Need some more specifics.
Is this a system that has run without issues for years, and suddenly got slow on finds today, or ???
What's going on with server activity levels?
Is this a dedicated FM Server?
5+ years with no problems. I have been noticing everyday this week it has got a little worse with today being almost dead stop. Yes it is dedicated but Im not sure of activity levels. How would I go about checking these and what should I look for and compare to?
Also, have there been new settings updated for anything in Server in the last week or so?
Can you stop FM Server and then restart it? If so, does it improved when restarted?
For tips on monitoring server load, follow the links in FMI's website from this KnowledgeBase asnwer:
That I am not sure of. I will check with my other tech guy and see if he permissioned it. I can try a restart tomorrow morning and let you know what happens. Thanks for the link.
Unfortunetly a restart had no effect. Still slow and worse than yesterday. Also no updates have been installed recently. Today I cannot even cancel the find without total lock up of FM.
Has anything changed in the database? Are you searching on related records?
Finds on related records will get painfully worse as the record count of those related records increases.
Nothing background has changed except a few colors and field locations. We do have an import every morning which can range anywhere from 1000 to 10000. My record count for one of my files is 1449750. (Out of 13 files) Yes they are related. Believe me it is very very painful. I will be going through the server tonight front to back side to side. I am having to let employees go home early as FM is being very frustrating.
Don't forget to check your physical network to make sure it isn't a network problem. Bad switch or router? Some activity hogging lots of bandwidth? Bad ethernet wires? Its just one additional possibillity.
Thanks much. That it something I never thought of but makes sense.
One quick test on this is to disconnect the server from the network to isolate it. Then log into the database on the server and run the find and test the speed. That way you are doing it without the network being involved. Usually running it on the local machine will be a little faster than over the LAN, but it should not be significantly different. If it is still slow, then back to looking at the file. Has there been a new join? Is there an external source (ESS or ODBC, etc.) that is no longer talking to FileMaker? Are there unstored fields being searched? Are the searched fields indexed? And you can always do a Recover and see if the recovered file behaves the same way. Is there another Server Scheduled Script running at the same time? You might also look at the Admin Console log files too.
Checking for network problems is an excellent suggestion. How's Internet speeds, etc. on your network?
You indicated this server has been running for 5 years without a problem. Has it been running on the same box for 5 years (i.e. are you on a 5 year old server)? If so, it might be time to think about getting a new server, 4-5 years running 24/7 is a long time - might want to run some diagnostics there as well.
I think someone else also suggested this, but if possible you might want to pull copies of you databases offline and run a recovery. What version of FM Server are you running?
I was having the same issue. I had this problem for over 2 years, find requests were taking ages to complete due to the record count in the file reaching 1 million. To help the system, I had to create yearly archive files and move records into the archive file for the year. I also found out that the version of FM Server I was using, the system would fall over when a file reaches 1gb. I started then to create automatic scripts to run reports etc out of hours as the system began to run very slow during the day.
The server I was using at the time was about 5 years old and using FM Server 6.
During February 2012, I purchased a new server, Dual Quad Processor, 16gb RAM & SAS drives and installed FM Server 11v3. I also did file recoveries and compressions. Amended various seaches to perform finds and constrain within found sets. So far so good.
I also did some network checks, replaced network cards, cables, hubs & switches where necessary. This helped alot. Still get network bandwidth hogging at times but this is caused by the reports users run each morning. But overall running of the system is much more better.
Fingers crossed !! - System is very stable at present even though the data file has reached 1 million records.