Thank you for posting:
We will need more information to narrow down the possible cause. You mentioned extended periods of reading and writing to the server - are these periods of user-driven data entry or are there scripts running?
Just to rule out any externalities, is this machine dedicated to FileMaker Server? Do you have any plugins?
When the issue occurs, does the Activity Monitor show any processes with heavy CPU, RAM or network usage? Are any FileMaker processes crashing? If you can still access the Admin Console, try forcefully disconnecting some clients.
The logs may also provide some more insight into the issue. In particular the Event.log may mention any timeouts, failed connections, or other errors. The Stats.log would give specific performance parameters if you have it enabled.
If you have any additional information about the history of this issue, please let us know.
Thanks for the quick answer,
The data is entered into the database via a filemaker script. There are no server side plugins, there are some client side plugins. This machine is dedicated to Filemaker server.
I'm going in this evening to try recreate this problem. So I can check the activity monitor, I haven't checked that yet. Once I get more information I will post the results.
I started off with an empty database this morning. Customers/Employees starting using the solution at 10AM. At around 3PM the system crashed. At this time approximately 8000 records had been created, also during that time those records could have been edited.
At 10 AM this morning not all the terminals were in use, but as the day progressed more terminals were being used. At the time of the crash, 3PM, all the terminals were in use.
What do you see in the logs?
In the events.log I just have some authentication warnings.
I Spent 4 + hours last night filling the database and trying to cause a system crash but was unable to successfully do it myself. I have turned on all the logging, hopefully next crash I will see some useful information in the logs.
As of right now there has been no errors in any of the log files.
I believe that Filemaker server is not crashing on me nor is any of my clients.
The network seems to slow to a crawl, which made me assume it was a crash of some sort. I was able to access the admin console and forcefully disconnected some clients.
Any suggestion on what would bring my network to its knees?
Did disconnecting some clients change the behavior?
Look at the stats in the Admin console to see what's there.
On the client machines where you see the slow down, can you get any kind of CPU usage stats? On windows, you'd use task manager for this, but I don't know the Mac equivalent.
If you have any kind of software you an use to monitor your network traffic, you might use it to rule out network issues.
Do you have any hourly backups with verification enabled? (That's a known issue for slowing things down.) If so, try disabling the verification option to see if it affects responsiveness.
My boss was complaining that his filemaker screens were being very unresponsive. Typed in text was not echoed promptly to the screen and lots of hour glasses and progress bars kepting popping up and then disappearing while other users didn't seem to have this problem. Then I used task manager to check his machine and found that an AOL application he insists on using was hogging up to 90% of the CPU cycles!
I have disbaled file sharing on my server and found out that there was a very important firmware upgrade for our router. I have been two days without a network slowdown.
I didnt see anything on the client machines hogging the CPU. Thanks for the help! I have been learning alot through this process.
Hi Jeff, I am having similar issues with delays in FMS 11, was all fine until we upgraded from 8.5 to 11 - and in fact was Ok until we rolled out the clients upgrades.
You say you upgraded the router? Is your network controlled by a router rather than a dedicated switch (this would cause bottlenecks) Did you do anything else that I might be able to try as I am pulling my hair out!
I had a firmware update available for my router. This router does control my network. I also cleaned up my ERD and got rid of alot of unstored calculations. If you have any unstored calcs make sure they are not used in any of your relationships as this will surely lead to some bottlenecks as well.
Ive been through the system line by line, cannot see why the upgrade has made things so unreliable. Its not that the system is slow - it stops completely. Then off it goes again, very odd.
We have also upgrade to a new more powerful server running win 2008 64 bit - we now think the operating system and FMS 11 are having issues together. There is a massive memory leak that is cured by restarting FMS, curious to know if anyone else out there has FMS11 running on a win2008 64 bit box?
We are using Windows Server 2008, 64bit with FileMaker Server 10 (not advanced). We've only had two major issues. One is an occasional hardware shut down and restart that we've traced to a firmware issue that hopefully has been fixed with a firmware update from HP. The other is that a scheduled backup hangs and fails to complete at fairly rare, random intervals. This "hang" doesn't interfere with clients who are using the system, but we have to carefully shut down and restart the server before the console and specifically server scheduled backups and other scripts will once again execute successfully.