Hi - I’ve read the threads on the scheduler threads issue but still have a question. My client is using 6.5SP3 and has set max scheduled threads to 25. Many of the tasks are complex so ‘repeat from last invocation’ is not an option. There are 3 or 4 tasks that run every 5 minutes. Recently the databases were changed to clustered and now they seem to max out the scheduled threads every few days. My feeling is the threads are hanging due to db hanging. The threads never appear to time out since it never recovers without a reboot.

We’re trying to find out if they are rebooting the db at certain times to see if that is causing the problem.

Any suggestions would be appreciated.

It may be worth checking your DB adapter services. If they don’t have a timeout then perhaps adding one would help.

they all have 1000 msec timeouts

I guess next would be to take thread dumps when the thread pool is exhausted. Perhaps wM support would be able to identify where the threads are stuck.

A 1 second timeout seems really short.

My client doesn’t currently have access to support as they are using 6.5

Does the Service Usage page show the services that are scheduled to be running? Perhaps that can help zero in on where the services get hung.

It shows a list of services and when they last ran. I’ll have a look next time it happens and see if anything stands out. Thanks Rob.