Logging this under 'general interest' as i'm not sure if any other c

jdegraaf over 11 years ago

Logging this under 'general interest' as i'm not sure if any other category fits the bill.

I've got a couple of processes that have existing tasks. However, when trying to open these tasks from the apps portal, the 'waiting' text appears and ultimately times out.
When trying to monitor the process, the process modeler opens but an error appears saying 'Unable to acquire a Read connection. Safe to retry.'. I cannot cancel, nor delete the processes.
Other processes from the same process model are behaving as normal.

Do you have any ideas how to fix this? It's currently a production issue....

Discussion posts and replies are publicly visible

0 jdegraaf over 11 years ago

Running Appian 7.2 by the way
Cancel
Vote Up 0 Vote Down

Sign in to reply

Verify Answer

Cancel
0 Eduardo Fuentes
Appian Employee
over 11 years ago

What is the "Caused by" that you see in the stack trace of this error in the server.log?
Cancel
Vote Up 0 Vote Down

Sign in to reply

Verify Answer

Cancel
0 jdegraaf over 11 years ago

This is what the log is indicating:
Caused by: com.appiancorp.kougar.driver.exceptions.SafeRetryException: Cannot deliver messages to engines: will be retried
Cancel
Vote Up 0 Vote Down

Sign in to reply

Verify Answer

Cancel
0 Eduardo Fuentes
Appian Employee
over 11 years ago

This error indicates a process to process message couldn't be delivered because the execution engines were too busy but the action will be retried.

I don't think this error is related, can you verify this message is the only one that appears when accessing these process instances?
Cancel
Vote Up 0 Vote Down

Sign in to reply

Verify Answer

Cancel
0 Eduardo Fuentes
Appian Employee
over 11 years ago

If you find out this is indeed the error that is logged when you reproduce the issue, can you try pausing the process and see if that allows you to monitor it?

If you can't pause it from the user interface use the Process Management Services plug-in available for download here in Appian Forum under the "Shared Components" record (forum.appian.com/.../all)- Smart Services and it's Pause Process smart service.

This will let us know if the issue is the process is falling into an infinite loop locking up the engines.
Cancel
Vote Up 0 Vote Down

Sign in to reply

Verify Answer

Cancel
0 jdegraaf over 11 years ago

Yesterday (after the overnight restart) these messages were still appearing in the log, but this morning (after another overnight restart) they seem to have disappeared. I can also monitor the processes as normal again.
Not entirely sure what has happened, especially for it not to be resolved after 1 restart...
Cancel
Vote Up 0 Vote Down

Sign in to reply

Verify Answer

Cancel
0 Eduardo Fuentes
Appian Employee
over 11 years ago

This is more an indication of high load on the execution engines, the error will not be present if there's no load which can explain why you are no longer seeing them.

You can verify if this is load-related by looking at the CPU usage and the perflogs folder.
Cancel
Vote Up 0 Vote Down

Sign in to reply

Verify Answer

Cancel
0 jdegraaf over 11 years ago

thanks Adriano.
I'd be very surprised if this was high-load related. The logs show that even yesterday Jan 1st this issue was happening, even though no one was working. It could be an automated trigger that happened the first of every month, but then that wouldn't explain why it was happening Dec 31st either. We definitely checked the load on dec 31st and it didn't look like any serious load was being handled.
Cancel
Vote Up 0 Vote Down

Sign in to reply

Verify Answer

Cancel