Khangelani Gama <Khangelani.Gama@xxxxxxxxxxxxxxxxxx> wrote: > the backups we have uses pg_dump command. The automatic > backup dumps file happen daily and gets saved in a certain > directory No off-site copies or archives to fall back on? If not, my ideas are: (1) I hope you made a file copy as soon as you found the problem. If not, stop the database and make one immediately. If the old versions of the rows haven't yet been vacuumed away, it might be possible to fish some or all of the corrupted data out of the database. This is very tricky work, though, and you might want to contract with an expert. (I don't do that anymore, but there are several good companies with people who do.) (2) Are there any source documents you can fall back on to reconstruct the data? (3) Is there any way to "reverse" the particular calculations or operations which mangled the data? Once you do get upgraded, you should look into Point In Time Recovery (PITR) backups. Had you been running this, you could have replayed the transactions right up to the moment of corruption. Well, if you kept a base backup from before the corruption and all the WAL files since that base backup. -Kevin -- Sent via pgsql-admin mailing list (pgsql-admin@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-admin
