[sheepdog] [PATCH v1 0/4] remove farm from sheepdog

Joseph Glanville joseph at cloudscaling.com
Mon May 13 09:43:19 CEST 2013


Is the problem how the snapshot is exported out of the cluster or how
the objects are marked read-only etc?

In terms of exporting it it would be really nice if a manifest of
read-only objects was created along with an interface make reading
said objects easy.
The manifest can then dealt with by another tool that understands what
you are trying to export to.

For instance you could backup to an NFS filer in parallel by mounting
the share on every sheep node, splitting up the manifest evenly
between the nodes and exporting in parallel.
This would work even better if you had a proper distributed storage
system to backup to, like S3 or Swift.
This strategy would also be compatible with tape and VTL.

Maybe I have grossly miss-understood the problem but it is atleast how
I would envision it working.

Joseph.



More information about the sheepdog mailing list