner and permissions so
>> underline has some other triggering to the kernel and how it reads
>> those layers.
>>
>> On 9/1/17 11:18 AM, Eric Broch wrote:
>>> 1)  Problem forking
>>>
>>> vfork() is used to execute qmail-queue
>>>
>>> The error 'simscan: error forking qmail-queue' will happen for either
>>> of the following reasons:
>>>
>>> a) [EAGAIN]
>>>  The system-wide limit on the total number of processes under
>>>  execution would be exceeded, or the system-imposed limit on the
>>>  total number of processes under execution by a single user would
>>>  be exceeded.
>>> b) [ENOMEM]
>>>  There is insufficient swap space for the new process.
>>>
>>>
>>> On 9/1/2017 11:53 AM, Eric Broch wrote:
 Are there any errors in /var/log/messages that correspond?


 On 9/1/2017 11:39 AM, Jeff Koch wrote:
> Eric / Remo / Rajesh - we reported this same problem a number of
> months ago and thought we had fixed it. But alas, it's still with
> us. The good news is that we haven't had any customer complaints.
> I'll be very interested if it can be resolved.
>
> Jeff Koch
>
>
> On 9/1/2017 12:01 PM, Eric Broch wrote:
>> I'm not sure what's going on here. Is this a relatively new
>> phenomenon?
>>
>> I wonder if it's a memory, or even a disk speed, issue since it
>> only happens at peak hours?
>>
>> I think I'll appy Johannes Weberhofer's patch and put it out there
>> for you...and cross or fingers.
>>
>>
>> On 9/1/2017 6:28 AM, Rajesh M wrote:
>>> remo / eric
>>>
>>> i have still not being able to resolve the qq soft reject error.
>>>
>>> these are my findings
>>>
>>> 1) the errors i see are "error forking qmail-queue" and "ripmime error" 
>>> which causes the qq soft reject.
>>>
>>> 2) the max concurrent connections in the logs is around 35.
>>>
>>> 3)  These errors come up during peak working hours when the server is 
>>> under a  load of 4 - 8, and they increase the load even more to over 
>>> 10-15.
>>>
>>> 4) i came across this link (not sure if this is related)
>>> https://github.com/qmail/simscan/blob/master/simscan.c
>>>
>>> 5) i can share with you my live smtp logs with simscan debug.
>>>
>>> have extracted some lines below
>>>
>>>
>>> Error forking qmail-queue
>>>
>>> @400059a8fa7b0a2ed1b4 tcpserver: status: 31/200
>>> @400059a8fa7b13162584 simscan: cdb looking up version spam
>>> @400059a8fa7b13166bd4 simscan: runned_scanners is  attach: 1.4.0 
>>> clamav: 0.98.6/m: spam: 3.3.2
>>> @400059a8fa7b13166fbc simscan: found 3.3.2
>>> @400059a8fa7b13168efc simscan:[10757]:CLEAN 
>>> (5.00/30.00):9.7712s:-Possible Spam- RE_ REQUIRE BOOKING // 1X20  //
>>> ICD TKD TO 
>>> BANDARABASS:103.241.181.228:cs@atlasdecargo.com:rathe...@radiant-india.net
>>> @400059a8fa7b1316cd7c simscan: done, execing qmail-queue
>>> @400059a8fa7b1316fc5c simscan: error forking qmail-queue
>>> @400059a8fa7b13199854 simscan: exit error code: 71
>>> @400059a8fa7b131c4004 qmail-smtpd: qq soft reject (mail server 
>>> temporarily rejected message (#4.3.0)): MAILFROM:
>>> RCPTTO:rathe...@y.net
>>>
>>>
>>> Error in ripmime
>>>
>>> @400059a8fa98045a4bc4 simscan: pelookup: domain is aissamaritime.in
>>> @400059a8fa98045a4bc4 simscan: cdb looking up aissamaritime.in
>>> @400059a8fa98045a4fac simscan: pelookup: local part is 
>>> shailesh_k_bom
>>> @400059a8fa98045a4fac simscan: lpart: local part is **
>>> @400059a8fa98045a5394 simscan: cdb looking 
>>> upshailesh_k_...@aissamaritime.in

Re: [qmailtoaster] qq soft reject errors on high load

[Eric Broch]

Number of user processes.


On 9/1/2017 1:57 PM, Rajesh M wrote:

eric

what is RLIMIT ? is it related to memory ?

rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To: qmailtoaster-list@qmailtoaster.com
Sent: Fri, 1 Sep 2017 13:25:49 -0600
Subject:

Here's a section of code in simscan.c that controls the number of processes.

#ifdef HAS_ULIMIT_NPROC
    /* Set ulimits to prevent hangs if it forks too many processes */
    getrlimit(RLIMIT_NPROC, );
    limits.rlim_cur = 1024;
    setrlimit(RLIMIT_NPROC, );
#endif

The RLIMIT is set to 1024.

I think this may be the issue and would explain why there would be
forking issues with qmail-queue, ripmime, and others. This is in place
to prevent hangs...could be a good thing.


On 9/1/2017 1:04 PM, Remo Mattei wrote:

interesting...

mine was an easy fix by changing the owner and permissions so
underline has some other triggering to the kernel and how it reads
those layers.

On 9/1/17 11:18 AM, Eric Broch wrote:

1)  Problem forking

vfork() is used to execute qmail-queue

The error 'simscan: error forking qmail-queue' will happen for either
of the following reasons:

a) [EAGAIN]
 The system-wide limit on the total number of processes under
 execution would be exceeded, or the system-imposed limit on the
 total number of processes under execution by a single user would
 be exceeded.
b) [ENOMEM]
 There is insufficient swap space for the new process.


On 9/1/2017 11:53 AM, Eric Broch wrote:

Are there any errors in /var/log/messages that correspond?


On 9/1/2017 11:39 AM, Jeff Koch wrote:

Eric / Remo / Rajesh - we reported this same problem a number of
months ago and thought we had fixed it. But alas, it's still with
us. The good news is that we haven't had any customer complaints.
I'll be very interested if it can be resolved.

Jeff Koch


On 9/1/2017 12:01 PM, Eric Broch wrote:

I'm not sure what's going on here. Is this a relatively new
phenomenon?

I wonder if it's a memory, or even a disk speed, issue since it
only happens at peak hours?

I think I'll appy Johannes Weberhofer's patch and put it out there
for you...and cross or fingers.


On 9/1/2017 6:28 AM, Rajesh M wrote:

remo / eric

i have still not being able to resolve the qq soft reject error.

these are my findings

1) the errors i see are "error forking qmail-queue" and "ripmime error" which 
causes the qq soft reject.

2) the max concurrent connections in the logs is around 35.

3)  These errors come up during peak working hours when the server is under a  
load of 4 - 8, and they increase the load even more to over 10-15.

4) i came across this link (not sure if this is related)
https://github.com/qmail/simscan/blob/master/simscan.c

5) i can share with you my live smtp logs with simscan debug.

have extracted some lines below


Error forking qmail-queue

@400059a8fa7b0a2ed1b4 tcpserver: status: 31/200
@400059a8fa7b13162584 simscan: cdb looking up version spam
@400059a8fa7b13166bd4 simscan: runned_scanners is  attach: 1.4.0 clamav: 
0.98.6/m: spam: 3.3.2
@400059a8fa7b13166fbc simscan: found 3.3.2
@400059a8fa7b13168efc simscan:[10757]:CLEAN (5.00/30.00):9.7712s:-Possible 
Spam- RE_ REQUIRE BOOKING // 1X20  //
ICD TKD TO 
BANDARABASS:103.241.181.228:cs@atlasdecargo.com:rathe...@radiant-india.net
@400059a8fa7b1316cd7c simscan: done, execing qmail-queue
@400059a8fa7b1316fc5c simscan: error forking qmail-queue
@400059a8fa7b13199854 simscan: exit error code: 71
@400059a8fa7b131c4004 qmail-smtpd: qq soft reject (mail server temporarily 
rejected message (#4.3.0)): MAILFROM:
RCPTTO:rathe...@y.net


Error in ripmime

@400059a8fa98045a4bc4 simscan: pelookup: domain is aissamaritime.in
@400059a8fa98045a4bc4 simscan: cdb looking up aissamaritime.in
@400059a8fa98045a4fac simscan: pelookup: local part is shailesh_k_bom
@400059a8fa98045a4fac simscan: lpart: local part is **
@400059a8fa98045a5394 simscan: cdb looking upshailesh_k_...@aissamaritime.in
@400059a8fa98045a5394 simscan: ripmime error
@400059a8fa98045a6334 simscan: exit error code: 71
@400059a8fa98045a95fc qmail-smtpd: qq soft reject (mail server temporarily 
rejected message (#4.3.0)):
MAILFROM:  RCPTTO:shailesh_k_...@yy.in


Wierd error logs

@400059a9032f3aa79a24 simscan: clamdscan: --- SCAN SUMMARY 
---
@400059a9032f3aa7b964 simscan: clamdscan: 
/var/qmail/simscan/1504248613.321653.5221: OK
@400059a9032f3aa86d14 simscan: clamdscan:
@400059a9032f3aa870fc simscan: clamdscan: ---simscan: cdb looking up 
version clamav
@400059a9032f3aa8a3c4  SCAN simscan: clamdscan: SIUnMfMeAcRtYe d- 
-fsimscan: runned_scanners is  attach: 1.4.0 clamav: 0.98.6/m:
@400059a9032f3aa8c304 -isimscan: found 0.98.6/m:
@400059a9032f3aa8c6ec -l-e-s-:- -0
@400059a9032f3aa8f1e4 --simscan: normal clamdscan return code: 0
@400059a9032f3aa8f1e4

Re: [qmailtoaster] qq soft reject errors on high load

[Rajesh M]

eric

what is RLIMIT ? is it related to memory ?

rajesh

- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To: qmailtoaster-list@qmailtoaster.com
Sent: Fri, 1 Sep 2017 13:25:49 -0600
Subject:

Here's a section of code in simscan.c that controls the number of processes.

#ifdef HAS_ULIMIT_NPROC
   /* Set ulimits to prevent hangs if it forks too many processes */
   getrlimit(RLIMIT_NPROC, );
   limits.rlim_cur = 1024;
   setrlimit(RLIMIT_NPROC, );
#endif

The RLIMIT is set to 1024.

I think this may be the issue and would explain why there would be
forking issues with qmail-queue, ripmime, and others. This is in place
to prevent hangs...could be a good thing.


On 9/1/2017 1:04 PM, Remo Mattei wrote:
> interesting...
>
> mine was an easy fix by changing the owner and permissions so
> underline has some other triggering to the kernel and how it reads
> those layers.
>
> On 9/1/17 11:18 AM, Eric Broch wrote:
>>
>> 1)  Problem forking
>>
>> vfork() is used to execute qmail-queue
>>
>> The error 'simscan: error forking qmail-queue' will happen for either
>> of the following reasons:
>>
>> a) [EAGAIN]
>> The system-wide limit on the total number of processes under
>> execution would be exceeded, or the system-imposed limit on the
>> total number of processes under execution by a single user would
>> be exceeded.
>> b) [ENOMEM]
>> There is insufficient swap space for the new process.
>>
>>
>> On 9/1/2017 11:53 AM, Eric Broch wrote:
>>>
>>> Are there any errors in /var/log/messages that correspond?
>>>
>>>
>>> On 9/1/2017 11:39 AM, Jeff Koch wrote:

 Eric / Remo / Rajesh - we reported this same problem a number of
 months ago and thought we had fixed it. But alas, it's still with
 us. The good news is that we haven't had any customer complaints.
 I'll be very interested if it can be resolved.

 Jeff Koch


 On 9/1/2017 12:01 PM, Eric Broch wrote:
>
> I'm not sure what's going on here. Is this a relatively new
> phenomenon?
>
> I wonder if it's a memory, or even a disk speed, issue since it
> only happens at peak hours?
>
> I think I'll appy Johannes Weberhofer's patch and put it out there
> for you...and cross or fingers.
>
>
> On 9/1/2017 6:28 AM, Rajesh M wrote:
>> remo / eric
>>
>> i have still not being able to resolve the qq soft reject error.
>>
>> these are my findings
>>
>> 1) the errors i see are "error forking qmail-queue" and "ripmime error" 
>> which causes the qq soft reject.
>>
>> 2) the max concurrent connections in the logs is around 35.
>>
>> 3)  These errors come up during peak working hours when the server is 
>> under a  load of 4 - 8, and they increase the load even more to over 
>> 10-15.
>>
>> 4) i came across this link (not sure if this is related)
>> https://github.com/qmail/simscan/blob/master/simscan.c
>>
>> 5) i can share with you my live smtp logs with simscan debug.
>>
>> have extracted some lines below
>>
>>
>> Error forking qmail-queue
>>
>> @400059a8fa7b0a2ed1b4 tcpserver: status: 31/200
>> @400059a8fa7b13162584 simscan: cdb looking up version spam
>> @400059a8fa7b13166bd4 simscan: runned_scanners is  attach: 1.4.0 
>> clamav: 0.98.6/m: spam: 3.3.2
>> @400059a8fa7b13166fbc simscan: found 3.3.2
>> @400059a8fa7b13168efc simscan:[10757]:CLEAN 
>> (5.00/30.00):9.7712s:-Possible Spam- RE_ REQUIRE BOOKING // 1X20  //
>> ICD TKD TO 
>> BANDARABASS:103.241.181.228:cs@atlasdecargo.com:rathe...@radiant-india.net
>> @400059a8fa7b1316cd7c simscan: done, execing qmail-queue
>> @400059a8fa7b1316fc5c simscan: error forking qmail-queue
>> @400059a8fa7b13199854 simscan: exit error code: 71
>> @400059a8fa7b131c4004 qmail-smtpd: qq soft reject (mail server 
>> temporarily rejected message (#4.3.0)): MAILFROM:
>> RCPTTO:rathe...@y.net
>>
>>
>> Error in ripmime
>>
>> @400059a8fa98045a4bc4 simscan: pelookup: domain is aissamaritime.in
>> @400059a8fa98045a4bc4 simscan: cdb looking up aissamaritime.in
>> @400059a8fa98045a4fac simscan: pelookup: local part is shailesh_k_bom
>> @400059a8fa98045a4fac simscan: lpart: local part is **
>> @400059a8fa98045a5394 simscan: cdb looking 
>> upshailesh_k_...@aissamaritime.in
>> @400059a8fa98045a5394 simscan: ripmime error
>> @400059a8fa98045a6334 simscan: exit error code: 71
>> @400059a8fa98045a95fc qmail-smtpd: qq soft reject (mail server 
>> temporarily rejected message (#4.3.0)):
>> MAILFROM:  RCPTTO:shailesh_k_...@yy.in
>>
>>
>> Wierd error logs
>>
>> @400059a9032f3aa79a24 simscan: clamdscan: --- SCAN SUMMARY 
>> ---
>> @400059a9032f3aa7b964 

Re: [qmailtoaster] qq soft reject errors on high load

[Rajesh M]

jeff

could you please let me know the cpu details of your machine

is it a dell machine or some other with  intel E5 processor ?

rajesh


- Original Message -
From: Jeff Koch [mailto:jeffk...@intersessions.com]
To: qmailtoaster-list@qmailtoaster.com
Sent: Fri, 1 Sep 2017 13:39:37 -0400
Subject:

Eric / Remo / Rajesh - we reported this same problem a number of months
ago and thought we had fixed it. But alas, it's still with us. The good
news is that we haven't had any customer complaints. I'll be very
interested if it can be resolved.

Jeff Koch


On 9/1/2017 12:01 PM, Eric Broch wrote:
>
> I'm not sure what's going on here. Is this a relatively new phenomenon?
>
> I wonder if it's a memory, or even a disk speed, issue since it only
> happens at peak hours?
>
> I think I'll appy Johannes Weberhofer's patch and put it out there for
> you...and cross or fingers.
>
>
> On 9/1/2017 6:28 AM, Rajesh M wrote:
>> remo / eric
>>
>> i have still not being able to resolve the qq soft reject error.
>>
>> these are my findings
>>
>> 1) the errors i see are "error forking qmail-queue" and "ripmime error" 
>> which causes the qq soft reject.
>>
>> 2) the max concurrent connections in the logs is around 35.
>>
>> 3)  These errors come up during peak working hours when the server is under 
>> a  load of 4 - 8, and they increase the load even more to over 10-15.
>>
>> 4) i came across this link (not sure if this is related)
>> https://github.com/qmail/simscan/blob/master/simscan.c
>>
>> 5) i can share with you my live smtp logs with simscan debug.
>>
>> have extracted some lines below
>>
>>
>> Error forking qmail-queue
>>
>> @400059a8fa7b0a2ed1b4 tcpserver: status: 31/200
>> @400059a8fa7b13162584 simscan: cdb looking up version spam
>> @400059a8fa7b13166bd4 simscan: runned_scanners is  attach: 1.4.0 clamav: 
>> 0.98.6/m: spam: 3.3.2
>> @400059a8fa7b13166fbc simscan: found 3.3.2
>> @400059a8fa7b13168efc simscan:[10757]:CLEAN 
>> (5.00/30.00):9.7712s:-Possible Spam- RE_ REQUIRE BOOKING // 1X20  //
>> ICD TKD TO 
>> BANDARABASS:103.241.181.228:cs@atlasdecargo.com:rathe...@radiant-india.net
>> @400059a8fa7b1316cd7c simscan: done, execing qmail-queue
>> @400059a8fa7b1316fc5c simscan: error forking qmail-queue
>> @400059a8fa7b13199854 simscan: exit error code: 71
>> @400059a8fa7b131c4004 qmail-smtpd: qq soft reject (mail server 
>> temporarily rejected message (#4.3.0)): MAILFROM:
>> RCPTTO:rathe...@y.net
>>
>>
>> Error in ripmime
>>
>> @400059a8fa98045a4bc4 simscan: pelookup: domain is aissamaritime.in
>> @400059a8fa98045a4bc4 simscan: cdb looking up aissamaritime.in
>> @400059a8fa98045a4fac simscan: pelookup: local part is shailesh_k_bom
>> @400059a8fa98045a4fac simscan: lpart: local part is **
>> @400059a8fa98045a5394 simscan: cdb looking 
>> upshailesh_k_...@aissamaritime.in
>> @400059a8fa98045a5394 simscan: ripmime error
>> @400059a8fa98045a6334 simscan: exit error code: 71
>> @400059a8fa98045a95fc qmail-smtpd: qq soft reject (mail server 
>> temporarily rejected message (#4.3.0)):
>> MAILFROM:  RCPTTO:shailesh_k_...@yy.in
>>
>>
>> Wierd error logs
>>
>> @400059a9032f3aa79a24 simscan: clamdscan: --- SCAN SUMMARY 
>> ---
>> @400059a9032f3aa7b964 simscan: clamdscan: 
>> /var/qmail/simscan/1504248613.321653.5221: OK
>> @400059a9032f3aa86d14 simscan: clamdscan:
>> @400059a9032f3aa870fc simscan: clamdscan: ---simscan: cdb looking up 
>> version clamav
>> @400059a9032f3aa8a3c4  SCAN simscan: clamdscan: SIUnMfMeAcRtYe 
>> d- -fsimscan: runned_scanners is  attach: 1.4.0 clamav: 0.98.6/m:
>> @400059a9032f3aa8c304 -isimscan: found 0.98.6/m:
>> @400059a9032f3aa8c6ec -l-e-s-:- -0
>> @400059a9032f3aa8f1e4 --simscan: normal clamdscan return code: 0
>> @400059a9032f3aa8f1e4
>> @400059a9032f3aa8f5cc simscan: clamdscan: Infected fsimscan: clamdscan: 
>> iTliemes::  00
>> @400059a9032f3aa93834 .simscan: clamdscan: 1T1i8m es:e c0 .1(002  ms 
>> e0simscan: calling spamc
>> @400059a9032f3aa96ee4 c  s(0)simscan: calling /usr/bin/spamc simscan: 
>> clamdscan:
>> @400059a9032f3aa999dc  spamcm
>> @400059a9032f3aa999dc 0 s)
>> @400059a9032f3aa9ad64 /var/qmail/simscan/1simscan: check_spam had an 
>> error ret: -1
>> @400059a9032f3aa9e02c 504248613.307311.5215: OK
>> @400059a9032f337c simscan: clamdscan:
>> @400059a9032f3764 simscan: clamdscan: --- SCAN SUMMARY 
>> ---
>> @400059a9032f3aab3fbc simscan: clamdscan: Infected files: 0
>> @400059a9032f3aaba164 simscan: clamdscan: Time: 0.135 sec (0 m 0 s)
>> @400059a9032f3aac39bc simscan: clamdscan: /var/qmail/simscan/15simscan: 
>> exit error code: 71
>> @400059a9032f3aaca334 04248613.308469.5216: OK
>> @400059a9032f3aacddcc simscan: clamdscan:
>> @400059a9032f3aace984 simscan: clamdscan: --- SCAN SUMMARY 
>> ---
>> 

Re: [qmailtoaster] qq soft reject errors on high load

[Eric Broch]

Here's a section of code in simscan.c that controls the number of processes.

#ifdef HAS_ULIMIT_NPROC
  /* Set ulimits to prevent hangs if it forks too many processes */
  getrlimit(RLIMIT_NPROC, );
  limits.rlim_cur = 1024;
  setrlimit(RLIMIT_NPROC, );
#endif

The RLIMIT is set to 1024.

I think this may be the issue and would explain why there would be 
forking issues with qmail-queue, ripmime, and others. This is in place 
to prevent hangs...could be a good thing.



On 9/1/2017 1:04 PM, Remo Mattei wrote:

interesting...

mine was an easy fix by changing the owner and permissions so 
underline has some other triggering to the kernel and how it reads 
those layers.


On 9/1/17 11:18 AM, Eric Broch wrote:


1)  Problem forking

vfork() is used to execute qmail-queue

The error 'simscan: error forking qmail-queue' will happen for either 
of the following reasons:


a) [EAGAIN]
The system-wide limit on the total number of processes under
execution would be exceeded, or the system-imposed limit on the
total number of processes under execution by a single user would
be exceeded. 
b) [ENOMEM]
There is insufficient swap space for the new process. 



On 9/1/2017 11:53 AM, Eric Broch wrote:


Are there any errors in /var/log/messages that correspond?


On 9/1/2017 11:39 AM, Jeff Koch wrote:


Eric / Remo / Rajesh - we reported this same problem a number of 
months ago and thought we had fixed it. But alas, it's still with 
us. The good news is that we haven't had any customer complaints. 
I'll be very interested if it can be resolved.


Jeff Koch


On 9/1/2017 12:01 PM, Eric Broch wrote:


I'm not sure what's going on here. Is this a relatively new 
phenomenon?


I wonder if it's a memory, or even a disk speed, issue since it 
only happens at peak hours?


I think I'll appy Johannes Weberhofer's patch and put it out there 
for you...and cross or fingers.



On 9/1/2017 6:28 AM, Rajesh M wrote:

remo / eric

i have still not being able to resolve the qq soft reject error.

these are my findings

1) the errors i see are "error forking qmail-queue" and "ripmime error" which 
causes the qq soft reject.

2) the max concurrent connections in the logs is around 35.

3)  These errors come up during peak working hours when the server is under a  
load of 4 - 8, and they increase the load even more to over 10-15.

4) i came across this link (not sure if this is related)
https://github.com/qmail/simscan/blob/master/simscan.c

5) i can share with you my live smtp logs with simscan debug.

have extracted some lines below


Error forking qmail-queue

@400059a8fa7b0a2ed1b4 tcpserver: status: 31/200
@400059a8fa7b13162584 simscan: cdb looking up version spam
@400059a8fa7b13166bd4 simscan: runned_scanners is  attach: 1.4.0 clamav: 
0.98.6/m: spam: 3.3.2
@400059a8fa7b13166fbc simscan: found 3.3.2
@400059a8fa7b13168efc simscan:[10757]:CLEAN (5.00/30.00):9.7712s:-Possible 
Spam- RE_ REQUIRE BOOKING // 1X20  //
ICD TKD TO 
BANDARABASS:103.241.181.228:cs@atlasdecargo.com:rathe...@radiant-india.net
@400059a8fa7b1316cd7c simscan: done, execing qmail-queue
@400059a8fa7b1316fc5c simscan: error forking qmail-queue
@400059a8fa7b13199854 simscan: exit error code: 71
@400059a8fa7b131c4004 qmail-smtpd: qq soft reject (mail server temporarily 
rejected message (#4.3.0)): MAILFROM:
RCPTTO:rathe...@y.net


Error in ripmime

@400059a8fa98045a4bc4 simscan: pelookup: domain is aissamaritime.in
@400059a8fa98045a4bc4 simscan: cdb looking up aissamaritime.in
@400059a8fa98045a4fac simscan: pelookup: local part is shailesh_k_bom
@400059a8fa98045a4fac simscan: lpart: local part is **
@400059a8fa98045a5394 simscan: cdb looking upshailesh_k_...@aissamaritime.in
@400059a8fa98045a5394 simscan: ripmime error
@400059a8fa98045a6334 simscan: exit error code: 71
@400059a8fa98045a95fc qmail-smtpd: qq soft reject (mail server temporarily 
rejected message (#4.3.0)):
MAILFROM:  RCPTTO:shailesh_k_...@yy.in


Wierd error logs

@400059a9032f3aa79a24 simscan: clamdscan: --- SCAN SUMMARY 
---
@400059a9032f3aa7b964 simscan: clamdscan: 
/var/qmail/simscan/1504248613.321653.5221: OK
@400059a9032f3aa86d14 simscan: clamdscan:
@400059a9032f3aa870fc simscan: clamdscan: ---simscan: cdb looking up 
version clamav
@400059a9032f3aa8a3c4  SCAN simscan: clamdscan: SIUnMfMeAcRtYe d- 
-fsimscan: runned_scanners is  attach: 1.4.0 clamav: 0.98.6/m:
@400059a9032f3aa8c304 -isimscan: found 0.98.6/m:
@400059a9032f3aa8c6ec -l-e-s-:- -0
@400059a9032f3aa8f1e4 --simscan: normal clamdscan return code: 0
@400059a9032f3aa8f1e4
@400059a9032f3aa8f5cc simscan: clamdscan: Infected fsimscan: clamdscan: 
iTliemes::  00
@400059a9032f3aa93834 .simscan: clamdscan: 1T1i8m es:e c0 .1(002  ms 
e0simscan: calling spamc
@400059a9032f3aa96ee4 c  s(0)simscan: calling /usr/bin/spamc simscan: 
clamdscan:

Re: [qmailtoaster] qq soft reject errors on high load

[Eric Broch]

2) Problem forking (ripmime)

vfork is also used to execute ripmime

The error 'simscan: ripmime error' will happen for the same reasons as 
mentioned in the previous email (a and b below):


a) [EAGAIN]
   The system-wide limit on the total number of processes under
   execution would be exceeded, or the system-imposed limit on the
   total number of processes under execution by a single user would be
   exceeded. 
b) [ENOMEM]
   There is insufficient swap space for the new process. 


Eric



On 9/1/2017 12:18 PM, Eric Broch wrote:


1)  Problem forking

vfork() is used to execute qmail-queue

The error 'simscan: error forking qmail-queue' will happen for either 
of the following reasons:


a) [EAGAIN]
The system-wide limit on the total number of processes under
execution would be exceeded, or the system-imposed limit on the
total number of processes under execution by a single user would
be exceeded. 
b) [ENOMEM]
There is insufficient swap space for the new process. 



On 9/1/2017 11:53 AM, Eric Broch wrote:


Are there any errors in /var/log/messages that correspond?


On 9/1/2017 11:39 AM, Jeff Koch wrote:


Eric / Remo / Rajesh - we reported this same problem a number of 
months ago and thought we had fixed it. But alas, it's still with 
us. The good news is that we haven't had any customer complaints. 
I'll be very interested if it can be resolved.


Jeff Koch


On 9/1/2017 12:01 PM, Eric Broch wrote:


I'm not sure what's going on here. Is this a relatively new 
phenomenon?


I wonder if it's a memory, or even a disk speed, issue since it 
only happens at peak hours?


I think I'll appy Johannes Weberhofer's patch and put it out there 
for you...and cross or fingers.



On 9/1/2017 6:28 AM, Rajesh M wrote:

remo / eric

i have still not being able to resolve the qq soft reject error.

these are my findings

1) the errors i see are "error forking qmail-queue" and "ripmime error" which 
causes the qq soft reject.

2) the max concurrent connections in the logs is around 35.

3)  These errors come up during peak working hours when the server is under a  
load of 4 - 8, and they increase the load even more to over 10-15.

4) i came across this link (not sure if this is related)
https://github.com/qmail/simscan/blob/master/simscan.c

5) i can share with you my live smtp logs with simscan debug.

have extracted some lines below


Error forking qmail-queue

@400059a8fa7b0a2ed1b4 tcpserver: status: 31/200
@400059a8fa7b13162584 simscan: cdb looking up version spam
@400059a8fa7b13166bd4 simscan: runned_scanners is  attach: 1.4.0 clamav: 
0.98.6/m: spam: 3.3.2
@400059a8fa7b13166fbc simscan: found 3.3.2
@400059a8fa7b13168efc simscan:[10757]:CLEAN (5.00/30.00):9.7712s:-Possible 
Spam- RE_ REQUIRE BOOKING // 1X20  //
ICD TKD TO 
BANDARABASS:103.241.181.228:cs@atlasdecargo.com:rathe...@radiant-india.net
@400059a8fa7b1316cd7c simscan: done, execing qmail-queue
@400059a8fa7b1316fc5c simscan: error forking qmail-queue
@400059a8fa7b13199854 simscan: exit error code: 71
@400059a8fa7b131c4004 qmail-smtpd: qq soft reject (mail server temporarily 
rejected message (#4.3.0)): MAILFROM:
RCPTTO:rathe...@y.net


Error in ripmime

@400059a8fa98045a4bc4 simscan: pelookup: domain is aissamaritime.in
@400059a8fa98045a4bc4 simscan: cdb looking up aissamaritime.in
@400059a8fa98045a4fac simscan: pelookup: local part is shailesh_k_bom
@400059a8fa98045a4fac simscan: lpart: local part is **
@400059a8fa98045a5394 simscan: cdb looking upshailesh_k_...@aissamaritime.in
@400059a8fa98045a5394 simscan: ripmime error
@400059a8fa98045a6334 simscan: exit error code: 71
@400059a8fa98045a95fc qmail-smtpd: qq soft reject (mail server temporarily 
rejected message (#4.3.0)):
MAILFROM:  RCPTTO:shailesh_k_...@yy.in


Wierd error logs

@400059a9032f3aa79a24 simscan: clamdscan: --- SCAN SUMMARY 
---
@400059a9032f3aa7b964 simscan: clamdscan: 
/var/qmail/simscan/1504248613.321653.5221: OK
@400059a9032f3aa86d14 simscan: clamdscan:
@400059a9032f3aa870fc simscan: clamdscan: ---simscan: cdb looking up 
version clamav
@400059a9032f3aa8a3c4  SCAN simscan: clamdscan: SIUnMfMeAcRtYe d- 
-fsimscan: runned_scanners is  attach: 1.4.0 clamav: 0.98.6/m:
@400059a9032f3aa8c304 -isimscan: found 0.98.6/m:
@400059a9032f3aa8c6ec -l-e-s-:- -0
@400059a9032f3aa8f1e4 --simscan: normal clamdscan return code: 0
@400059a9032f3aa8f1e4
@400059a9032f3aa8f5cc simscan: clamdscan: Infected fsimscan: clamdscan: 
iTliemes::  00
@400059a9032f3aa93834 .simscan: clamdscan: 1T1i8m es:e c0 .1(002  ms 
e0simscan: calling spamc
@400059a9032f3aa96ee4 c  s(0)simscan: calling /usr/bin/spamc simscan: 
clamdscan:
@400059a9032f3aa999dc  spamcm
@400059a9032f3aa999dc 0 s)
@400059a9032f3aa9ad64 /var/qmail/simscan/1simscan: check_spam had an error 
ret: -1
@400059a9032f3aa9e02c 

Re: [qmailtoaster] qq soft reject errors on high load

[Eric Broch]

1)  Problem forking

vfork() is used to execute qmail-queue

The error 'simscan: error forking qmail-queue' will happen for either of 
the following reasons:


a) [EAGAIN]
   The system-wide limit on the total number of processes under
   execution would be exceeded, or the system-imposed limit on the
   total number of processes under execution by a single user would be
   exceeded. 
b) [ENOMEM]
   There is insufficient swap space for the new process. 



On 9/1/2017 11:53 AM, Eric Broch wrote:


Are there any errors in /var/log/messages that correspond?


On 9/1/2017 11:39 AM, Jeff Koch wrote:


Eric / Remo / Rajesh - we reported this same problem a number of 
months ago and thought we had fixed it. But alas, it's still with us. 
The good news is that we haven't had any customer complaints. I'll be 
very interested if it can be resolved.


Jeff Koch


On 9/1/2017 12:01 PM, Eric Broch wrote:


I'm not sure what's going on here. Is this a relatively new phenomenon?

I wonder if it's a memory, or even a disk speed, issue since it only 
happens at peak hours?


I think I'll appy Johannes Weberhofer's patch and put it out there 
for you...and cross or fingers.



On 9/1/2017 6:28 AM, Rajesh M wrote:

remo / eric

i have still not being able to resolve the qq soft reject error.

these are my findings

1) the errors i see are "error forking qmail-queue" and "ripmime error" which 
causes the qq soft reject.

2) the max concurrent connections in the logs is around 35.

3)  These errors come up during peak working hours when the server is under a  
load of 4 - 8, and they increase the load even more to over 10-15.

4) i came across this link (not sure if this is related)
https://github.com/qmail/simscan/blob/master/simscan.c

5) i can share with you my live smtp logs with simscan debug.

have extracted some lines below


Error forking qmail-queue

@400059a8fa7b0a2ed1b4 tcpserver: status: 31/200
@400059a8fa7b13162584 simscan: cdb looking up version spam
@400059a8fa7b13166bd4 simscan: runned_scanners is  attach: 1.4.0 clamav: 
0.98.6/m: spam: 3.3.2
@400059a8fa7b13166fbc simscan: found 3.3.2
@400059a8fa7b13168efc simscan:[10757]:CLEAN (5.00/30.00):9.7712s:-Possible 
Spam- RE_ REQUIRE BOOKING // 1X20  //
ICD TKD TO 
BANDARABASS:103.241.181.228:cs@atlasdecargo.com:rathe...@radiant-india.net
@400059a8fa7b1316cd7c simscan: done, execing qmail-queue
@400059a8fa7b1316fc5c simscan: error forking qmail-queue
@400059a8fa7b13199854 simscan: exit error code: 71
@400059a8fa7b131c4004 qmail-smtpd: qq soft reject (mail server temporarily 
rejected message (#4.3.0)): MAILFROM:
RCPTTO:rathe...@y.net


Error in ripmime

@400059a8fa98045a4bc4 simscan: pelookup: domain is aissamaritime.in
@400059a8fa98045a4bc4 simscan: cdb looking up aissamaritime.in
@400059a8fa98045a4fac simscan: pelookup: local part is shailesh_k_bom
@400059a8fa98045a4fac simscan: lpart: local part is **
@400059a8fa98045a5394 simscan: cdb looking upshailesh_k_...@aissamaritime.in
@400059a8fa98045a5394 simscan: ripmime error
@400059a8fa98045a6334 simscan: exit error code: 71
@400059a8fa98045a95fc qmail-smtpd: qq soft reject (mail server temporarily 
rejected message (#4.3.0)):
MAILFROM:  RCPTTO:shailesh_k_...@yy.in


Wierd error logs

@400059a9032f3aa79a24 simscan: clamdscan: --- SCAN SUMMARY 
---
@400059a9032f3aa7b964 simscan: clamdscan: 
/var/qmail/simscan/1504248613.321653.5221: OK
@400059a9032f3aa86d14 simscan: clamdscan:
@400059a9032f3aa870fc simscan: clamdscan: ---simscan: cdb looking up 
version clamav
@400059a9032f3aa8a3c4  SCAN simscan: clamdscan: SIUnMfMeAcRtYe d- 
-fsimscan: runned_scanners is  attach: 1.4.0 clamav: 0.98.6/m:
@400059a9032f3aa8c304 -isimscan: found 0.98.6/m:
@400059a9032f3aa8c6ec -l-e-s-:- -0
@400059a9032f3aa8f1e4 --simscan: normal clamdscan return code: 0
@400059a9032f3aa8f1e4
@400059a9032f3aa8f5cc simscan: clamdscan: Infected fsimscan: clamdscan: 
iTliemes::  00
@400059a9032f3aa93834 .simscan: clamdscan: 1T1i8m es:e c0 .1(002  ms 
e0simscan: calling spamc
@400059a9032f3aa96ee4 c  s(0)simscan: calling /usr/bin/spamc simscan: 
clamdscan:
@400059a9032f3aa999dc  spamcm
@400059a9032f3aa999dc 0 s)
@400059a9032f3aa9ad64 /var/qmail/simscan/1simscan: check_spam had an error 
ret: -1
@400059a9032f3aa9e02c 504248613.307311.5215: OK
@400059a9032f337c simscan: clamdscan:
@400059a9032f3764 simscan: clamdscan: --- SCAN SUMMARY 
---
@400059a9032f3aab3fbc simscan: clamdscan: Infected files: 0
@400059a9032f3aaba164 simscan: clamdscan: Time: 0.135 sec (0 m 0 s)
@400059a9032f3aac39bc simscan: clamdscan: /var/qmail/simscan/15simscan: 
exit error code: 71
@400059a9032f3aaca334 04248613.308469.5216: OK
@400059a9032f3aacddcc simscan: clamdscan:
@400059a9032f3aace984 simscan: clamdscan: --- SCAN 

Re: [qmailtoaster] qq soft reject errors on high load

[Jeff Koch]

Eric / Remo / Rajesh - we reported this same problem a number of months 
ago and thought we had fixed it. But alas, it's still with us. The good 
news is that we haven't had any customer complaints. I'll be very 
interested if it can be resolved.


Jeff Koch


On 9/1/2017 12:01 PM, Eric Broch wrote:


I'm not sure what's going on here. Is this a relatively new phenomenon?

I wonder if it's a memory, or even a disk speed, issue since it only 
happens at peak hours?


I think I'll appy Johannes Weberhofer's patch and put it out there for 
you...and cross or fingers.



On 9/1/2017 6:28 AM, Rajesh M wrote:

remo / eric

i have still not being able to resolve the qq soft reject error.

these are my findings

1) the errors i see are "error forking qmail-queue" and "ripmime error" which 
causes the qq soft reject.

2) the max concurrent connections in the logs is around 35.

3)  These errors come up during peak working hours when the server is under a  
load of 4 - 8, and they increase the load even more to over 10-15.

4) i came across this link (not sure if this is related)
https://github.com/qmail/simscan/blob/master/simscan.c

5) i can share with you my live smtp logs with simscan debug.

have extracted some lines below


Error forking qmail-queue

@400059a8fa7b0a2ed1b4 tcpserver: status: 31/200
@400059a8fa7b13162584 simscan: cdb looking up version spam
@400059a8fa7b13166bd4 simscan: runned_scanners is  attach: 1.4.0 clamav: 
0.98.6/m: spam: 3.3.2
@400059a8fa7b13166fbc simscan: found 3.3.2
@400059a8fa7b13168efc simscan:[10757]:CLEAN (5.00/30.00):9.7712s:-Possible 
Spam- RE_ REQUIRE BOOKING // 1X20  //
ICD TKD TO 
BANDARABASS:103.241.181.228:cs@atlasdecargo.com:rathe...@radiant-india.net
@400059a8fa7b1316cd7c simscan: done, execing qmail-queue
@400059a8fa7b1316fc5c simscan: error forking qmail-queue
@400059a8fa7b13199854 simscan: exit error code: 71
@400059a8fa7b131c4004 qmail-smtpd: qq soft reject (mail server temporarily 
rejected message (#4.3.0)): MAILFROM:
RCPTTO:rathe...@y.net


Error in ripmime

@400059a8fa98045a4bc4 simscan: pelookup: domain is aissamaritime.in
@400059a8fa98045a4bc4 simscan: cdb looking up aissamaritime.in
@400059a8fa98045a4fac simscan: pelookup: local part is shailesh_k_bom
@400059a8fa98045a4fac simscan: lpart: local part is **
@400059a8fa98045a5394 simscan: cdb looking upshailesh_k_...@aissamaritime.in
@400059a8fa98045a5394 simscan: ripmime error
@400059a8fa98045a6334 simscan: exit error code: 71
@400059a8fa98045a95fc qmail-smtpd: qq soft reject (mail server temporarily 
rejected message (#4.3.0)):
MAILFROM:  RCPTTO:shailesh_k_...@yy.in


Wierd error logs

@400059a9032f3aa79a24 simscan: clamdscan: --- SCAN SUMMARY 
---
@400059a9032f3aa7b964 simscan: clamdscan: 
/var/qmail/simscan/1504248613.321653.5221: OK
@400059a9032f3aa86d14 simscan: clamdscan:
@400059a9032f3aa870fc simscan: clamdscan: ---simscan: cdb looking up 
version clamav
@400059a9032f3aa8a3c4  SCAN simscan: clamdscan: SIUnMfMeAcRtYe d- 
-fsimscan: runned_scanners is  attach: 1.4.0 clamav: 0.98.6/m:
@400059a9032f3aa8c304 -isimscan: found 0.98.6/m:
@400059a9032f3aa8c6ec -l-e-s-:- -0
@400059a9032f3aa8f1e4 --simscan: normal clamdscan return code: 0
@400059a9032f3aa8f1e4
@400059a9032f3aa8f5cc simscan: clamdscan: Infected fsimscan: clamdscan: 
iTliemes::  00
@400059a9032f3aa93834 .simscan: clamdscan: 1T1i8m es:e c0 .1(002  ms 
e0simscan: calling spamc
@400059a9032f3aa96ee4 c  s(0)simscan: calling /usr/bin/spamc simscan: 
clamdscan:
@400059a9032f3aa999dc  spamcm
@400059a9032f3aa999dc 0 s)
@400059a9032f3aa9ad64 /var/qmail/simscan/1simscan: check_spam had an error 
ret: -1
@400059a9032f3aa9e02c 504248613.307311.5215: OK
@400059a9032f337c simscan: clamdscan:
@400059a9032f3764 simscan: clamdscan: --- SCAN SUMMARY 
---
@400059a9032f3aab3fbc simscan: clamdscan: Infected files: 0
@400059a9032f3aaba164 simscan: clamdscan: Time: 0.135 sec (0 m 0 s)
@400059a9032f3aac39bc simscan: clamdscan: /var/qmail/simscan/15simscan: 
exit error code: 71
@400059a9032f3aaca334 04248613.308469.5216: OK
@400059a9032f3aacddcc simscan: clamdscan:
@400059a9032f3aace984 simscan: clamdscan: --- SCAN SUMMARY 
---
@400059a9




@400059a9032f3aa86d14 simscan: clamdscan:
@400059a9032f3aa870fc simscan: clamdscan: ---simscan: cdb looking up 
version clamav
@400059a9032f3aa8a3c4  SCAN simscan: clamdscan: SIUnMfMeAcRtYe d- 
-fsimscan: runned_scanners is  attach: 1.4.0 clamav: 0.98.6/m:
@400059a9032f3aa8c304 -isimscan: found 0.98.6/m:
@400059a9032f3aa8c6ec -l-e-s-:- -0
@400059a9032f3aa8f1e4 --simscan: normal clamdscan return code: 0
@400059a9032f3aa8f1e4
@400059a9032f3aa8f5cc simscan: clamdscan: Infected fsimscan: clamdscan: 
iTliemes::  

Re: [qmailtoaster] qq soft reject errors on high load

[Rajesh M]

eric

the machine i have is a dell,  hex core with hyperthreading, 16 gb ram -- 
hardly 4 gb being used, 600 gb 15krpm drive, 2 drives of 2000 gb each 
separately handling data.

rajesh


- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To: qmailtoaster-list@qmailtoaster.com
Sent: Fri, 1 Sep 2017 10:01:01 -0600
Subject:

I'm not sure what's going on here. Is this a relatively new phenomenon?

I wonder if it's a memory, or even a disk speed, issue since it only
happens at peak hours?

I think I'll appy Johannes Weberhofer's patch and put it out there for
you...and cross or fingers.


On 9/1/2017 6:28 AM, Rajesh M wrote:
> remo / eric
>
> i have still not being able to resolve the qq soft reject error.
>
> these are my findings
>
> 1) the errors i see are "error forking qmail-queue" and "ripmime error" which 
> causes the qq soft reject.
>
> 2) the max concurrent connections in the logs is around 35.
>
> 3)  These errors come up during peak working hours when the server is under a 
>  load of 4 - 8, and they increase the load even more to over 10-15.
>
> 4) i came across this link (not sure if this is related)
> https://github.com/qmail/simscan/blob/master/simscan.c
>
> 5) i can share with you my live smtp logs with simscan debug.
>
> have extracted some lines below
>
>
> Error forking qmail-queue
>
> @400059a8fa7b0a2ed1b4 tcpserver: status: 31/200
> @400059a8fa7b13162584 simscan: cdb looking up version spam
> @400059a8fa7b13166bd4 simscan: runned_scanners is  attach: 1.4.0 clamav: 
> 0.98.6/m: spam: 3.3.2
> @400059a8fa7b13166fbc simscan: found 3.3.2
> @400059a8fa7b13168efc simscan:[10757]:CLEAN 
> (5.00/30.00):9.7712s:-Possible Spam- RE_ REQUIRE BOOKING // 1X20  //
> ICD TKD TO 
> BANDARABASS:103.241.181.228:cs@atlasdecargo.com:rathe...@radiant-india.net
> @400059a8fa7b1316cd7c simscan: done, execing qmail-queue
> @400059a8fa7b1316fc5c simscan: error forking qmail-queue
> @400059a8fa7b13199854 simscan: exit error code: 71
> @400059a8fa7b131c4004 qmail-smtpd: qq soft reject (mail server 
> temporarily rejected message (#4.3.0)): MAILFROM:
> RCPTTO:rathe...@y.net
>
>
> Error in ripmime
>
> @400059a8fa98045a4bc4 simscan: pelookup: domain is aissamaritime.in
> @400059a8fa98045a4bc4 simscan: cdb looking up aissamaritime.in
> @400059a8fa98045a4fac simscan: pelookup: local part is shailesh_k_bom
> @400059a8fa98045a4fac simscan: lpart: local part is **
> @400059a8fa98045a5394 simscan: cdb looking 
> upshailesh_k_...@aissamaritime.in
> @400059a8fa98045a5394 simscan: ripmime error
> @400059a8fa98045a6334 simscan: exit error code: 71
> @400059a8fa98045a95fc qmail-smtpd: qq soft reject (mail server 
> temporarily rejected message (#4.3.0)):
> MAILFROM:  RCPTTO:shailesh_k_...@yy.in
>
>
> Wierd error logs
>
> @400059a9032f3aa79a24 simscan: clamdscan: --- SCAN SUMMARY 
> ---
> @400059a9032f3aa7b964 simscan: clamdscan: 
> /var/qmail/simscan/1504248613.321653.5221: OK
> @400059a9032f3aa86d14 simscan: clamdscan:
> @400059a9032f3aa870fc simscan: clamdscan: ---simscan: cdb looking up 
> version clamav
> @400059a9032f3aa8a3c4  SCAN simscan: clamdscan: SIUnMfMeAcRtYe d- 
> -fsimscan: runned_scanners is  attach: 1.4.0 clamav: 0.98.6/m:
> @400059a9032f3aa8c304 -isimscan: found 0.98.6/m:
> @400059a9032f3aa8c6ec -l-e-s-:- -0
> @400059a9032f3aa8f1e4 --simscan: normal clamdscan return code: 0
> @400059a9032f3aa8f1e4
> @400059a9032f3aa8f5cc simscan: clamdscan: Infected fsimscan: clamdscan: 
> iTliemes::  00
> @400059a9032f3aa93834 .simscan: clamdscan: 1T1i8m es:e c0 .1(002  ms 
> e0simscan: calling spamc
> @400059a9032f3aa96ee4 c  s(0)simscan: calling /usr/bin/spamc simscan: 
> clamdscan:
> @400059a9032f3aa999dc  spamcm
> @400059a9032f3aa999dc 0 s)
> @400059a9032f3aa9ad64 /var/qmail/simscan/1simscan: check_spam had an 
> error ret: -1
> @400059a9032f3aa9e02c 504248613.307311.5215: OK
> @400059a9032f337c simscan: clamdscan:
> @400059a9032f3764 simscan: clamdscan: --- SCAN SUMMARY 
> ---
> @400059a9032f3aab3fbc simscan: clamdscan: Infected files: 0
> @400059a9032f3aaba164 simscan: clamdscan: Time: 0.135 sec (0 m 0 s)
> @400059a9032f3aac39bc simscan: clamdscan: /var/qmail/simscan/15simscan: 
> exit error code: 71
> @400059a9032f3aaca334 04248613.308469.5216: OK
> @400059a9032f3aacddcc simscan: clamdscan:
> @400059a9032f3aace984 simscan: clamdscan: --- SCAN SUMMARY 
> ---
> @400059a9
>
>
> 
>
> @400059a9032f3aa86d14 simscan: clamdscan:
> @400059a9032f3aa870fc simscan: clamdscan: ---simscan: cdb looking up 
> version clamav
> @400059a9032f3aa8a3c4  SCAN simscan: clamdscan: SIUnMfMeAcRtYe d- 
> -fsimscan: runned_scanners is  attach: 1.4.0 clamav: 0.98.6/m:
> @400059a9032f3aa8c304 -isimscan: found 

Re: [qmailtoaster] qq soft reject errors on high load

[Eric Broch]

I'm not sure what's going on here. Is this a relatively new phenomenon?

I wonder if it's a memory, or even a disk speed, issue since it only 
happens at peak hours?


I think I'll appy Johannes Weberhofer's patch and put it out there for 
you...and cross or fingers.



On 9/1/2017 6:28 AM, Rajesh M wrote:

remo / eric

i have still not being able to resolve the qq soft reject error.

these are my findings

1) the errors i see are "error forking qmail-queue" and "ripmime error" which 
causes the qq soft reject.

2) the max concurrent connections in the logs is around 35.

3)  These errors come up during peak working hours when the server is under a  
load of 4 - 8, and they increase the load even more to over 10-15.

4) i came across this link (not sure if this is related)
https://github.com/qmail/simscan/blob/master/simscan.c

5) i can share with you my live smtp logs with simscan debug.

have extracted some lines below


Error forking qmail-queue

@400059a8fa7b0a2ed1b4 tcpserver: status: 31/200
@400059a8fa7b13162584 simscan: cdb looking up version spam
@400059a8fa7b13166bd4 simscan: runned_scanners is  attach: 1.4.0 clamav: 
0.98.6/m: spam: 3.3.2
@400059a8fa7b13166fbc simscan: found 3.3.2
@400059a8fa7b13168efc simscan:[10757]:CLEAN (5.00/30.00):9.7712s:-Possible 
Spam- RE_ REQUIRE BOOKING // 1X20  //
ICD TKD TO 
BANDARABASS:103.241.181.228:cs@atlasdecargo.com:rathe...@radiant-india.net
@400059a8fa7b1316cd7c simscan: done, execing qmail-queue
@400059a8fa7b1316fc5c simscan: error forking qmail-queue
@400059a8fa7b13199854 simscan: exit error code: 71
@400059a8fa7b131c4004 qmail-smtpd: qq soft reject (mail server temporarily 
rejected message (#4.3.0)): MAILFROM:
RCPTTO:rathe...@y.net


Error in ripmime

@400059a8fa98045a4bc4 simscan: pelookup: domain is aissamaritime.in
@400059a8fa98045a4bc4 simscan: cdb looking up aissamaritime.in
@400059a8fa98045a4fac simscan: pelookup: local part is shailesh_k_bom
@400059a8fa98045a4fac simscan: lpart: local part is **
@400059a8fa98045a5394 simscan: cdb looking upshailesh_k_...@aissamaritime.in
@400059a8fa98045a5394 simscan: ripmime error
@400059a8fa98045a6334 simscan: exit error code: 71
@400059a8fa98045a95fc qmail-smtpd: qq soft reject (mail server temporarily 
rejected message (#4.3.0)):
MAILFROM:  RCPTTO:shailesh_k_...@yy.in


Wierd error logs

@400059a9032f3aa79a24 simscan: clamdscan: --- SCAN SUMMARY 
---
@400059a9032f3aa7b964 simscan: clamdscan: 
/var/qmail/simscan/1504248613.321653.5221: OK
@400059a9032f3aa86d14 simscan: clamdscan:
@400059a9032f3aa870fc simscan: clamdscan: ---simscan: cdb looking up 
version clamav
@400059a9032f3aa8a3c4  SCAN simscan: clamdscan: SIUnMfMeAcRtYe d- 
-fsimscan: runned_scanners is  attach: 1.4.0 clamav: 0.98.6/m:
@400059a9032f3aa8c304 -isimscan: found 0.98.6/m:
@400059a9032f3aa8c6ec -l-e-s-:- -0
@400059a9032f3aa8f1e4 --simscan: normal clamdscan return code: 0
@400059a9032f3aa8f1e4
@400059a9032f3aa8f5cc simscan: clamdscan: Infected fsimscan: clamdscan: 
iTliemes::  00
@400059a9032f3aa93834 .simscan: clamdscan: 1T1i8m es:e c0 .1(002  ms 
e0simscan: calling spamc
@400059a9032f3aa96ee4 c  s(0)simscan: calling /usr/bin/spamc simscan: 
clamdscan:
@400059a9032f3aa999dc  spamcm
@400059a9032f3aa999dc 0 s)
@400059a9032f3aa9ad64 /var/qmail/simscan/1simscan: check_spam had an error 
ret: -1
@400059a9032f3aa9e02c 504248613.307311.5215: OK
@400059a9032f337c simscan: clamdscan:
@400059a9032f3764 simscan: clamdscan: --- SCAN SUMMARY 
---
@400059a9032f3aab3fbc simscan: clamdscan: Infected files: 0
@400059a9032f3aaba164 simscan: clamdscan: Time: 0.135 sec (0 m 0 s)
@400059a9032f3aac39bc simscan: clamdscan: /var/qmail/simscan/15simscan: 
exit error code: 71
@400059a9032f3aaca334 04248613.308469.5216: OK
@400059a9032f3aacddcc simscan: clamdscan:
@400059a9032f3aace984 simscan: clamdscan: --- SCAN SUMMARY 
---
@400059a9




@400059a9032f3aa86d14 simscan: clamdscan:
@400059a9032f3aa870fc simscan: clamdscan: ---simscan: cdb looking up 
version clamav
@400059a9032f3aa8a3c4  SCAN simscan: clamdscan: SIUnMfMeAcRtYe d- 
-fsimscan: runned_scanners is  attach: 1.4.0 clamav: 0.98.6/m:
@400059a9032f3aa8c304 -isimscan: found 0.98.6/m:
@400059a9032f3aa8c6ec -l-e-s-:- -0
@400059a9032f3aa8f1e4 --simscan: normal clamdscan return code: 0
@400059a9032f3aa8f1e4
@400059a9032f3aa8f5cc simscan: clamdscan: Infected fsimscan: clamdscan: 
iTliemes::  00
@400059a9032f3aa93834 .simscan: clamdscan: 1T1i8m es:e c0 .1(002  ms 
e0simscan: calling spamc
@400059a9032f3aa96ee4 c  s(0)simscan: calling /usr/bin/spamc simscan: 
clamdscan:
@400059a9032f3aa999dc  spamcm
@400059a9032f3aa999dc 0 s)
@400059a9032f3aa9ad64 

Re: [qmailtoaster] qq soft reject errors on high load

[Rajesh M]

remo / eric

i have still not being able to resolve the qq soft reject error.

these are my findings

1) the errors i see are "error forking qmail-queue" and "ripmime error" which 
causes the qq soft reject.

2) the max concurrent connections in the logs is around 35.

3)  These errors come up during peak working hours when the server is under a  
load of 4 - 8, and they increase the load even more to over 10-15.

4) i came across this link (not sure if this is related)
https://github.com/qmail/simscan/blob/master/simscan.c

5) i can share with you my live smtp logs with simscan debug.

have extracted some lines below


Error forking qmail-queue

@400059a8fa7b0a2ed1b4 tcpserver: status: 31/200
@400059a8fa7b13162584 simscan: cdb looking up version spam
@400059a8fa7b13166bd4 simscan: runned_scanners is  attach: 1.4.0 clamav: 
0.98.6/m: spam: 3.3.2
@400059a8fa7b13166fbc simscan: found 3.3.2
@400059a8fa7b13168efc simscan:[10757]:CLEAN (5.00/30.00):9.7712s:-Possible 
Spam- RE_ REQUIRE BOOKING // 1X20  //
ICD TKD TO BANDAR 
ABASS:103.241.181.228:cs@atlasdecargo.com:rathe...@radiant-india.net
@400059a8fa7b1316cd7c simscan: done, execing qmail-queue
@400059a8fa7b1316fc5c simscan: error forking qmail-queue
@400059a8fa7b13199854 simscan: exit error code: 71
@400059a8fa7b131c4004 qmail-smtpd: qq soft reject (mail server temporarily 
rejected message (#4.3.0)): MAILFROM:
RCPTTO:rathe...@y.net


Error in ripmime

@400059a8fa98045a4bc4 simscan: pelookup: domain is aissamaritime.in
@400059a8fa98045a4bc4 simscan: cdb looking up aissamaritime.in
@400059a8fa98045a4fac simscan: pelookup: local part is shailesh_k_bom
@400059a8fa98045a4fac simscan: lpart: local part is **
@400059a8fa98045a5394 simscan: cdb looking up 
shailesh_k_...@aissamaritime.in
@400059a8fa98045a5394 simscan: ripmime error
@400059a8fa98045a6334 simscan: exit error code: 71
@400059a8fa98045a95fc qmail-smtpd: qq soft reject (mail server temporarily 
rejected message (#4.3.0)):
MAILFROM: RCPTTO:shailesh_k_...@yy.in


Wierd error logs

@400059a9032f3aa79a24 simscan: clamdscan: --- SCAN SUMMARY 
---
@400059a9032f3aa7b964 simscan: clamdscan: 
/var/qmail/simscan/1504248613.321653.5221: OK
@400059a9032f3aa86d14 simscan: clamdscan:
@400059a9032f3aa870fc simscan: clamdscan: ---simscan: cdb looking up 
version clamav
@400059a9032f3aa8a3c4  SCAN simscan: clamdscan: SIUnMfMeAcRtYe d- 
-fsimscan: runned_scanners is  attach: 1.4.0 clamav: 0.98.6/m:
@400059a9032f3aa8c304 -isimscan: found 0.98.6/m:
@400059a9032f3aa8c6ec -l-e-s-:- -0
@400059a9032f3aa8f1e4 --simscan: normal clamdscan return code: 0
@400059a9032f3aa8f1e4
@400059a9032f3aa8f5cc simscan: clamdscan: Infected fsimscan: clamdscan: 
iTliemes::  00
@400059a9032f3aa93834 .simscan: clamdscan: 1T1i8m es:e c0 .1(002  ms 
e0simscan: calling spamc
@400059a9032f3aa96ee4 c  s(0)simscan: calling /usr/bin/spamc simscan: 
clamdscan:
@400059a9032f3aa999dc  spamcm
@400059a9032f3aa999dc 0 s)
@400059a9032f3aa9ad64 /var/qmail/simscan/1simscan: check_spam had an error 
ret: -1
@400059a9032f3aa9e02c 504248613.307311.5215: OK
@400059a9032f337c simscan: clamdscan:
@400059a9032f3764 simscan: clamdscan: --- SCAN SUMMARY 
---
@400059a9032f3aab3fbc simscan: clamdscan: Infected files: 0
@400059a9032f3aaba164 simscan: clamdscan: Time: 0.135 sec (0 m 0 s)
@400059a9032f3aac39bc simscan: clamdscan: /var/qmail/simscan/15simscan: 
exit error code: 71
@400059a9032f3aaca334 04248613.308469.5216: OK
@400059a9032f3aacddcc simscan: clamdscan:
@400059a9032f3aace984 simscan: clamdscan: --- SCAN SUMMARY 
---
@400059a9




@400059a9032f3aa86d14 simscan: clamdscan:
@400059a9032f3aa870fc simscan: clamdscan: ---simscan: cdb looking up 
version clamav
@400059a9032f3aa8a3c4  SCAN simscan: clamdscan: SIUnMfMeAcRtYe d- 
-fsimscan: runned_scanners is  attach: 1.4.0 clamav: 0.98.6/m:
@400059a9032f3aa8c304 -isimscan: found 0.98.6/m:
@400059a9032f3aa8c6ec -l-e-s-:- -0
@400059a9032f3aa8f1e4 --simscan: normal clamdscan return code: 0
@400059a9032f3aa8f1e4
@400059a9032f3aa8f5cc simscan: clamdscan: Infected fsimscan: clamdscan: 
iTliemes::  00
@400059a9032f3aa93834 .simscan: clamdscan: 1T1i8m es:e c0 .1(002  ms 
e0simscan: calling spamc
@400059a9032f3aa96ee4 c  s(0)simscan: calling /usr/bin/spamc simscan: 
clamdscan:
@400059a9032f3aa999dc  spamcm
@400059a9032f3aa999dc 0 s)
@400059a9032f3aa9ad64 /var/qmail/simscan/1simscan: check_spam had an error 
ret: -1
@400059a9032f3aa9e02c 504248613.307311.5215: OK
@400059a9032f337c simscan: clamdscan:
@400059a9032f3764 simscan: clamdscan: --- SCAN SUMMARY 
---
@400059a9032f3aab3fbc simscan: clamdscan: Infected files: 0

Re: [qmailtoaster] qq soft reject errors on high load

[Rajesh M]

thank for sharing your experience remo,

so would the following be correct ?

chmod 2750  simscan
chown clamav.clamav simscan

do i need to stop qmail while applying this ?

rajesh


- Original Message -
From: Remo Mattei [mailto:r...@mattei.org]
To: qmailtoaster-list@qmailtoaster.com,24x7ser...@24x7server.net
Sent: Thu, 31 Aug 2017 07:23:20 -0700
Subject:

I had the same issue with qq.

I changed the permissions and all good. I also raised the value of the
softlimits exec /usr/bin/softlimit -m 104000

and my permissions are drwxr-s---   2 clamav clamav 4096 Aug 31 07:19
simscan

So hopefully this helps.

Remo


On 8/31/17 6:51 AM, Rajesh M wrote:
> eric
>
> on my qmailtoaster server (centos 6 64 bit)  there are several domains --- 
> high traffic
>
> about 80 percent of them are routed via the sophos antispam filter and then 
> passed on to the mailserver with spamassassin/clam disabled by adding the ip 
> of the sophos server to the tcp.smtp file to exclude the line 
> /var/qmail/bin/simscan
>
> the balance 20 are directly reaching the mailserver ie mx pointed to 
> mailserver instead of sophos server and are scanned with clam / spamassassin.
>
> today i turned on the spamassassin / clam for the above 80 percent domains 
> too and this started throwing qq reject errors on  a large scale but randomly.
>
> spamassassin child-processes was set to 20 to handle the load
>
> i checked the log files for the last 30 days till yesterday and found that 
> the same qq reject errors were present but very rarely ie 3 to 4 times on a 
> single day (occasional high load ??)
> so it seems that this error comes up only when the load on the server is 
> high, all take place when there are attachments.
>
> reading the thread you sent me, i am not sure whether it applies to my case 
> but i can try out.
>
> #
> drwxr-x---  2 clamav root 6 May  6 17:57 simscan/
>
> changing this to the following fixed the problem:
> drwxr-s---  2 clamav clamav 6 May  6 17:57 simscan/
> ###
>
> should i just
> chown clamav.clamav simscan
>
>
> thanks
> rajesh
>
>
>
>
>
> - Original Message -
> From: Eric Broch [mailto:ebr...@whitehorsetc.com]
> To: 24x7ser...@24x7server.net
> Sent: Thu, 31 Aug 2017 06:42:08 -0600
> Subject:
>
> Have a look here, Rajesh, I'm not sure if it will help:
>
> https://www.mail-archive.com/qmailtoaster-list@qmailtoaster.com/msg40456.html
>
>
> On 8/31/2017 4:51 AM, Rajesh M wrote:
>> hi
>>
>> centos 6 64 bit
>> SpamAssassin version 3.3.2
>> ClamAV 0.98.6/23745/Thu Aug 31 14:07:08 2017
>>
>> we have a sophos antispam firewall which receives our emails and then routes 
>> them to our mail server (qmailtoaster, dovecot, spamassassin, clam)
>>
>> in the mail server we had stopped spamassassin
>>
>> however due to some custom requirements by our clients we enabled the same
>>
>> however we get the following error when the load on the server is high on a 
>> random basis.
>>
>> 2017-08-31 14:48:05.753994500 simscan: calling spamc
>> 2017-08-31 14:48:05.754001500 simscan: calling /usr/bin/spamc  spamc
>> 2017-08-31 14:48:05.754010500 simscan: check_spam had an error ret: -1
>> 2017-08-31 14:48:05.754101500 simscan: exit error code: 71
>> 2017-08-31 14:48:05.754220500 qmail-smtpd: qq soft reject (mail server 
>> temporarily rejected message (#4.3.0)): MAILFROM: 
>> RCPTTO:c...@.com
>>
>>
>> 2017-08-31 14:48:05.956137500 simscan: ripmime error
>> qmail-smtpd: qq soft reject (mail server temporarily rejected message
>>
>>
>>
>> if i disable simscan in tcp.smtp then errors do not take place.
>>
>> i enabled simscan debug and this is the result.
>>
>> 2017-08-31 14:48:05.956144500 qmail-smtpd: qq soft reject (mail server 
>> temporarily rejected message (#4.3.0)): MAILFROM: 
>> RCPTTO:inv1...@.net
>>
>>
>> clam=yes,spam=yes,spam_hits=30,spam_passthru=yes,attach=.ace:.arc:.arj:.b64:.bat:.bhx:.cab:.chm:.com:.cpl:.dll:.exe:.gz:.hqx:.hta:.inf:.ins:.iso:.isp:.jse:.lib:.lnk:.lzh:.mim:.msp:.mst:.pif:.reg:.scf:.scr:.sct:.shb:.shs:.sys:.taz:.tgz:.tz:.url:.uu:.uue:.vb:.vbe:.vbs:.wsc:.wsf:.wsh:.xxe:.docm:.z:.jar
>> 2017-08-31 14:48:05.746592500 simscan: pelookup clam = yes
>> 2017-08-31 14:48:05.746593500 simscan: pelookup spam = yes
>> 2017-08-31 14:48:05.746593500 simscan: pelookup spam_hits = 30
>> 2017-08-31 14:48:05.746595500 simscan: Per Domain Hits set to : 30.00
>> 2017-08-31 14:48:05.746595500 simscan: pelookup spam_passthru = yes
>> 2017-08-31 14:48:05.746596500 simscan: unimplemented flag spam_passthru = yes
>> 2017-08-31 14:48:05.746596500 simscan: pelookup attach = 
>> .ace:.arc:.arj:.b64:.bat:.bhx:.cab:.chm:.com:.cpl:.dll:.exe:.gz:.hqx:.hta:.inf:.ins:.iso:.isp:.jse:.lib:.lnk:.lzh:.mim:.msp:.mst:.pif:.reg:.scf:.scr:.sct:.shb:.shs:.sys:.taz:.tgz:.tz:.url:.uu:.uue:.vb:.vbe:.vbs:.wsc:.wsf:.wsh:.xxe:.docm:.z:.jar
>> 2017-08-31 14:48:05.746599500 simscan: attachment flag attach = 
>> 

Re: [qmailtoaster] qq soft reject errors on high load

[Remo Mattei]

I had the same issue with qq.

I changed the permissions and all good. I also raised the value of the
softlimits exec /usr/bin/softlimit -m 104000

and my permissions are drwxr-s---   2 clamav clamav 4096 Aug 31 07:19
simscan

So hopefully this helps.

Remo


On 8/31/17 6:51 AM, Rajesh M wrote:
> eric
>
> on my qmailtoaster server (centos 6 64 bit)  there are several domains --- 
> high traffic
>
> about 80 percent of them are routed via the sophos antispam filter and then 
> passed on to the mailserver with spamassassin/clam disabled by adding the ip 
> of the sophos server to the tcp.smtp file to exclude the line 
> /var/qmail/bin/simscan
>
> the balance 20 are directly reaching the mailserver ie mx pointed to 
> mailserver instead of sophos server and are scanned with clam / spamassassin.
>
> today i turned on the spamassassin / clam for the above 80 percent domains 
> too and this started throwing qq reject errors on  a large scale but randomly.
>
> spamassassin child-processes was set to 20 to handle the load
>
> i checked the log files for the last 30 days till yesterday and found that 
> the same qq reject errors were present but very rarely ie 3 to 4 times on a 
> single day (occasional high load ??)
> so it seems that this error comes up only when the load on the server is 
> high, all take place when there are attachments.
>
> reading the thread you sent me, i am not sure whether it applies to my case 
> but i can try out.
>
> #
> drwxr-x---  2 clamav root 6 May  6 17:57 simscan/
>
> changing this to the following fixed the problem:
> drwxr-s---  2 clamav clamav 6 May  6 17:57 simscan/
> ###
>
> should i just 
> chown clamav.clamav simscan
>
>
> thanks
> rajesh
>
>
>
>
>
> - Original Message -
> From: Eric Broch [mailto:ebr...@whitehorsetc.com]
> To: 24x7ser...@24x7server.net
> Sent: Thu, 31 Aug 2017 06:42:08 -0600
> Subject: 
>
> Have a look here, Rajesh, I'm not sure if it will help:
>
> https://www.mail-archive.com/qmailtoaster-list@qmailtoaster.com/msg40456.html
>
>
> On 8/31/2017 4:51 AM, Rajesh M wrote:
>> hi
>>
>> centos 6 64 bit
>> SpamAssassin version 3.3.2
>> ClamAV 0.98.6/23745/Thu Aug 31 14:07:08 2017
>>
>> we have a sophos antispam firewall which receives our emails and then routes 
>> them to our mail server (qmailtoaster, dovecot, spamassassin, clam)
>>
>> in the mail server we had stopped spamassassin
>>
>> however due to some custom requirements by our clients we enabled the same
>>
>> however we get the following error when the load on the server is high on a 
>> random basis.
>>
>> 2017-08-31 14:48:05.753994500 simscan: calling spamc
>> 2017-08-31 14:48:05.754001500 simscan: calling /usr/bin/spamc  spamc
>> 2017-08-31 14:48:05.754010500 simscan: check_spam had an error ret: -1
>> 2017-08-31 14:48:05.754101500 simscan: exit error code: 71
>> 2017-08-31 14:48:05.754220500 qmail-smtpd: qq soft reject (mail server 
>> temporarily rejected message (#4.3.0)): MAILFROM: 
>> RCPTTO:c...@.com
>>
>>
>> 2017-08-31 14:48:05.956137500 simscan: ripmime error
>> qmail-smtpd: qq soft reject (mail server temporarily rejected message
>>
>>
>>
>> if i disable simscan in tcp.smtp then errors do not take place.
>>
>> i enabled simscan debug and this is the result.
>>
>> 2017-08-31 14:48:05.956144500 qmail-smtpd: qq soft reject (mail server 
>> temporarily rejected message (#4.3.0)): MAILFROM: 
>> RCPTTO:inv1...@.net
>>
>>
>> clam=yes,spam=yes,spam_hits=30,spam_passthru=yes,attach=.ace:.arc:.arj:.b64:.bat:.bhx:.cab:.chm:.com:.cpl:.dll:.exe:.gz:.hqx:.hta:.inf:.ins:.iso:.isp:.jse:.lib:.lnk:.lzh:.mim:.msp:.mst:.pif:.reg:.scf:.scr:.sct:.shb:.shs:.sys:.taz:.tgz:.tz:.url:.uu:.uue:.vb:.vbe:.vbs:.wsc:.wsf:.wsh:.xxe:.docm:.z:.jar
>> 2017-08-31 14:48:05.746592500 simscan: pelookup clam = yes
>> 2017-08-31 14:48:05.746593500 simscan: pelookup spam = yes
>> 2017-08-31 14:48:05.746593500 simscan: pelookup spam_hits = 30
>> 2017-08-31 14:48:05.746595500 simscan: Per Domain Hits set to : 30.00
>> 2017-08-31 14:48:05.746595500 simscan: pelookup spam_passthru = yes
>> 2017-08-31 14:48:05.746596500 simscan: unimplemented flag spam_passthru = yes
>> 2017-08-31 14:48:05.746596500 simscan: pelookup attach = 
>> .ace:.arc:.arj:.b64:.bat:.bhx:.cab:.chm:.com:.cpl:.dll:.exe:.gz:.hqx:.hta:.inf:.ins:.iso:.isp:.jse:.lib:.lnk:.lzh:.mim:.msp:.mst:.pif:.reg:.scf:.scr:.sct:.shb:.shs:.sys:.taz:.tgz:.tz:.url:.uu:.uue:.vb:.vbe:.vbs:.wsc:.wsf:.wsh:.xxe:.docm:.z:.jar
>> 2017-08-31 14:48:05.746599500 simscan: attachment flag attach = 
>> .ace:.arc:.arj:.b64:.bat:.bhx:.cab:.chm:.com:.cpl:.dll:.exe:.gz:.hqx:.hta:.inf:.ins:.iso:.isp:.jse:.lib:.lnk:.lzh:.mim:.msp:.mst:.pif:.reg:.scf:.scr:.sct:.shb:.shs:.sys:.taz:.tgz:.tz:.url:.uu:.uue:.vb:.vbe:.vbs:.wsc:.wsf:.wsh:.xxe:.docm:.z:.jar
>> 2017-08-31 14:48:05.746602500 simscan: add_attach called with 
>> 

Re: [qmailtoaster] qq soft reject errors on high load

[Rajesh M]

eric

on my qmailtoaster server (centos 6 64 bit)  there are several domains --- high 
traffic

about 80 percent of them are routed via the sophos antispam filter and then 
passed on to the mailserver with spamassassin/clam disabled by adding the ip of 
the sophos server to the tcp.smtp file to exclude the line 
/var/qmail/bin/simscan

the balance 20 are directly reaching the mailserver ie mx pointed to mailserver 
instead of sophos server and are scanned with clam / spamassassin.

today i turned on the spamassassin / clam for the above 80 percent domains too 
and this started throwing qq reject errors on  a large scale but randomly.

spamassassin child-processes was set to 20 to handle the load

i checked the log files for the last 30 days till yesterday and found that the 
same qq reject errors were present but very rarely ie 3 to 4 times on a single 
day (occasional high load ??)
so it seems that this error comes up only when the load on the server is high, 
all take place when there are attachments.

reading the thread you sent me, i am not sure whether it applies to my case but 
i can try out.

#
drwxr-x---  2 clamav root 6 May  6 17:57 simscan/

changing this to the following fixed the problem:
drwxr-s---  2 clamav clamav 6 May  6 17:57 simscan/
###

should i just
chown clamav.clamav simscan


thanks
rajesh





- Original Message -
From: Eric Broch [mailto:ebr...@whitehorsetc.com]
To: 24x7ser...@24x7server.net
Sent: Thu, 31 Aug 2017 06:42:08 -0600
Subject:

Have a look here, Rajesh, I'm not sure if it will help:

https://www.mail-archive.com/qmailtoaster-list@qmailtoaster.com/msg40456.html


On 8/31/2017 4:51 AM, Rajesh M wrote:
> hi
>
> centos 6 64 bit
> SpamAssassin version 3.3.2
> ClamAV 0.98.6/23745/Thu Aug 31 14:07:08 2017
>
> we have a sophos antispam firewall which receives our emails and then routes 
> them to our mail server (qmailtoaster, dovecot, spamassassin, clam)
>
> in the mail server we had stopped spamassassin
>
> however due to some custom requirements by our clients we enabled the same
>
> however we get the following error when the load on the server is high on a 
> random basis.
>
> 2017-08-31 14:48:05.753994500 simscan: calling spamc
> 2017-08-31 14:48:05.754001500 simscan: calling /usr/bin/spamc  spamc
> 2017-08-31 14:48:05.754010500 simscan: check_spam had an error ret: -1
> 2017-08-31 14:48:05.754101500 simscan: exit error code: 71
> 2017-08-31 14:48:05.754220500 qmail-smtpd: qq soft reject (mail server 
> temporarily rejected message (#4.3.0)): MAILFROM: 
> RCPTTO:c...@.com
>
>
> 2017-08-31 14:48:05.956137500 simscan: ripmime error
> qmail-smtpd: qq soft reject (mail server temporarily rejected message
>
>
>
> if i disable simscan in tcp.smtp then errors do not take place.
>
> i enabled simscan debug and this is the result.
>
> 2017-08-31 14:48:05.956144500 qmail-smtpd: qq soft reject (mail server 
> temporarily rejected message (#4.3.0)): MAILFROM: 
> RCPTTO:inv1...@.net
>
>
> clam=yes,spam=yes,spam_hits=30,spam_passthru=yes,attach=.ace:.arc:.arj:.b64:.bat:.bhx:.cab:.chm:.com:.cpl:.dll:.exe:.gz:.hqx:.hta:.inf:.ins:.iso:.isp:.jse:.lib:.lnk:.lzh:.mim:.msp:.mst:.pif:.reg:.scf:.scr:.sct:.shb:.shs:.sys:.taz:.tgz:.tz:.url:.uu:.uue:.vb:.vbe:.vbs:.wsc:.wsf:.wsh:.xxe:.docm:.z:.jar
> 2017-08-31 14:48:05.746592500 simscan: pelookup clam = yes
> 2017-08-31 14:48:05.746593500 simscan: pelookup spam = yes
> 2017-08-31 14:48:05.746593500 simscan: pelookup spam_hits = 30
> 2017-08-31 14:48:05.746595500 simscan: Per Domain Hits set to : 30.00
> 2017-08-31 14:48:05.746595500 simscan: pelookup spam_passthru = yes
> 2017-08-31 14:48:05.746596500 simscan: unimplemented flag spam_passthru = yes
> 2017-08-31 14:48:05.746596500 simscan: pelookup attach = 
> .ace:.arc:.arj:.b64:.bat:.bhx:.cab:.chm:.com:.cpl:.dll:.exe:.gz:.hqx:.hta:.inf:.ins:.iso:.isp:.jse:.lib:.lnk:.lzh:.mim:.msp:.mst:.pif:.reg:.scf:.scr:.sct:.shb:.shs:.sys:.taz:.tgz:.tz:.url:.uu:.uue:.vb:.vbe:.vbs:.wsc:.wsf:.wsh:.xxe:.docm:.z:.jar
> 2017-08-31 14:48:05.746599500 simscan: attachment flag attach = 
> .ace:.arc:.arj:.b64:.bat:.bhx:.cab:.chm:.com:.cpl:.dll:.exe:.gz:.hqx:.hta:.inf:.ins:.iso:.isp:.jse:.lib:.lnk:.lzh:.mim:.msp:.mst:.pif:.reg:.scf:.scr:.sct:.shb:.shs:.sys:.taz:.tgz:.tz:.url:.uu:.uue:.vb:.vbe:.vbs:.wsc:.wsf:.wsh:.xxe:.docm:.z:.jar
> 2017-08-31 14:48:05.746602500 simscan: add_attach called with 
> .ace:.arc:.arj:.b64:.bat:.bhx:.cab:.chm:.com:.cpl:.dll:.exe:.gz:.hqx:.hta:.inf:.ins:.iso:.isp:.jse:.lib:.lnk:.lzh:.mim:.msp:.mst:.pif:.reg:.scf:.scr:.sct:.shb:.shs:.sys:.taz:.tgz:.tz:.url:.uu:.uue:.vb:.vbe:.vbs:.wsc:.wsf:.wsh:.xxe:.docm:.z:.jar
> 2017-08-31 14:48:05.746606500 simscan: .ace is attachment number 0
> 2017-08-31 14:48:05.746607500 simscan: .arc is attachment number 1
> 2017-08-31 14:48:05.746607500 simscan: .arj is attachment number 2
> 2017-08-31 14:48:05.746607500 simscan: .b64 is attachment number 3
> 2017-08-31 

[qmailtoaster] qq soft reject errors on high load

[Rajesh M]

hi

centos 6 64 bit
SpamAssassin version 3.3.2
ClamAV 0.98.6/23745/Thu Aug 31 14:07:08 2017

we have a sophos antispam firewall which receives our emails and then routes 
them to our mail server (qmailtoaster, dovecot, spamassassin, clam)

in the mail server we had stopped spamassassin

however due to some custom requirements by our clients we enabled the same

however we get the following error when the load on the server is high on a 
random basis.

2017-08-31 14:48:05.753994500 simscan: calling spamc
2017-08-31 14:48:05.754001500 simscan: calling /usr/bin/spamc  spamc
2017-08-31 14:48:05.754010500 simscan: check_spam had an error ret: -1
2017-08-31 14:48:05.754101500 simscan: exit error code: 71
2017-08-31 14:48:05.754220500 qmail-smtpd: qq soft reject (mail server 
temporarily rejected message (#4.3.0)): MAILFROM: 
RCPTTO:c...@.com


2017-08-31 14:48:05.956137500 simscan: ripmime error
qmail-smtpd: qq soft reject (mail server temporarily rejected message



if i disable simscan in tcp.smtp then errors do not take place.

i enabled simscan debug and this is the result.

2017-08-31 14:48:05.956144500 qmail-smtpd: qq soft reject (mail server 
temporarily rejected message (#4.3.0)): MAILFROM: 
RCPTTO:inv1...@.net


clam=yes,spam=yes,spam_hits=30,spam_passthru=yes,attach=.ace:.arc:.arj:.b64:.bat:.bhx:.cab:.chm:.com:.cpl:.dll:.exe:.gz:.hqx:.hta:.inf:.ins:.iso:.isp:.jse:.lib:.lnk:.lzh:.mim:.msp:.mst:.pif:.reg:.scf:.scr:.sct:.shb:.shs:.sys:.taz:.tgz:.tz:.url:.uu:.uue:.vb:.vbe:.vbs:.wsc:.wsf:.wsh:.xxe:.docm:.z:.jar
2017-08-31 14:48:05.746592500 simscan: pelookup clam = yes
2017-08-31 14:48:05.746593500 simscan: pelookup spam = yes
2017-08-31 14:48:05.746593500 simscan: pelookup spam_hits = 30
2017-08-31 14:48:05.746595500 simscan: Per Domain Hits set to : 30.00
2017-08-31 14:48:05.746595500 simscan: pelookup spam_passthru = yes
2017-08-31 14:48:05.746596500 simscan: unimplemented flag spam_passthru = yes
2017-08-31 14:48:05.746596500 simscan: pelookup attach = 
.ace:.arc:.arj:.b64:.bat:.bhx:.cab:.chm:.com:.cpl:.dll:.exe:.gz:.hqx:.hta:.inf:.ins:.iso:.isp:.jse:.lib:.lnk:.lzh:.mim:.msp:.mst:.pif:.reg:.scf:.scr:.sct:.shb:.shs:.sys:.taz:.tgz:.tz:.url:.uu:.uue:.vb:.vbe:.vbs:.wsc:.wsf:.wsh:.xxe:.docm:.z:.jar
2017-08-31 14:48:05.746599500 simscan: attachment flag attach = 
.ace:.arc:.arj:.b64:.bat:.bhx:.cab:.chm:.com:.cpl:.dll:.exe:.gz:.hqx:.hta:.inf:.ins:.iso:.isp:.jse:.lib:.lnk:.lzh:.mim:.msp:.mst:.pif:.reg:.scf:.scr:.sct:.shb:.shs:.sys:.taz:.tgz:.tz:.url:.uu:.uue:.vb:.vbe:.vbs:.wsc:.wsf:.wsh:.xxe:.docm:.z:.jar
2017-08-31 14:48:05.746602500 simscan: add_attach called with 
.ace:.arc:.arj:.b64:.bat:.bhx:.cab:.chm:.com:.cpl:.dll:.exe:.gz:.hqx:.hta:.inf:.ins:.iso:.isp:.jse:.lib:.lnk:.lzh:.mim:.msp:.mst:.pif:.reg:.scf:.scr:.sct:.shb:.shs:.sys:.taz:.tgz:.tz:.url:.uu:.uue:.vb:.vbe:.vbs:.wsc:.wsf:.wsh:.xxe:.docm:.z:.jar
2017-08-31 14:48:05.746606500 simscan: .ace is attachment number 0
2017-08-31 14:48:05.746607500 simscan: .arc is attachment number 1
2017-08-31 14:48:05.746607500 simscan: .arj is attachment number 2
2017-08-31 14:48:05.746607500 simscan: .b64 is attachment number 3
2017-08-31 14:48:05.746610500 simscan: .bat is attachment number 4
2017-08-31 14:48:05.746610500 simscan: .bhx is attachment number 5
2017-08-31 14:48:05.746610500 simscan: .cab is attachment number 6
2017-08-31 14:48:05.746611500 simscan: .chm is attachment number 7
2017-08-31 14:48:05.746611500 simscan: .com is attachment number 8
2017-08-31 14:48:05.746611500 simscan: .cpl is attachment number 9
2017-08-31 14:48:05.746611500 simscan: .dll is attachment number 10
2017-08-31 14:48:05.746612500 simscan: .exe is attachment number 11
2017-08-31 14:48:05.746614500 simscan: .gz is attachment number 12
2017-08-31 14:48:05.746614500 simscan: .hqx is attachment number 13
2017-08-31 14:48:05.746614500 simscan: .hta is attachment number 14
2017-08-31 14:48:05.746615500 simscan: .inf is attachment number 15
2017-08-31 14:48:05.746615500 simscan: .ins is attachment number 16
2017-08-31 14:48:05.746615500 simscan: .iso is attachment number 17
2017-08-31 14:48:05.746616500 simscan: .isp is attachment number 18
2017-08-31 14:48:05.746616500 simscan: .jse is attachment number 19
2017-08-31 14:48:05.746618500 simscan: .lib is attachment number 20
2017-08-31 14:48:05.746618500 simscan: .lnk is attachment number 21
2017-08-31 14:48:05.746619500 simscan: .lzh is attachment number 22
2017-08-31 14:48:05.746622500 simscan: .mim is attachment number 23
2017-08-31 14:48:05.746622500 simscan: .msp is attachment number 24
2017-08-31 14:48:05.746623500 simscan: .mst is attachment number 25
2017-08-31 14:48:05.746623500 simscan: .pif is attachment number 26
2017-08-31 14:48:05.746623500 simscan: .reg is attachment number 27
2017-08-31 14:48:05.746624500 simscan: .scf is attachment number 28
2017-08-31 14:48:05.746624500 simscan: .scr is attachment number 29
2017-08-31 14:48:05.746624500 simscan: .sct is