Sometimes when I restart my server where Folwable Engine is running. Jobs get stuck.
Meaning though the lock_exp_time_ has passed , the lock never clears.
I have checked asyncExecutorResetExpiredJobsInterval this is set to 1 minute(Default).
Now I want a way to best way to remove locks for such jobs.
Currently I am clearing them manually
update act_ru_job set lock_owner_ = null , lock_exp_time_ = null where lock_exp_time_ < now()
The reset expired thread should pick this up. You’re saying it doesn’t remove the lock after a reboot, even after a minute?
Yes, that’s similar to what the reset expired logic should do.
We are also seeing similar behavior where jobs are getting locked and works fine after restart - updating from database doesn’t seem a solution. This is even caused with normal job execution (few jobs) - looking at the Flowable documentation they recommend using dedicated Message queues which might be a viable option if there is too many job execution to be processed which is not the case with our app.
I am trying to fine tuning my job executor pool settings
Other fine tuning could be done using some good article here
https://blog.camunda.com/post/2019/10/job-executor-what-is-going-on-in-my-process-engine/
Still trying to explore above options but are there any known issue or better suggestions.
Thanks!
There was a bug with resetting jobs, that is fixed on current master. Which version are you using?
Missed your reply - we are on flowable 6.5 version.