Re: I think i've understood because VirtualFull consolidation stall...

Mandi! Arno Lehmann via Bacula-users
  In chel di` si favelave...

> One is to set "Maximum Concurrent Read Jobs" to less than half the 
> available devices of your autochanger. That would be one in your case.

Better later then ever.


I've tested this and wait some weeks to be sure that all work as expected; i
can confirm that simply limiting 'read jobs' work as expected, leaving
sufficient write device to be able the consolidation tasks to end and not
stall each other.


Looking wider, seems me that all the 'Maximum Concurrent( Read)? Jobs' stuff
in director, storage and client need a deepening: apart prevent locks, wrong
chices can impact performance and some 'rule' (ot thumb?) can be assessed,
at least.


Or i've missed some 'bacula whitepaper'? ;-)

--