Does someone know any info about this problem?
I’m trying to to do a remote control stop of the deadline slave for testing and I get the error “No Connection could be made because the target machine actively refused it…” certain times when I run the command. I’ve tried running it from my computer or directly on a slave and I would say 3/5 times get that error. Ive tried with the ForceStopSlave and get the same error.
When a remote command is sent to another machine, it is sent to the Deadline Launcher running on that remote machine. So if the Launcher isn’t running, that’s usually the reason for this error message. However, that doesn’t explain why you get it randomly. When you send the command, do you send it to a single slave at a time or to multiple slaves at the same time? Also, do you get this error when sending remote commands from the Monitor? The Monitor and Command apps use the same code to send these commands, but I’m just curious to see if it is also reproduceable from the Monitor.
Finally, which version of Deadline are you using?
Cheers,
Hey Ryan,
The deadline version we are running is 5.0 and the launcher is running on the slave that I try running the StopSlave command. I’ve just tried it from the deadline monitor and it worked for one and failed for the other. I run it from my command prompt usually and its very wishy-washy…sometimes it will work, sometimes I’ll have to run the ForceStopSlave command after getting an error for the StopSlave, and sometimes I just get the error mentioned above for both StopSlave and ForceStopSlave and is just unresponsive to any commands. Its possible to run the command for numerous machines at the same time??
Thanks for your help
Just to confirm, do you mean it worked for a specific slave, and not another slave, or is it random for the same slave? I guess the same question applies to the command line. If it is random for the same slave, we can have you look at the Launcher log on the remote machine when you get a failed command. To find the launcher log, right-click on the Launcher in the system tray on the remote machine and select Explore Log Folder. Find the most recent Launcher log, which will be from the current Launcher session. If you open it, you can check the end of it to see if there was an error getting the message. If there is no sign of the Launcher ever receiving the message, then something else must have blocked it from reaching the Launcher.
If you need to have the slave shutdown, then ForceStopSlave is probably the way to go. Sometimes the slave isn’t immediately responsive to the stop command because it’s busy, and forcing it to shutdown ensure that it closes.
Yes, it is. You can specify a comma separated list of slave names in the command line.
deadlinecommand.exe -RemoteControl slave01,slave02,slave03 ...Cheers,
Hey Ryan,
I’ll check the logs in the meantime but did you mention that the Remote Control command wont work if the launcher isn’t running? All the slaves are running without the launcher started except for the ones I’m using to test with.
Thanks
The Launcher needs to be running on the machines that are receiving the remote commands. So if the Launcher is running on the remote machines that you are sending the commands to, you should be good.
Cheers,
It might have been because of the launcher… I tried a bunch of times with the launcher shutdown and with it on and ran the Stop Slave from the Monitor. The error I was getting was replicated on the Monitor every time the Launcher was closed and it was fine with it started (however the StopSlave command from the Monitor is a ForceStropSlave?). From my command line on my desktop, however, I’m getting a whole new error altogether - “The requested address is not valid in its context 255.255.255.255:5042”. I believe this is related to my network connection and I havent rebooted yet to see if that has any effect whatsoever. Ill let you know but so far using the Deadline Monitor for the Remote Control commands has proven to be more effective for me.