How to debug a slave that won't book

Deadline
1

I have a machine that just sits idle when it’s running a slave. I’ve compared the slave options between that host and a working one, and can’t see anything different that’s not expected (differences in machine, etc). At this point, I really don’t know how to figure out why it’s not booking. Is there any way for me to see why a given slave is rejecting pending tasks?

2

Actually, right as I submitted that ticket the slave process crashed. Maybe this will help track the problem down?

Performing Job Cleanup Scan...
    Job Cleanup Scan - Loading completed jobs
An unhandled exception occurred: Object reference not set to an instance of an object (System.NullReferenceException)
  at System.Threading.Timer+Scheduler.SchedulerThread () [0x00000] in <filename unknown>:0 
  at System.Threading.Thread.StartInternal () [0x00000] in <filename unknown>:0 
[ERROR] FATAL UNHANDLED EXCEPTION: System.NullReferenceException: Object reference not set to an instance of an object
  at System.Threading.Timer+Scheduler.SchedulerThread () [0x00000] in <filename unknown>:0 
  at System.Threading.Thread.StartInternal () [0x00000] in <filename unknown>:0
3

Can I ask for the OS running your DB and the slave in question? Can I ask your Deadline version? Thanks

4

I’m running the beta of Deadline 7.2 on the repository, but the slave is running 7.0 on Ubuntu (all of our slaves except one are running 7.0 while I qualify 7.2). The database is on Windows Server 2008. Let me know if there’s anything else I can provide to help with the debugging.

5

Hi,
All Deadline applications, incl. Slave should be running exactly the same version of Deadline which should match the version of the DB/Repository that they are currently connected to. Running a mismatch of versions could potentially cause issues here.

6

I just tried upgrading the slave to 7.2, and it still exhibits the same behavior. All other slaves (whether on 7.0 or 7.2) are processing tasks correctly.

7

Just to double-check, can you confirm via using the “Job Candidate Filter” that this particular slave is eligible for this job:
docs.thinkboxsoftware.com/produc … ate-filter

With the Filter button enabled, if the slave is still shown, then please could you enable verbose “Application Logging” in Repo Options:
docs.thinkboxsoftware.com/produc … on-logging

the logs should then be able to explain what is happening here. Feel free to post them here.

Location of logs:
In Monitor, select the Help drop-down. Choose Explore Log Folder.
Sort folder by Date Modified. Slave logs are named in the following format:
deadlineslave-[hostname]-.log

8

Awesome, that’s a huge help. Looks like it was the the render process crashing quickly and immediately blacklisting the host again and again that was making it unbookable. Thanks for the help!