Posted on September 19, 2018 7:02 pm
Print Friendly, PDF & Email

I have implemented an external loop and it is getting executed as expected.
I have not extended agent_handler as the event_loop’ wait_for_initialized() method provide us the required functionality to wait for the initial state sync from Sysdb.

Now, the point is when I execute daemon shutdown, I see that the daemon is always in shutting down state.
Secondly, ProcMgr is seen to restart the agent endlessly without accounting for the SHUTDOWN.

Is this some sort of bug that needs to be handled in EOS SDK when external loops are used?
Supposing that the external-loop is an infinite loop with an optional sleep time, how does the agent and hence the EOS SDK get notified that the agent should be SHUTDOWN?


Posted by Alexis Dacquay
Answered on August 14, 2020 9:40 pm

Hi Shrikanth,

Maybe you can raise a Github issue at:
Or raise a ticket with the Arista support team?


Post your Answer

You must be logged in to post an answer.