xlogdump fixups and WAL log question.

From: Theo Schlossnagle <jesus(at)omniti(dot)com>
To: pgsql-hackers(at)postgresql(dot)org
Cc: Theo Schlossnagle <jesus(at)omniti(dot)com>
Subject: xlogdump fixups and WAL log question.
Date: 2006-10-20 17:18:04
Message-ID: 8BC09726-CF44-4DFD-B00D-210D0AE45A3D@omniti.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Not sure who cares, so xzilla indicated I should drop a note here. I
just made the xlogdump stuff work for 8.1 (trivial) and fixed a few
other small issues that caused it to not work right both generally
and in our environment.

http://pgfoundry.org/tracker/index.php?
func=detail&aid=1000760&group_id=1000202&atid=772

We're using it to track down what's causing some wal log ruckus.
We're generating about a quarter terabyte of WAL logs a day (on bad
days) which is posing some PITR backup pains. That amount isn't a
severe challenge to backup, but our previous install was on Oracle
and it generated substantially less archive redo logs (10-20 gigs per
day).

Is it possible to create tables in fashion that will not write info
to the WAL log -- knowingly and intentionally making them
unrecoverable? This is very desirable for us. We snapshot tables
from a production environment. If the database goes down and we
recover, the old snapshots are out of date anyway and serve no useful
purpose. The periodic snapshot procedure would re-snap them in short
order anyway. I'd like to do:

INSERT INTO TABLE tblfoo_snap1 AS SELECT * from <table on remote
database> NO LOGGING;

(NO LOGGING being the only part we're currently missing) Is something
like this possible?

Cheers ;-)
Theo

// Theo Schlossnagle
// CTO -- http://www.omniti.com/~jesus/
// OmniTI Computer Consulting, Inc. -- http://www.omniti.com/

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Magnus Hagander 2006-10-20 17:54:51 zic with msvc
Previous Message mark 2006-10-20 17:11:16 Re: [SPAM?] Re: Asynchronous I/O Support