Mitel 5000 V6 SP1 email exim socket bind to port 25

[telcoengineer]

We recently installed 3 5000 systems networked, we are also using voicemail networking (VPIM). at Node 1 I am constantly seeing errors n the mail_mainlog. Emails from a VPIM mailbox as well as just voicemail to email delivery does not get sent out until the system is rebooted and then shortly after boot up it stops again.

Aug  4 21:16:56 Voiplewis exim[1980]: socket bind() to port 25 for address (any IPv4) failed: Address already in use: waiting 30s before trying again (1 more try)
Aug  4 21:17:26 Voiplewis exim[1980]: socket bind() to port 25 for address (any IPv4) failed: Address already in use: daemon abandoned
Aug  4 21:17:26 Voiplewis exim[1980]: socket bind() to port 25 for address (any IPv4) failed: Address already in use: daemon abandoned
Aug  4 21:33:50 Voiplewis exim[995]: exim 4.80.1 daemon started: pid=995, -q30m, listening for SMTP on port 25 (IPv4)
Aug  4 21:35:13 Voiplewis exim[1992]: socket bind() to port 25 for address (any IPv4) failed: Address already in use: waiting 30s before trying again (9 more tries)

[Tech Electronics]

Telcoengineer,

I hate saying this, but upgrade them to 6.0 SP1 PR4 and retest.

Thanks,

TE

[telcoengineer]

TE

Sorry I should have been more specific with the version, the system is at pr4 sp1.

[telcoengineer]

TE

This was a new install last week Friday, our SOP is to download the most current version from the FTP site Emergency Builds. 
I have been severely under impressed with mitel support lately. I wish it was inter-tel. It's sad when you are certified on the product, call for support and can't seem to find someone that has any knowledge about the system or at times has less real knowledge than you have.

Sorry for ranting, I'm done now.

[bhackbarth]

It's no secret that v6 has had its bugs.   Don't worry too much about the phone support;  when I was calling about the loopstart issues they said that no fix was in the works, but the patch for it appeared just a week later on the engineering site.  There may be a bit of a disconnect between the phone support staff and the engineers writing the patches.

I think Mitel does less software testing than Inter-Tel may have done, which means we must rely on the customer to find the bugs. But at least they are fixing them

[Tech Electronics]

Telcoengineer,

Is the UM synchronization Enhanced or Forward and Copy? If it is Enhanced try Forward and Copy.

Also, try using a relay instead going directly to the email server if you are not already.

If you can turn off the VM Networking too and test it that way. It has to be one of those or a combination of them, we just need to narrow down the suspects.

I would probably start off by turning everything off, perform a reset and see if the system calms down. Then start up each one separately and see if there are any problems. Then try both at the same time again. After that I would look at changing parameters based on my findings.

Thanks,

TE

[telcoengineer]

Hello TE,

The UM was forward only and that was on only 2 mailboxes. We did try it with VPIM disabled. I had been noticing that the CPU usage was running between 90 to 100% all the time with only 2 active calls. I had asked a tech from Mitel and she implied that that was acceptable. I have installed quite a number of these and have only experienced that when they were SIP of SSH attacks.

I ended up calling into Mitel again and got the right person this time. He escalated the issue to level 3 and they found all kinds of cool things. He found that the CPU was primarily being used up by AVDAP so he started digging around there, found several emails stuck in queue most likely corrupted he said. He cleared those out and also discovered that Enhanced Reporting in the UVM was potentially causing an issue. Once her cleared emails and disabled the Enhanced reporting the CPU usage came down to normal.

I am now receiving voicemail to email. They emphatically told me that VPIM with an offsite mail server will not work and is not supported, mostly revolving around DNS issues. I do however have it working so hopefully there will be no problems that require support.
Also, apparently this is a potential issue caused by converting an Axxess database even though we saved the Axxess DB tested it, converted to 5000 HX and tested it, there were still some underlying causes that they could not explain.

Thanks for all of the input and suggestions.

[Tech Electronics]

Telcoengineer,

Were you able to see the steps they took to clear those corrupted emails out or were they remoted in and explaining things to you?

I know they can get down into the Linux OS and perform some functions there, but I have yet to find out how they do it.

Thanks,

TE