Multi Instance (multi lane) flows not Completing
I have the followig issue:
Flows with parallel execution (multi-lanes) does not progress. (each individual lane complete, but the flow never progresses from there).
Tried modifying the worker threads/heap size, etc... But is working on another environments with same configuration. I saw a same issue rised here years ago but with no final known resolution.
OO is 10.80 upgraded from 10.60
00 is 10.80 running on windows 2008 R2
DB is SQL 2012
Found similar issue in the forum
But no resolution was shown.
Take a look at /oo/monitoring
Jobs section toward bottom - (split join trigger) delays in processing/ too long to process/ triggering too frequently?
Can you advise number of records per OO_EXECUTION* tables
Thank you for your reply.
- Do you have other MI steps in the env with the same issue?
YES. All flows with MI on this environment fails the same, even a test with onky that
- /oo/monitoring (See attached) at the moment the split were completed (each lane)
- OO_EXECUTION* table ?
See tables on attached pic. Which table do you mean?
Thanks again for your reply
I mean the number of records per table -> can you screenshot (In SSMS Right click DB -> Reports -> Disk Usage by Table) for that section of tables.
Also, I don't see the attachment for /oo/monitoring jobs section
I haven't heard back
You could always try to clear the execution data stores - disable scheduled jobs, stop central, execute below, start central (you don't necessarily have to stop central) -- this will remove all running jobs -- you should allow previously running jobs to finish unless you don't care about their execution...
TRUNCATE TABLE OO_FINISHED_BRANCHES
TRUNCATE TABLE OO_EXECUTION_INTERRUPTS
DELETE FROM OO_SUSPENDED_EXECUTIONS
TRUNCATE TABLE OO_EXECUTION_QUEUES
TRUNCATE TABLE OO_EXECUTED_FLOW_GRAPH
TRUNCATE TABLE OO_EXECUTION_STATE
TRUNCATE TABLE OO_EXECUTION_STATES
TRUNCATE TABLE OO_EXECUTION_STATES_1
TRUNCATE TABLE OO_PAUSE_DATA
DELETE FROM OO_RUNTIME_VALUE_STORE
UPDATE OO_EXECUTION_SUMMARY SET STATUS = 'CANCELED' , END_TIME=START_TIME WHERE STATUS IN ('RUNNING','PAUSED','PENDING_PAUSE','PENDING_CANCEL')
Thank you for your reply, I didn't find your response until today.
Below the tables. I was also able to run the queries you suggested against the SQL. I stopped Central, ran the queries and startes again after deleting also the Tomcat temp files .../Central/Tomat/Temp .... And the same result, all multi-lane flows completes the multi-step operation and just sit there. See also below the screen capture for the "jobs" under /monitoring.
The one I modified is still in the same date and time I did the modification in.
Do you have any other suggestion?
Yes, some of your jobs are running way too frequently
Did you compare the values with the other environment these MI steps are successfully executing in?
are you seeing any errors in the central logs?