> Now, only the messages:
> "...is recorded as being in the process of being deleted, ignoring it
> during drain" are left. Can this be ignored?
>
I hope so, as I've been ignoring it for years now!
Once I've got a pool node to a point where I think it has drained I like
to delete anything that the DPM might think is on there but isn't, using
something like (on the headnode):
#for i in `dpm-disk-to-dpns -s pool.your.dpm:/fs| awk {'print $1'}`; do
dpm-delreplica $i; done
This is to prevent future zombie entries in the dpm database. It'd be a
good idea to sanity check what the dpm-disk-to-dpns gives you first (and
maybe record it just in case).
Of course what I do might be a far cry from what you should be doing...
Cheers!
Matt
|