i’m not sure if this is actually a specific Deadline7 issue but couldn’t find anything about it in the online documentation for v6.
There are a few specific Limits in the 3ds max submitter that seem to be internal/built-in. We activated the Krakatoa Limit for two jobs last week but somehow we got a lot of errors for almost all the tasks of those jobs that the maximum user count for the licenses was reached. Now i’m wondering if those limits have to be configured somewhere by us - which i guess we must not otherwise they wouldn’t be any better than defining a limit ourselves, right? The interesting thing is that as soon as we added our manually-configured Krakatoa Limit the jobs didn’t fail anymore. How does Deadline actually know all those limit for Krakatoa, VRay and the others? Does it query the license servers?
You would still have to create the limits it’s referring to in Deadline (ie: Krakatoa, V-Ray, Maxwell, etc). These are just helper options in the submitter to allow you to not worry about setting them.
i see. May i suggest to put a description of how to use this into the manual? I couldn’t find any info there on how exactly this works. Also, our artists are confused by the term ‘autotatically’ at the beginning as it sounds like it will figure out everything by itself. E.g. if you have a scene that does NOT have any Krakatoa in it we’d expect the limit not to be set although it is checked as it will dicover this situation automatically. Is this correct?
Yes, those Limits check the class of the currently assigned renderer and if it is one of the supported ones AND the respective checkbox was checked to request it AND the Limit was not already checked manually, it will be added. It assumes (incorrectly) in your case, that the Limit “Krakatoa” exists in the Repository.
I think we could improve SMTD by checking whether the Limit “Krakatoa” actually exists on the Repository and only set it automatically if it does… The reason we have not done this is that asking the Repository for the list of Limits can take a few seconds and we did not want to waste that time each time SMTD is launched (or worse, each time a job is submitted). We do cache the list of Limits between sessions, so we could check against the last known list of Limits since that does not change too often.