There is no change from 11 to 12 that would have any impact on connections.  SmarterMail doesn't really even manage the connections, we rely on the .NET Framework and Windows to do most of the work and notify us when there is a new connection and then we act on it.

 

SmarterMail itself has no set limits, we are only limited by the available hardware or performance of Windows.  When connections are being dropped, that is the OS making the decision and it often  related to issues from disk i/o, cpu, or some VM limit by the host etc.

 

If you want to provide a lot more info about your servers and configuration, we can give you some ideas.  But, there is a lot to evaluate and if your looking for quick results, it would be best to have our support team look at your server and evaluate the issue. We find crazy problems like Netowork Drivers or the need to disable CRC checks on a network card etc.  Our support often has to go FAR beyond SmarterMail and more into your Windows install.

 

Thanks,

 

Tim

Hi Tim,

 

I just posted support ticket and will work from there.

 

The most irritating thing is it just started to happen after I switched to new server and new SmarterMail. I understand in this case we have million things to consider but it's just irritating that with enterprise class hardware, brand new operating system and latest version of SmarterMail software I get issues I never had on previous, crappy system.

 

It's bit strange to point on hardware because I should see the issue with all services not just IMAP.

 

Let's stop here. I'll update this thread once I get it solved within support ticket.

 

Thanks,

 

PP

I completely understand.  We will do out best to help you get to the bottom of it.

 

I have had my own experiences as well.  I had built one of the larger Windows hosting companies back in the day.  In the early stages we had finally made the decision that we were going to go with better hardware and build out an entire segment because we were growing so quick.  It was a big decision for us at the time.  So, we had purchased a bunch of hardware and found what we believed to be the perfect hardware configuration.  Then, we purchased enough hardware to build out a 100+ servers and built out the segment all at once.  We were sending shared sites in a round robin fashion to these 100 servers, so each server was slowly going from 1 to 2 to 3 sites etc.  After about 50 sites were on each of the servers they all started to fail.   We had tested this hardware configuration and it could have handled 400+ sites on each of the servers but after 50, everything was collapsing.  After many long sleepless nights, it came down to the RAID controllers.  Well, it turned out that the manufacturer had produced a bad batch of RAID controllers.  We had to hold things together for awhile until the manufacturer sent us new ones.  I wish this was the only story I had, but the list is endless.  There is so much that can go wrong hardware or software.  What is nice is that we have been there before and we try to help our customers through these things. Often, its not SmarterMail related but then sometimes it is.  In any case, I / We can sympathize with your situation :)

Hi,

 

It's been more than 2 weeks without seeing this issue again. It stopped the same day I submitted support ticket to SmarterTools. You did the magic without touching anything :)

 

Seriously, the reason why it happened is still unknown. Right now I can just guess it was kind of DoS directed to IMAP. It's a bad guess because I would expect whole server going down in case of DoS.

 

It's not coming back now and I'm happy with this. Please consider this case closed.

 

Regards,

 

PP