Many apologies, I got things completely the wrong way round I think! My fault entirely.
The snapshot volume holds the *original* data from the frozen volume, using copy-on-write, so neither taking nor
releasingthe
snapshot involve any copying of data.
If you run out of room on the snapshot volume because you've written too much data to the frozen volume then you are
buggered,pure
and simple.
> -----Original Message-----
> From: Tom Lane [mailto:tgl@sss.pgh.pa.us]
> Sent: 17 March 2003 14:32
> To: Matt Clark
> Cc: Ragnar KjÃzrstad; Peter Eisentraut; Murthy Kambhampaty;
> pgsql-admin@postgresql.org
> Subject: Re: [ADMIN] LVM snapshots
>
>
> "Matt Clark" <matt@ymogen.net> writes:
> > So when you take a snapshot there is no data to be copied, and when
> > you release a snapshot some data needs to be written back.
>
> Does the write-back preserve the original write ordering? In general,
> what happens if there's a system crash in this state? What of errors
> encountered during write-back, eg out of disk space (or is disk space
> accounting still done in terms of the original volume)?
>
> regards, tom lane
>
