We migrated over many thousands of accounts. Surely up there on the "Largest servers that SM powers" list. And we had zero issues with conversion. We tested on a small virtual server, copied over all the data (minus the GRP files) and ran a test conversion, worked without issues. We have accounts from a few MB to into the very large 50GB+ mailboxes. Took about 20 minutes or so, but it finished and we were off. If you have the resources, copy over just the directory structure and all the config files etc. for the users / domains (minus the GRP files) and run a test conversion on a test server, just to see how it goes. I did it 3 times on fresh installs before we actually did it to our production system.

It's been rocky to say the least, but Matt has worked his butt of tweaking SM to the point of what I can only consider, perfection, all of the major issues we've had are strictly due to the size of our server and how much traffic and data we handle.

I'd say you are safe to upgrade just based on how well SM is working in our massive environment.

I upgraded a small server (<250 Users) on Win 2012r2 from v15.x to 17.x two weeks ago and it worked perfectly.

I'm doing another this weekend.

We too are considering the update from 15.x to 17 (Build 7016) this weekend. Many users are EAS integrated wtih 25GB+ mailboxes.

Should we be worried?

I had some bad luck again with SmarterMail. So far found one whole domain's users does not exist. Now I get to spend time trying figure out how to get them all back how can an entire domain of users not transfer.

Update, I look all the users exist on the server, I can see them, but they do not show up in the console.

So three domains out of 82 did not transfer properly. Looks like a long night of email recovery.

Did you open a ticket? The data is there and can be reconnected without too much hassle.  I’ve heard of it before but it’s not common. 

i was going to submit a ticket, but the response is the next business day and of course the domains this happened to are my biggest customers. I was not going to pay for a support ticket since it was the installations issue, not something I did. The only thing I did was update my software.

After a lot of trial an error I am going to leave this out there in case someone else runs into. I use the default location. If you do not have a backup of your files, make a copy of this file first.

Open the file, I use Notepad++, C:\Program Files (x86)\SmarterTools\SmarterMail\Service\Settings\domains.json

My data is on the D drive

Check for your domain, the format looks similar to this, I have masked my domain:

    "example.com": {
      "data_path": "d:\\SmarterMail\\Domains\\example.com"
    },

I went to the end make sure you have a comma (,) on all except the last one. I missed the one before where I copied and pasted.

Restart SmarterMail service

Restart the Website, or the web server

So far I can send and receive from one of the users, they may have lost a few days of emails, but luckily it was over the weekend.

We upgraded last week from v15 to build 7016. So far no major issues. 

The biggest complaint we got was the spotty search functionality the day after the upgrade. This is just a side affect of the re-indexing process. By default SM is set to use 1 indexing thread and with our 4600 plus accounts the re-indexing looked like it might take days. We increased the indexing threads to 12 and closely watched system CPU. As a result we got the re-indexing done in 3-4 hours.

How do you modify the indexing thread settings?

Max Threads is in Troubleshooting -> Options under Indexing.

After changing the max threads do you need to restart the SM service?

No restart needed.