Hi, we have a problem where a render fails because a connection to a mapped drive is lost, however the system fails all the frames assigned to a machine and does not move on to the next render machine , how do i fix this?
This might be due to the plugin failing the job instead of the task? What plugin is this?
we do a mayabach first and then a redshift render , it normally fails when having to fetch a redshift.rs file from drive, and when it cant find the file it fails, and does not resume with next machine in que, just fails the whole job
Right, this might be some Deadline behavior then, perhaps the base plugin fails on certain types of errors? Speculating here 
need to look at the code.
which code would that be?
For instance <repo>/plugins/MayaBatch/MayaBatch.py and such, you could look for clues by searching for occurrences of fail.