Hello
We’re having some issues with the slave status updating in Monitor: there’s been no change since we installed RC3 yesterday. The farm seems to be functioning correctly otherwise, it’s just not updating the slave status. Everything’s marked as Offline.
Here’s the Console log:
2014-12-02 10:13:27: Main Window shown
2014-12-02 10:13:27: Time to show main window: 196.000 ms
2014-12-02 10:13:27: Error occurred while updating pulse cache: Object reference not set to an instance of an object (System.NullReferenceException)
2014-12-02 10:13:27: at Python.Runtime.PyString…ctor (System.String s) [0x00000] in :0
2014-12-02 10:13:27: at Deadline.Importers.PythonImporter`1[Deadline.Pulses.PulseInfoSettings].UpdateData (Deadline.Pulses.PulseInfoSettings[] modifiedData, System.String[] deletedIDs) [0x00000] in :0
2014-12-02 10:13:27: at Deadline.Importers.PulseListImporter.a (Deadline.Pulses.PulseInfoSettings[] A_0, System.String[] A_1) [0x00000] in :0
2014-12-02 10:13:27: at Deadline.StorageDB.PulseStorage.a (System.Object A_0) [0x00000] in :0
2014-12-02 10:13:57: WARNING: disabling profiling because of error begin profiling section twice without ending the first one (Pulse Importer).
Something to do with Pulse perhaps? We did have a license server conflict, so our Deadline license is not running on the default port. Works fine though.
Many thanks
Tom
Hey Tom,
It sounds like you’re hitting this regression with RC3:
viewtopic.php?f=205&t=12495&start=40#p56555
We should be getting a new build out tomorrow (RC4) that addresses this issue.
Sorry for the inconvenience!
Cheers,
Ryan
Just an FYI that Deadline 7 RC4 has been released!
viewtopic.php?f=204&t=12744
Cheers,
Ryan
That fixed it. Thanks Ryan!
Awesome, thanks for confirming!
Might have spoke too soon. The slave status is updating correctly now, but I’m still getting this error from the console in the Monitor:
2014-12-04 17:42:31: Error occurred while updating pulse cache: Object reference not set to an instance of an object (System.NullReferenceException)
2014-12-04 17:42:31: at Python.Runtime.PyString…ctor (System.String s) [0x00000] in :0
2014-12-04 17:42:31: at Deadline.Importers.PythonImporter`1[Deadline.Pulses.PulseInfoSettings].UpdateData (Deadline.Pulses.PulseInfoSettings[] modifiedData, System.String[] deletedIDs) [0x00000] in :0
2014-12-04 17:42:31: at Deadline.Importers.PulseListImporter.a (Deadline.Pulses.PulseInfoSettings[] A_0, System.String[] A_1) [0x00000] in :0
2014-12-04 17:42:31: at Deadline.StorageDB.PulseStorage.a (System.Object A_0) [0x00000] in :0
And WOL is working, but I now can’t remotely shut down the render nodes. (Either manually, or with Deadline’s Power Management. The connection just times out.)
Many thanks
Tom
We actually stumbled across this specific error this morning, and have already fixed it internally. Needless to say, there will be an RC5 released for v7 next week.
For the remote control timeout issue, is the Launcher running on the remote nodes?
Cheers,
Ryan
Yes it is. I checked on all of them.
Thanks
Tom
Thanks for checking. Are you able to ping the remote machines from your machine (by both host name and IP address)?
Yep, everything else is working normally: responding to ping, VNC login etc.
Could there maybe be a firewall involved here (running on the remote machines)? We’ve had no issues running remote commands here with RC4, so maybe it’s something external to Deadline…
Just to confirm, did this remote control issue just start with RC4, or has it been a longer ongoing issue?