Re: 9.1 doesn't start when died mid-backup - Mailing list pgsql-bugs

From Fujii Masao
Subject Re: 9.1 doesn't start when died mid-backup
Date
Msg-id BANLkTikVrTUL7aB=jWZRd6VFqsDD=WTJ_Q@mail.gmail.com
Whole thread Raw
In response to Re: 9.1 doesn't start when died mid-backup  (Heikki Linnakangas <heikki.linnakangas@enterprisedb.com>)
List pgsql-bugs
On Wed, Apr 13, 2011 at 8:18 PM, Heikki Linnakangas
<heikki.linnakangas@enterprisedb.com> wrote:
> Yep :-(. I'm thinking we need to revert that change, now that we know there
> was a reason for only checking the end-of-WAL when doing archive recovery
> after all. We can keep it as a warning, though. With a hint explaining that
> if you're doing crash recovery from a database that crashed while an online
> backup was in progress, it's ok, but if you're recovering from an online
> backup, the backup is broken.

Yeah, we cannot distinguish those two cases. I agree to emit such a warning.

> For backups taken with pg_basebackup in 9.1, we can add a flag to the
> backup_label, indicating that the backup was taken with pg_basebackup. For
> such backups, the above scenario really should not happen, and we can still
> make it a hard error if it does.

Agreed.

Regards,

--
Fujii Masao
NIPPON TELEGRAPH AND TELEPHONE CORPORATION
NTT Open Source Software Center

pgsql-bugs by date:

Previous
From: Tom Lane
Date:
Subject: Re: BUG #5974: UNION construct type cast gives poor error message
Next
From: "Kevin Grittner"
Date:
Subject: Re: BUG #5974: UNION construct type cast gives poor error message