I have a driver that i'm working on which after a few hours of running the console reports that there are more than 1000 groups in the stack to be executed in the connection 2.
I've tried to use the pending calls to see which group(s) if any are taking long or are getting executed too often, but so far I have not been able to pinpoint the issue.
Except for this message after putting the element logs to level 2. Not sure if this is connected.
The driver is multi-threaded and also has multiple connections so im not really sure how the stack looks in this case. Does this mean that there is one stack per connection, since the message points to connection 2?
Also the driver is already using execute one and execute one on top on most actions that poll a group.
What is the best method to find which group(s) is causing the stack to grow that big.
Hi Geovanny,
The pendingcalls you're seeing related to the multi-threaded timer are normal, each row in table 200 will trigger QAction 200. The time of these threads is relatively low so initially I don't think the multi-threaded timer is the issue.
However the only thing worrying in your screenshot of the pendingcalls is the Timer with id 2 which is already running for a long time so something must be blocking there (most likey this timer is also polling connection 2).
Ofcourse this is just an educated guess, if you could provide the protocol name and version I could take a deeper look.
The issue was with a group which was getting added by a timer or via a checktrigger before it fully completed. To find which one it was, I put logs in the logic which the group executed and saw one that was taking about 2-5 minutes which was in a 30- second timer. I added a condition with a flag to the group so that it only executes if the previous one was done.
Hi Jelle
The name of the driver is Verizon Newtec Dialog Platform Collector the latest version is 1.2.0.6.
Regarding the timer in the pending calls I thought it was ok and sometimes normal to see timers for a long time, the only thing that should be worrying is if you see groups which are your leading cause to RTEs.
Also timer 2 does indeed deals with connection 1 which is the second connection in the driver, ill take a second look now that I have something to check.
Thanks,