Can't connect to worker logs after machines are resumed from sleep

Deadline
1

Title pretty much says it all. None of our workers can connect to the log viewer, remotely or locally after the machines are resumed from sleep.

Connecting to Ruby...

RemoteLog: connecting to machine '192.168.0.100' which resolved to '192.168.0.100' port 5055

Making a connection to '192.168.0.100' port 5055

That’s all it ever does. Restarting the worker immediately solves the problem.

This happens on win10 and 11. IP addresses are static, ports are forwarded and specified in the settings.

2

Worker’s remote log port is random, did you make it static?
There are some tests you can perform to check if it is a networking problem:

  1. Try to connect with the Worker log, get the IP and Port, then try the following:
  • Telnet the IP with the same port by running Telnet <Worker’s IP here> <Port_here>
  • If you are able to connect with Telnet send some data like a key from keyboard
  • Then try to connect with the Worker log again
  1. Try this from any client machine:
cd /Applications/Thinkbox/Deadline10/Resources # For Mac
cd /opt/Thinkbox/Deadline10/bin                # For Linux
cd "c:\Program Files\Thinkbox\Deadline10\bin"  # For Windows
deadlinecommand -InternalConnectToSlaveLog "<workername>" -1

Share your findings.

3

Hi,

I usually see that type of output when the firewall is blocking the port. However you stated that it works again if you restart the deadlineworker.

Maybe the deadlineworker process (state LISTEN) has somehow gotten borked after the computer went to sleep?

If you use TCPView (easier than reading netstat output on windows), you should be able to view the deadlineworker processes and see if the state is LISTEN on that port after waking from sleep.

And thank you again for the Redshift/Houdini task progress!