Hey guys, I keep seeing this error report when I ctrl i on a slave to see the error reports. It doesn’t matter which entry I click on; they all say the same thing. Is there anything I can do to fix this so that I can view the errors for the job? Thanks.
This is on Deadline 6.
-Christian
That would imply that your slave machines are having trouble writing the report files to the Repository. Check your permissions on the repository share to make sure that it has open read and write access.
Cheers,
ok, I checked that and everything seems to be as it should be. I even went onto several slaves and went to that location on the network, copied a file to it and everything and it seemed to work fine.
The reports/jobs folder on the repository is completely empty BTW, which would seem to indicated that this is a problem with every slave machine on the network.
Could it be anything else?
That’s strange. The Deadline 6 repository installer should have created a bunch of subfolders here. This definitely explains your problem though.
Can you try running the Deadline 6 repository installer again to see if it creates the folders properly? You can just point to your existing Deadline 6 repository as the installation folder. It’s essentially like doing a repair.
Cheers,
Thanks, I have some custom scripts in my jobs and submissions folders, will these be overwritten?
-Christian
If they’re in the ‘custom’ folder in the repository, then you’ll be fine. Also, if they are new files (in other words, you didn’t modify existing scripts), you should also be fine. In any case, we always backup those files during the repository installation, so you can always find them afterwards in the ‘backup’ folder in the repository.
Cheers,
Yup, I re-installed the repository and this fixed it with all the proper folders in the right place and the logs working. Thanks alot.
-Christian
thanks, i have the same problem. i’ll give it a try
so i re installed the repo and i gt all of the proper sub folders.
then today, they disappeared again 
Hello omere,
I wonder if all of the slaves that are trying to connect to this repository are upgraded to the same version. If a slave was trying to connect, and did not have the correct updates, it could see that some of those files are in the wrong place, and thus delete or move them. Could you check on that? Thanks.
Dwight
Yeah, now that you mention it… a few days later the folders went bye bye here as well. I checked to see if there were any machines with the older version and didn’t see any but I am going to double check.
-Christian
so in the repo folder i found “slave” folder. in that folder i found folders with names of workstations that are running deadline 5. i guess it was mistake to give the deadline 6 repo folder the same name as the deadline 5 folder even after uninstalling it.
i’ve located the deadline 5 workstations and uninstalled them. then i’ve installed a deadline 6 repo on another station and copied the missing folders from there back to the original repo.
for now everything works.
I found that slave folder as well but it just has a ton of folders in it with random numbers. Is there another way I can see if there is an older version slave trying to get onto the repo?
-Christian
That’s weird. What were the date/times on those folders? Do they imply any recent activity?
Also, if you blow the Slaves folder away, does it appear later? Deadline 6 doesn’t have a Slaves folder in the repository, so if you blow it away, it won’t affect Deadline 6.
Also, maybe check if there is a Pulses folder, as v5 of Pulse would do that sort of cleanup as well.
ok, so there is a Pulse folder and a slaves folder in my deadline6 repo. Do I blow these away? Tks.
-Christian
Maybe just blow away their contents for now (but keep the actual Slaves and Pulses folders). If something gets created in them afterwards, hopefully that points you to whatever is still trying to connect to it.
Ok, so I found the offending machine, handled it, and copied the logs back into the reports folder and that seemed to have fixed it. Tks!
-Christian