Issues seem to persist when issuing slave or machine commands from the monitor. Ex: Shutting down slaves after task completion, or shutting down machines. I seem to get about 30% success in slaves actually responding. The common error is: A connection attempt failed because the connected party did not properly respond after a period of time, or established connection failed because connected host has failed to respond.
Perhaps I have a configuration problem?
I am running beta 7
Win Server 2008 R2 (server) Win 7 x64 (slaves)
Is it a random 30%, or is it always the same machines that succeed/fail?
By default, Deadline relies on DNS to resolve the machine’s host name that shows up in the slave panel so that it can connect to the remote machine. If there are DNS issues, that could explain the problem. A quick way to check would be to ping a machine you’re unable to remote control via its host name to see if you can connect to it.
If there are DNS issues, you can configure Deadline to use the IP address shown in the slave panel instead. This can be done in the Slave Settings in the Repository Options by enabling the “Use Slave’s IP address for Remote Control” option.
Thanks for the reply. It has been random for several months.
I launched into some network troubleshooting. The first thing I tried was to configure deadline to use the IP address instead of DNS names. That still gave the same problem. I then started looking into what was happening on the network. Apparently we had IPv6 installed on all the slaves but the network was not configured properly for that. The other issue was that on some machines there were some old configurations in the advanced section of the firewall for previous deadline installs. After cleaning up the entries I was able to get 100% response from the slaves in either IP address or DNS modes.