what title says
plus authenticating into each node using ssh host keys instead of private keys.
how bad is it? how can it be mitigated?
can we offer to alter skeleton file to not let users log into compute nodes
(configurable)?
can we offer to alter all current user profiles or pam to not let users log
into compute nodes?
this must be first be done in 1 run first (maybe in parallel) before the actual job is done, which means parsing results.
even if run in parallel, it should be able to abort after finishing $current_alteration
native auth in windows with samba4?
Original issue reported on code.google.com by geo...@marsel.is on 12 May 2013 at 10:29
Original issue reported on code.google.com by
geo...@marsel.is
on 12 May 2013 at 10:29