P3scan stop to work

Apr 27 18:26:23 proxy p3scan[32149]: ERR: Exiting now...
Apr 27 18:26:23 proxy p3scan: ERR: Sending mail to client
Apr 27 18:26:23 proxy p3scan[31216]: ERR: Attention: child with pid 32149 died with abnormal termsignal (11)! This is probably a bug. Please report to the author. numprocs is now 1
Apr 27 18:26:23 proxy kernel: p3scan[32149]: segfault at 241 ip 00007f742e013694 sp 00007ffed04c7110 error 4 in libc-2.17.so[7f742dfcb000+1b6000]
Apr 27 18:26:28 proxy p3scan[31216]: P3Scan terminates now
Apr 27 18:26:28 proxy p3scan[25355]: ERR: We cot SIGTERM!

After 2 days of work, pop3 client stop to download mail.
When disable and reenable pop3 proxy, client connection restart.

i have more user with bad habits to do it continuously send and receive
It is just the tick to the finger tick-tick-tick-tick-tick…

Today it is going to crash continuously
I saw on sourceforge that version 2.3.2 is 2005 :cry:

i have insert a cron in > /etc/crontab

5 * * * * * root /root/p3scan.sh

with

#:/bin/sh
systemctl restart p3scan

May 23 15:51:37 proxy p3scan[5495]: ERR: Attention: child with pid 5800 died with abnormal termsignal (11)! This is probably a bug. Please report to the author. numprocs is now 0
May 23 15:51:37 proxy kernel: p3scan[5800]: segfault at 241 ip 00007fcf2db1a694 sp 00007ffd5038f460 error 4 in libc-2.17.so[7fcf2dad2000+1b6000]

Thank you for sharing the workaround.

Yes the project is dead, very dead :smiley:
AFAIK there is no valid alternative.

new log message

WARNING: Your scanner returned neither 0, a viruscode, nor a good viruscode, but 2
May 29 09:08:21 proxy p3scan[10402]: ERR: We can't say if it is a virus! So we have to give the client the mail! You should check your configuration/system
May 29 09:08:21 proxy p3scan: ERR: Scanner returned unexpected error code. You should check your configuration/system.
May 29 09:08:21 proxy p3scan[10402]: ERR: Scanner returned unexpected error code. You should check your configuration/system.
May 29 09:08:21 proxy p3scan[10402]: ERR: Exiting now...
May 29 09:08:21 proxy p3scan[10383]: ERR: Attention: child with pid 10402 died with abnormal termsignal (11)! This is probably a bug. Please report to the author. numprocs is now 0
May 29 09:08:21 proxy kernel: p3scan[10402]: segfault at 241 ip 00007f66a26e3694 sp 00007ffd504c6340 error 4 in libc-2.17.so[7f66a269b000+1b7000]
May 29 09:08:22 proxy p3scan[10383]: P3Scan terminates now

linked problem? ClamAV ERROR: Failed to load new database: Malformed database

@filippo_carletti @davidep any clue on this?

Probably clamd died. It may happen because of a ram shortage.

Perhaps it is the only explanation [quote=“filippo_carletti, post:6, topic:6696”]
ram shortage.
[/quote]

Today or tomorrow migrate proxy on a xeon server 32 gb raid 5, then let’s see if the ram ends

Ok, this morning i have completed the migration on a new machine. For now no problem. Thanks all

1 Like

Looks like I got bitten by another manifestation. It’s not the issue I previously reported as that only killed the worker thread. And I don’t see any other issues, like clamd dying.

 p3scan[15954]: POP3S Connection from 192.168.0.3:55269
 p3scan[15954]: Real-server address is 216.55.149.49:995
 p3scan[15954]: Ignoring servers TOP capability...
 p3scan[15954]: Ignoring servers STLS capability...
 p3scan[15954]: USER '**********@********.***'
 p3scan[15954]: ERR: Sending mail to client
 p3scan[15954]: ERR: Exiting now...
 p3scan[1066]: ERR: Sending mail to client
 p3scan[1071]: ERR: Attention: child with pid 15954 died with abnormal termsignal (11)! This is probably a bug. Please report to the author. numprocs is now 4
 p3scan[1071]: P3Scan terminates now
 p3scan[15940]: ERR: We cot SIGTERM!
 p3scan[2754]: ERR: We cot SIGTERM!
 p3scan[9374]: ERR: We cot SIGTERM!
 p3scan[8666]: ERR: We cot SIGTERM!

Cheers.

Which kind of client is 192.168.0.3 ? I guess that could put you on the right track

It’s Windows 7 running Outlook 2013 and it runs on average 8 -> 10 hours per day, so I don’t think that could be the issue otherwise I’d be seeing this a lot more.

Cheers.

update:
another crash of p3scan.
the problem (i think) isnt ram.

free -m
              total        used        free      shared  buff/cache   available
Mem:          23933        3782       17698          90        2452       19601
Swap:         11443           0       11443

i have insert a better cron with check on process pid every minute

nano /root/p3scan-check.sh

    #!/bin/bash
    ps ax |grep -v grep |grep p3scan.pid >  /dev/null
    if [[ $? -ne 0 ]] ; then
            echo "Restarting P3scan:   $(date)" >> /var/log/p3scan-check
            /root/p3scan.sh &
    fi

nano /root/p3scan.sh

#!/bin/sh
systemctl restart p3scan

nano /etc/crontab
*/1 * * * * root /root/p3scan-check.sh

does the outlook leave the mail in the external server for “x” days? Try not to and purge the “inbox” folder till it has no mail. I had the same problem with fetchmail and my final solution for 100% working p3scan was to NOT leave messages in the server and ALWAYS retrieve all messages. P3Scan has problem with POP3 “TOP” command.

You could try adding enabletop in p3scan.conf.

Does it happen always with the same client IP? I saw a similar scenario with always the same IP.

Yes! here more user leave message on server.

I have boxes with at least 3-4 gb of mail
So p3scan every time reads all inboxes?

done!

In fact they are always the same boxes, at least a couple

Thanks all