Exim Broke

[tmallardi]

Exim broke last night, presumable after the cPanel update. Below is an example from the exim_mainlog.

failed to expand ACL string
"${if eq {${acl_m0}}{1}{0}{${perl{acl_checksa_deliver}{$domain}{${extract{5}{:}
{${lookup{${lookup{$domain}lsearch*{/etc/userdomains}{$value}}}lsearch{/etc/pass
wd}{$value}}}}}}}}": Undefined subroutine &main::acl_checksa_deliver called.

External mail is being rejected.

I'm currently on FC2 by the way. (Yes, I will upgrade very soon)
I have MailScanner and Firewall installed by Chirpy.

Any suggestions are greatly appreciated.

[tmallardi]

Exim broke last night, presumable after the cPanel update. Below is an example from the exim_mainlog.

failed to expand ACL string
"${if eq {${acl_m0}}{1}{0}{${perl{acl_checksa_deliver}{$dom ain}{${extract{5}{:}
{${lookup{${lookup{$domain}lsearch*{/etc/userdomains}{$value}}}lsearch{/etc/pass
wd}{$value}}}}}}}}": Undefined subroutine &main::acl_checksa_deliver called.

External mail is being rejected.

I'm currently on FC2 by the way. (Yes, I will upgrade very soon)
I have MailScanner and Firewall installed by Chirpy.

Any suggestions are greatly appreciated.

[cPanelDavidG]

** Merged Duplicate Threads **

[tmallardi]

My Hosting provider figure out a work around...

"Service Configuration" --> "Exim Configuration Editor" in the left panel, you will see a checkbox in the main window with the following line preceding it:

"Use the old transport based spamassassin system instead of the new acl style one. (not recommended, slow)"


Looks like he bypassed Chirpy's MailScanner.

I will proceed with getting a new server now....

[Bailey]

I can confirm this issue also exists on CENTOS Enterprise 3.9.

Client is running cPanel 11.15.0-C17938, and the latest CS Mailscanner. Identical issue & errors as of last night's cPanel update.

Bailey

[cpanelkenneth]

YOu need to rebuild exim.conf. /scripts/buildeximconf should do it.

[rpage]

I did /scripts/buildeximconf, restarted exim and restarted MailScanner and I still get the same error. Please more suggestions.
Rick

[tmallardi]

same here... the buildeximconf script did not fix it.

I'm just going to leave the old transport turned on until I can migrate to a new OS.

[rpage]

I did a force cpanel update, and it seems to be working now.
He is the version I am running, WHM 11.11.0 cPanel 11.15.0-C17938.
Rick

[Bailey]

Here is the advice prescribed by ConfigServer:

We've seen this happen to a couple of people, looks like cPanel may have messed up an exim configuration file update. I think the solution is to reset the ACL configuration to the defaults. In WHM - Exim Configuration Editor, scroll to the bottom and click Reset ACL Config to defaults. If there were any modifications to the exim configuration you should add them back in. In this case I think it would only be the exim dictionary attack ACL, which you can get from the instructions here:
http://www.configserver.com/free/eximdeny.html

You only need to do the part referring to the WHM Exim Configuration Editor, not the bits that are done in shell as they should already be there, assuming that a MailScanner Service was done on this server.

Hope this is helpful!!

Bailey

[l3git]

do we have a fix for this.. happened to a server of mine a few days ago and today on another...

[nickp666]

I had this as well, from memory (but not entirely sure - was half asleep when it happened) im sure it was the 'Ratelimit incoming SMTP connections that violate RFCs (this is usually spammers and broken mta)' and 'Trusted Mail Host Whitelist (/etc/trustedmailhosts) -- bypass spam checks' checkboxes activating themselves when I ran upcp, failing that it was definately one of the ACL conditions, as I already had HELO acls in place, those options conflict with them.