Hi -
I'm looking at ways to do clean PITR backups. Currently we're
pg_dumping our data in some cases when compressed is about 100GB.
Needless to say it's slow and IO intensive on both the host and the
backup server.
All of our databases are on NetApp storage and I have been looking
at SnapMirror (PITR RO copy ) and FlexClone (near instant RW volume
replica) for backing up our databases. The problem is because there
is no write-suspend or even a 'hot backup mode' for postgres it's
very plausible that the database has data in RAM that hasn't been
written and will corrupt the data. NetApp suggested that if we do a
SnapMirror, we do a couple in succession ( < 1s) so should one be
corrupt, we try the next one. They said oracle does something similar.
Is there a better way to quiesce the database without shutting it
down? Some of our databases are doing about 250,000 commits/min.
Best Regards,
Dan Gorman
