We have also the same problem since a long time. This ocuurses every two weeks.
Solition: I reboote my serevr every weekend

The same happens to us since around one year. But the MTA did not stops, mails still flows normally. It is the console that reports the 3 MTA queues stopped.  The solution for us is the same: rebooting the server. Trying to restart doesn't work.

It seems that the monitoring of the MTA queues is what fails after a period of time, for us variable between 2 to 6 weeks.

Hope Symantec fix it soon.

Regards,
Raúl Batista
Drogueria del Sud S.A.

try restarting services in this order so you don't have to reboot.
1. sms conduit
2. sms agent
3. sms exchange 5.5 notification agent
4. sms active directory notification agent

doing that seemed to fix it for me without rebooting.

Looks like this has become a "me too" thread. Well, then... me too. :robotwink:

I also have this problem.  The only changes that we have made to the product is to increase our log size from 2 Gig to 4 Gig and retention to 90 days.  I also have enough disk space so I'm not running out of space.  Could this be a database size issue?  I just changed the log size to 3 Gig so I will see if that is the issue.

Hi all,

in order to better assess what the possible cause could be, could you please reply with the product and version you're using along with the environmental details such as OS, network etc?

Thanks very much

-m

Thanks Dave, yes it's the latest build.

I'm afraid I'll have to fire some questions at you, in order to get a clear idea of what can be done.
Generally, there may be several causes for the MTAs to stop, therefore it's fundamental we look at some basic things first.

Could you please let me know if there's any pattern you've noticed with this issue? i.e. does it stop around the same time, after a certain amount of time/days etc.

Is this a Windows deployment?

What does the maillog show right before the problem? if possible please include a snippet or send me the log file

Another interesting element to look at would be the network connections status of the machine when/right before the problem shows up... We've seen sometimes increasing patterns of open/time_wait connections until the MTA stopped.

Most of these cases were related to environmental networking issues, such as DNS latencies (reverse lookup etc) or firewall settings limiting or filtering the smtp conversations.

Thanks

-m

thanks Dave,

I think the best way would be just to wait until the next problem and get a maillog snippet (maybe the filter-hub as well) and have a look at it.

Something else you may want to consider/look at is the Email Firewall Policies, where you can enable features such as the Spam and Virus Attack. 

If you run into a condition where your environment is under either a spam or a virus attack, you may experience an 'overload' of requests these components make towards the filter-hub, thus exhausting the available communications channels. If that's the case, you should also notice several errors in the filterhub log about loosing connection to the firewall channel (port 41013).

Hope this will help,

Regards

-m

Hi Matia,

It happened again over the weekend and shows mta status of stopped, but pleanty of connections on port 25.  Can you tell me exactly where access to maillog is?  I find the logging confusing on v5.

Thanks,

Dave

We've been having this issue as well which happens around once a month or so.

I talked to a engineer at Symantec and he told me there was no fix for this via a patch and the only work around is to create a cron job/scheduled job to restart the SMS Agent.

That does work but it doesn't solve the issue at hand.  Mail Security says the MTA service has stopped but we're still able to send and receive e-mail to and from the outside world so I'm to believe that there is something wrong/configured wrong with the management software.  In another environment of ours, the MTA service has been running for about 6 months with no issues.  Now I'm going to call Symantec again and see if another engineer has any insight about this.

Here's the KB for this: http://service1.symantec.com/support/ent-gate.nsf/...

Thanks jcurry,

That does fix the issue of it incorrectly reporting that it has stoped.  I wish Symantec would fix the issue that causes it to start.

Dave

An update:

I talked to a very informative engineer and he said that Mail Security wasn't designed for a VM environment.  If you have Mail Security running on a VM, that's the cause of having the MTA being shown as "down" in MS.  Their Virtual Appliances software is designed for a VM environment which you won't see these type of issues.

For now, the only work around is to create a cron job/scheduled job to restart the SMS Agent.

Here's a link to their Virtual Appliance software: http://www.symantec.com/business/products/trialwar...

Thanks jcurry,

This is not a VM install though.

Dave

Ahh.

Best of luck, though!

Hi Dave,

you can find the maillog under the logs folder within the product's installation path.

Please make sure it's set to INFO or DEBUG logging level. Should the issue occur again, could you please reply with a log snippet of the very moment when the issue occurred?

thanks

I have noticed that sometimes I see this when the MTA is still actually running (mail is still passing through).

Our console server is not a scanner and I believe it loses it's sync with what's happening.  The checkbox to 'start mta' on the control center merely sends out a notification that it's started without really doing anything.

I built a mini application that drops a test email into the pickup folder to test if the MTA is still operating but other than that I still end up rebooting either the console or the scanner to get it to report correctly and allow access to the queue viewer. 

Hi Matia,

It happened last night, reporting that all MTA's have stopped, but it is not true.  I have set logging to debug and will let you know if anything showes up.  I am assuming to look in the audit_mat_log.txt file, correct?  So far all I see in there are sender/orcpts lines.

Thanks,

Dave

Hi Matia,

I just got another email alert msg and checked to log, I dont see anything.

Dave

you are right, it is still running just fine, it just reports otherwise. i finally just created a script that restarts the sms agent at midnight every night using scheduled tasks.

batch file code:

net stop "BMIAGENTSVC"
net start "BMIAGENTSVC"

i haven't gotten emailed about any issues with mta in several weeks now!

Thanks for the tip compu-tech, it is a viable workaround.  On the scanner control center with the problem it occours in the weeks to months time frame.  I also have a scanner/control center that runs in a vm and controls another scanner and to the best of my memory I don't recall it ever happening on them.

Dave