Re: Drawbacks of implementing undelete entirely in user

Steffen Grunewald (steffen@gfz-potsdam.de)
Tue, 25 Jun 1996 15:14:03 +0200 (MET DST)

"Bob Allison writes"
|>
|> I have a couple of file servers here made by a company called Network Appliance.
|> They have a feature called "snapshot" which seems to do everything about
|> undelete and more.
|>
[...]
|>
|> I'm not really sure of the kernel/user separation of this idea, but I think the
|> only kernel support that would be needed would be to not deallocate blocks until
|> all files in all snapshots are no longer referencing the block. Creating and
|> deleting snapshots could probably be done in user space from a cron job.
|>
So it's doing sth. like hardlinking all files, right ? Then kernel support
is already available. As long as the link count stays above zero the file
can be accessed.

|> (BTW, making a snapshot of my 32GB file system (all drives form a single RAID
|> array) takes about 15-30 seconds.)
|>
Sounds quite fast to me. But then, all the information should be in the cache
of the system already. Maybe it's a special feature of your RAID array ?
--
Steffen Grunewald |  email steffen@gfz-potsdam.de |  fax (+49)-331-8877 520
GeoForschungsZentrum Potsdam, Telegrafenberg A17, D-14473 Potsdam,  Germany
/* Disclaimer : I don't speak for my employers, they don't speak for me  */