It would appear that setting PYTHONHOME for the current session could solve this issue. I think the reason why this cropped up in beta 6 is that in previous versions, the Slave’s PYTHONHOME variable was being inherited by the rendering process.
I’m going to do some additional digging on my end, but I’ve uploaded an updated MayaBatch script here that explicitly sets the PYTHONHOME variable to the correct Maya location for the given section. Give it a try and let us know if it makes a difference.
Glad to hear the MayaBatch patch fixed the python errors. We’ll be including this fix in beta 7.
So for now, it sounds like you should be fine by enabling the option to only map if the drive is unmapped. Then, when beta 7 comes out, things should work like they did back in beta 2 where the drive will only be mapped for the first task of a job.
Also will the changes you made to mayabatch.py be included in 7 or were those specific to our issues?
I think Seth still needs to look at how all the python paths are being set with the new mayabatch.py you sent.