PHP Send Mail Crashes EXIM

P

polo

Verified User
Joined
Apr 2, 2005
Messages
122
when php send mail sends mail to any domain on my server it works fine, but when it sends out side of domains on the server such as gmail hotmail etc... it starts to loop exim 2 99% processes each time before i know loads are out the roof. Any ideas?
 
/var/log/exim/mainlog
/var/log/exim/paniclog
/var/log/exim/rejectlog
/var/log/messages
 
I checked all of them nothing seeems out of the ordinary....
 
maillog nothing either... anytime wordpress or anything tries to send mail i get two proccesses that appear in the high 90%'s. Anyone else have any ideas? Please it sucks having to restart exim every 20 mins.
 
Ive tried the update thing with the exim.conf from DA. Ive watched panic log, i dont see anything weird in php.ini, this is driving me nuts. John any ideas?
 
Another piece of the puzzle if this helps. I found that writing php send mail works run from the shell php mail.php but if i access the same file from the webserver http://nerdytechs.com/mail.php it causes exim to loop and does not get sent.
 
Thanks John,

I followed the steps. Now the mail is not locking up exim anymore when sending from httpd, but it is not getting sent, when i check the exim queue the are "frozen" any ideas?

mail( '[email protected]', 'This is a test email', 'This is the message in the
email', 'From: [email protected]');
0m 462 1N1MUt-000572-FU <[email protected]> *** frozen ***
[email protected]

Mainlog spits out this:

2009-10-23 12:09:11 1N1MhT-0005QK-3h == [email protected] R=lookuphost T=remote_smtp defer (-1): smtp transport process returned non-zero status 0x000b: terminated by signal 11
2009-10-23 12:09:11 1N1MhT-0005QK-3h Frozen
2009-10-23 12:09:12 1N1MhU-0005QO-AT <= [email protected] U=apache P=local S=462 T="This is a test email" from <[email protected]> for [email protected]

I've tried this too http://help.directadmin.com/item.php?id=114 but still no luck still freezing up.
 
Last edited:
Signal 11 still means something is messed up.

I would just run /usr/local/directadmin/scripts/exim.sh

Then follow:

http://help.directadmin.com/item.php?id=51

If that doesnt fix it then maybe something is messed up with php.
Click to expand...

After running that im back to square 1 again processes are looping... and overlaoding the server

PID USERNAME THR PRI NICE SIZE RES STATE C TIME WCPU COMMAND
26473 mail 1 113 0 7272K 3960K CPU3 3 3:06 100.69% exim
26472 root 1 113 0 7272K 3960K CPU2 2 3:06 100.64% exim
 
Last edited:
i dont think its php i can run php mail.php from the shell and it runs fine its jsut when its run through the webserver that it doesnt work.
 
yeah but the shell php might be cli and the web one cgi?
 
Yeah not sure if thats really gonna help. Waiting to see what John has to say.
 
------- Comment #4 From Floris van den Berg 2006-10-04 23:06:13 -------

Recompiling exim fixed it for me. Perhaps try to recompile php as well. The
problem was that exim was linking with the wrong openssl version. This may be
the case for php as well.

You should read here:

http://bugs.exim.org/show_bug.cgi?id=384
 
cd /usr/local/directadmin/custombuild
./build update
./build php d
 
You must log in or register to reply here.
Back
Top