I’m not sure what this is or what could cause such behavior, but I run Deadline on two virtual machines. The server is a Windows Server 2008 x86 machine for the Repository and Pulse. The test machine is a VM of Windows XP x86. The thing is, when I run Slave app manually on the XP machine, it closes soon after. Automatically, no warnings and the slave is then marked as stalled in the monitor. I can manually reboot the Slave app, no problem, but it’ll close soon again.
I don’t have automatic Slave shutdown setup anywhere. Well, not anywhere I’d remember. But this behavior puzzles me. Never seen this before.
[code]2011-03-21 19:06:08: <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
2011-03-21 19:06:43: LICENSE-FREE MODE: Slave info file was deleted. Terminating slave.
2011-03-21 19:06:43: Slave - Exception: Failed to update slaveInfo: The network path was not found.
2011-03-21 19:06:43: Slave - No slave update error notification address specified in Repository Options - cannot send notification
2011-03-21 19:06:44: OnFormClosing
2011-03-21 19:06:44: Info Thread - requesting slave info thread quit.
2011-03-21 19:06:44: Listener Thread - OnConnect: Listener Socket has been closed.
2011-03-21 19:06:44: LICENSE-FREE MODE: Slave info file was deleted. Terminating slave.
2011-03-21 19:06:44: Slave - Exception: Failed to update slaveInfo: The network path was not found.
2011-03-21 19:06:44: Info Thread - shutdown complete
2011-03-21 19:06:45: Main window closing
2011-03-21 19:06:49: Scheduler Thread - shutdown complete
2011-03-21 19:06:50: Slave - Exception: Failed to update slaveInfo: The network path was not found.
2011-03-21 19:06:50: OnFormClosing
2011-03-21 19:06:50: Main window closing
2011-03-21 19:06:50: Main window closed[/code]
Hmm… there seems to be a network problem somewhere. Not sure where or what it’s causing, so, if it cannot be related to Deadline, please ignore this thread. Thank you.
Sounds to me like the slave can’t find the repository path. Best check the path is correctly defined in the *.ini file @ C:\ProgramData\Prime Focus\Deadline (or Thinkbox on v5)
If this is OK, it may be the VM environment.
I kind of remember issues between using “Bridged” instead of “NAT” on the VM player / virtubox player network settings caused some grieve on VM’s. (best to keep it on NAT)
I indeed did revert from NAT as it wasn’t working properly on my setup here (not sure why, to be honest), so I use bridged cons atm.
I have all the paths properly set in the deadline.ini as you suggested to check. Also no problems in regards to user priviledges etc…
And another thing, I had this issue again, on the monitor app, though, and it crashed on me. So, I guess it might be something in both the network and the Deadline 5 beta? Not sure. Here’s the screenshot. After clicking “OK”, the monitor just dissappeared
Can you post the log from the session when the Monitor crashed? You can find the log folder by right-clicking on the Launcher app and selecting Explore Log Folder. The reason for the Monitor crash should be in the log.
However, I really don’t know which one to look at, so, here are all of the logs for that day (EDIT: it seems it’s the 0007.log, according to the time stamp).
By the way, I reverted back to NAT in the VMs (although still have no internet access in the machines) and it seems that both the Monitor and the Slave apps are stable. Haven’t had a crash in two days of runtime now. So, thanks Mike (I believe it was you who suggested this?) for the tip.
@rrussell - cool (I just think you should shout louder about this achievement in your release notes )
@loocas - cool. Yeah, switching to NAT fixed our issues as well with our flat file VM’s as they would be run on the top of an OS. This problem doesn’t occur on our main ESX hosts as they have physical connections.
