Nick,
For production I recommend that you use full backups.
Full backups are used by the GemTools for backups. There a number of
options for restoring from full backups so the restore procedure can be
bit complicated (just look at the GemTools restore code for an example)
but for production it is worth reading the documentation and being
familiar with the various restore procedures.
With the "suspend checkpoints" technique you make a copy of the extent
and to restore you shut down the stone, copy in the backup and restart
the stone ... no need to read the manual to figure out how to do the
restore. The disadvantage is that you don't have as many options as you
with with restore from backup...
I'll probably switch the runBackup script to use full backup in the
future...
Dale
On 01/31/2011 06:35 AM, Nick Ager wrote:
> Hi,
>
> Looking at the /opt/gemstone/product/seaside/bin/runBackup, the keys
> lines appear to be:
>
> (System suspendCheckpointsForMinutes: 5)
> ifTrue: [
> System performOnServer: 'cp $GEMSTONE_DATADIR/extent0.dbf
> $GEMSTONE_DATADIR/backups'.
> ].
> ] ensure: [ System resumeCheckpoints].
>
> However page 187 of the GS64 SysAdminGuide shows:
> SystemRepository fullBackupTo: 'backup' MBytes: 0
>
> Are the two methods equivalent? If not is one to be preferred over the
> other?
>
> Thanks
>
> Nick
>