On 12/22/17 00:29, Michael Paquier wrote:
> exclusive backup API can lead to problems? Imagine the case where
> you take a exclusive backup and the instance from which a backup is
> taken crashes, *with* a backup_label file on disk. Oops. That's one
> reason behind non-exclusive backups, which is what pg_basebackup
I was noticing that terminology in the long backup-from-standby thread
I was reading, but it wasn't clear to me how the terms originated.
What's exclusive about pg_start_backup/copy/pg_stop_backup? And what's
nonexclusive about pg_basebackup (which, AFAICS, is following roughly
the same sequence under the hood)?
By the way, what does happen in that case? I'm guessing it wakes up,
sees the backup_label file, decides it's doing a PITR, and starts
replaying already-applied WAL from the start-of-backup checkpoint,
rather than from the most recent one? Oops.
-Chap