Yes, it's the classic problem. I understand the cause, but I have a question.
a little background for those who don't know about this one... ============= Last night I got a bunch of these... elmer /hr lev 3 FAILED [dumps way too big, 9065 KB, must skip incremental dumps] As a result, lots of DLEs were just skipped. This can happen if, for instance, one DLE had lots of changes and its incremental dump is much larger than normal. Or you have one DLE that is extremely large and when it does its level 0 dump, it squeezes out all the other DLEs. So the planner decides it can't fit all the dumps for the configured tape size. The question... ============= It's been a while since I've looked at all the shiny new knobs that amanda has grown (generally we just leave amanda as is), but have we grown anything that allows amanda to at least _try_ to dump such 'skipped' incrementals to holding disk? [note... our server is still amanda-2.4.2p2, so maybe this has been fixed in a different way in 2.5 or later and I just haven't noticed]