Can't send email Off the server using PHP mail()

motobrandt

Verified User
Joined
Jan 8, 2004
Messages
217
Yesterday my server underwent some health checks (thanks Martynas ;) ) one of which included upgrading exim from 4.60 to 4.66 using the rpm from http://files.directadmin.com/services/es_4.1_64/. Ever since then my webstores cannot send receipts to users that do not reside on the server (ie gmail.com)

I get this in the /var/log/httpd/error_log for the message.

T=remote_smtp defer (-1): smtp transport process returned non-zero status 0x000b: terminated by signal 11

I searched and tried this solution http://www.directadmin.com/help/item.php?id=114

Also tried this solution. http://help.directadmin.com/item.php?id=9

No luck yet. Can someone give me any ideas? Or should I just email support? :D

thanks
 
Last edited:

MadHag

Verified User
Joined
Aug 26, 2008
Messages
79
Location
French Pyrenees
If it ain't broke don't try and fix it.

Sure you've heard than one before. Sorry to rub salt in the wound.

David
 

nobaloney

NoBaloney Internet Svcs - In Memoriam †
Joined
Jun 16, 2003
Messages
26,119
Location
California
The version of Exim shouldn't affect whether or not gmail likes your emails; that should be based on DNS and settings in your exim.conf file.

Jeff
 

motobrandt

Verified User
Joined
Jan 8, 2004
Messages
217
The version of Exim shouldn't affect whether or not gmail likes your emails; that should be based on DNS and settings in your exim.conf file.

Jeff
Is that what that error is? A reject from the recipient? It happens so fast in the logs I figured it was a local configuration problem.

I've tried everything I can think of here. I guess I'll try to fallback to the previous version of exim to see if that fixes it.

Still open to suggestions. ;)
 

Chrysalis

Verified User
Joined
Aug 25, 2004
Messages
1,584
Location
uk
I am seeing the exact same issue on one server, it has only started happening recently.

All mails from php apps fail with T=remote_smtp defer (-1): smtp transport process returned non-zero status 0x000b: terminated by signal 11

If retry the delivery from mail queue they deliver fine.

The server in question is still using customapache, php version is 5.2.6 and apache 2.2.10.

I am scratching my hair over it as have tried a lot of things such as.

reverting to default DA exim config template.
trying different exim versions.
disabling mod security.
recompiling php and apache.
checked file descriptors as described in the DA knowledgebase.
checked file inodes which according to dh -i are fine.
bypassed clamav and spamassassin.
all mail sent via smtp and even squirrelmail is fine.

Server is running freebsd 6, latest DA version.
 

nobaloney

NoBaloney Internet Svcs - In Memoriam †
Joined
Jun 16, 2003
Messages
26,119
Location
California
Chrysalis,

It's the remote server deferring your email. Is it happening with all remote servers? Or just one or some?

You write that when you retry from the queue they deliver fine. Do they deliver from the queue without your intervention?

It could be something as simply as greylisting by the recipient server.

Jeff
 

Chrysalis

Verified User
Joined
Aug 25, 2004
Messages
1,584
Location
uk
It is basically any remote server, whether it been my own smtp server or gamil/yahoo or some other 3rd party.

We did some searching on this forum and seen it is an issue that has come up before multiple times, php and apache have been both recompiled with the larger file descriptors and we still using the binary for exim supplied by directadmin, the problem remained.

In the end we have modified the vhost template for apache so it doesnt instantly try sending anymore and adds to the queue as unfrozen instead, I then crontabbed a exim queue run command to run every 5 minutes (only on unfrozen messages), this seems to be a good workaround. the vhost change was swapping as below.

/usr/sbin/sendmail -t -i -f
to
/usr/sbin/sendmail -t -odq -f

Out of many servers it is the first time I have seen this occur.
 
Top