Thank you for your post.
I found your case into Technical Support, and I have forwarded the case to my manager. You will receive a response from a Support Technician.
In the case, I see the crash log with a 701 error, which is generally caused by out of memory. Did you ever have this running on a single-machine configuration?
I noticed in the web Event log file, everything was working normally until September 29 when the Web Publishing Engine stopped, and then every few days after that it would stop. I know it has been a couple of months, but do you recall if any updates to the system were made around that time?
We never tried this particular solution as a single server. We have approximately 200 users (not all simultaneous, but usage can be heavy at times).
We never saw any high memory utilization issues reported by Windows. We increased the web server RAM to 12GB at the time the issue started to become a real problem.
It didn't actually start on Sept 29... it just became much more frequent. Previously, it happened approximately once every few months, and the error message was different (we no longer get that old error message).
I checked with our Server admin back when I opened the ticket, and he said no changes were made in the few days before the problem intensified.
I don't know if this is actually related or not, but I just noticed that our FM13 split server deployment is also stopped, but our usage of that is seasonal, and it's not active right now, so I have no idea how long it's been stopped, or if it has been a recurring problem.
Thanks for the additional information.
The web event log started in mid-June, so I did not see any error until beginning with Sept 29.
Of the "approximately 200 users", has there been an increase in activity in the last few months?
Are the "approximately 200 users" accessing the server 24 hours/day? To be more specific, since nothing appears to have changed, I'm wondering if FileMaker Server somehow became damaged, as this would require an uninstall and reinstall of FileMaker Server. In other words, is there a time when nobody is accessing the server?
Since the issue is occurring with FileMaker Server 14, the FileMaker Server 13 deployment would be irrelevant since it is installed on its own server(s). Check the FileMaker Server 13 logs to see how the service was stopped.
The prior crash was not logged (at all, that I could find -- neither system/application logs, nor FileMaker logs). The prior crash did not cause the actual service to stop, it was still "running", just not working.
I believe the change in behavior was related to an FMS update, but I'm not 100% certain.
There has been a stepwise increase in usage since July, as we've been phasing in the system with various groups.
The largest increase in usage should have been at the beginning of September (technically the last few days of August).
We do have usage nearly 24 hours/day, but we can arrange a maintenance window.
I already tried to launch a whole new VM, but I was having trouble getting it to talk to the DB server, and ran out of time in my maintenance window, and had to switch back.
Is there a secret to switching Web servers? I initially tried to switch it "live", and when that didn't work, I switched the DB server back to single server setup, and then switched it back to two server setup. It still wouldn't talk to the new web server.
I restarted the old server, and it began communicating immediately.
Also, you're not asking for both servers to reinstalled, are you? If so, is there a way to export/import the config? If not, I fear the process would take quite a while.
Thank you for the additional information.
Sorry for any confusion. Only reinstall FileMaker Server 14.0.4 on the relevant server since that is where the problems occurred.
I have just been informed that one of our senior Support Technicians is now reassigned to your original case, and he has recently contacted you. He will now work with you directly.
Thank you. I'll work with the Tech and see if we get anywhere.
Have you had any resolution to this issue? We recently updated and started using a two-server deployment, and the web services in particular stop for no reason and a 701 error is all we get. Both machines are dedicated to FMS, nothing else, are just a few months old with 16 GB of RAM each. Stats never show an out of memory condition, so I'm not sure where else to turn.
Restarting the servers resolves the issue for a while, but not long enough. I did notice that depending on how the connection is closed might have some input to the crash, but I'm still looking for some time to further test that.
Did not have any of these problems with our two server FMS13 deployment.
In short, no.
The tech I'm working with has asked me to reinstall the server software (both halves), but I used our winter holiday maintenance window for something else, and haven't made another one long enough yet.
Your symptoms sound almost exactly like ours. We aren't running with quite as much memory... but they're both VMs, and we've never seen it go anywhere near full, so it didn't seem like there was any need to increase it. I can bump it if someone tells me it's going to solve this problem.
I've found myself wondering if it's not a *system* memory limit, but a FMS memory limit, and that there's no obvious way to increase the setting on Web Direct front end.
It seems like either that or some kind of allocation bug.
Thanks for the update. To clarify, we're running two Mac minis, these are not virtual machines. So, since we appear to experiencing the same issue on different hardware and setups, it seems that this is a much deeper issue beyond simple configuration.
I have sent a message to the Support Technician that was last handling your case. If you don't hear anything by Monday, please let me know.