Mandi! Josh Fisher via Bacula-users
In chel di` si favelave...
> The NextPool needs to be specified in the
> FVG-SV-ObitoFilePoolIncremental pool resource, not in the job resource.
OK, moved 'Next Job' in pool.
> Other than that, you are specifically telling the job to run on a single
> Storage resource. That will not work, unless the single Storage resource
> is an autochanger with multiple devices. You need to somehow ensure that
> Bacula can select a different device for writing the new virtual full
> job.
[...]
> Otherwise, you can use a virtual disk autochanger.
OK, i've rerun the job and effectively still stalled. I supposed that i can
explicitly use a (write) storage resource with:
run job=FVG-SV-Obito level=VirtualFull storage=SVPVE3VirtualFile
eg, expliciting the storage; still i've not understood why this simple
solution does not work.
So, for a sake of completeness, i try to summarize:
1) for a VirtualFull job, it is needed TWO devices, one for read, one for
write. This is obvious for tape device, for file device a bit less.
2) because devices are binded to Storages, the easy solution is to use TWO
storages. And because Storages are defined on Pools, TWO pools (supposing
READ and WRITE pools).
But because consolidation put the new virtualfull on WRITE pool, you need
also to move the full back to the READ pool, to make subsequent virtualfull.
[Indeed, this seems a bit mad to me]
3) the only escape line seems an autochanger (eg, don't blame me for
simplification: a storage with two, or more, devices) so we have a single
pool and storage, but still able to read and write to the same pool.
This seems to me literaly the ONLY way to implement 'Progressive Virtual
Full'...
Next week i'll give it a try. Thanks.
--
...I CARE. Significa: me ne importa, mi sta a cuore.
E` il contrario del motto fascista: me ne frego. (Don Milani)
_______________________________________________
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
