Is this a new behavior? IE were they working before and now they hang? Or has the problem plagued you for the FM12's servers entire existence?
If these scripts perform a "multitude of tasks" on live data as people are connected, there's always the chance that there's record locking issues, or other looping issues that are getting hung up. backup schedules are entirely different, and don't affect your scheduled scripts.
Sometimes, rolling your own robust inside-filemaker logging as part of your scripts is the only way to track down hangups on FMS scripts. Some solutions, like audit logging and commit triggers, might be able to assist in some way.
I've never experienced that issue without it being my fault in a script somewhere.
Unfortunately, the usual thing to do for absolute inexplicable behavior of FMS is to wipe OS/reinstall or uninstall/reinstall. If that is the case, maybe you can manage until FM13 and it won't be an issue.
It has happened randomly since these scripts were implemented. Sometimes i can go months without an issue, sometimes it'll happen twice in same month. It's not often enough to cause an issue (maybe 5 times in a year) but i was just curious if anyone else had experienced it.
2 of the scripts loop through records that are flagged to be emailed.
The script sequence does an hourly export of records, followed by a system level script to upload to a SQL database and FTP server.
The send email is suspect to me. Are you using a plug-in or the native send mail?
One thing to consider would be to have a context-free subscript that writes to a log table. IE:
perform script "log subscript" script parameter "whatever you want here" - placed after whatever script step you are testing
subscript would be:
go to layout – log
set field – notes – get(scriptparameter)
go to layout – original layout
You could run that subscript with negligible overhead and put in a script parameter of "sent email successfully" after the send mail step in your loop.
Then you could modify this to serve other areas of a test as well. It sounds like your scripts run often enough that this type of testing would be easy for you.
So you have the schedules set to "force abort" after the pre-configured duration?
Check the FMS logs for any errors. Validation errors in particular can make server-side schedules "hang".
I am using the native send mail.
I have some minor logging in place now to stamp a field as "sent" if there are no errors with sending the message but i could expand on that a little further to see if there are other issues.
Thank you for the guidance.
The scehdules are not set to force abort.
The only errors in the FMS log that pertain to any of the scripts are 401 errors for records not found which shouldnt cause them to hang, but nothing else.