Home > Archive > PostgreSQL Administration > November 2006 > [Fwd: Checkpoint Location Format]









You are viewing an archived Text-only version of the thread. To view this thread in it's original format and/or if you want to reply to this thread please [click here]

 

Author [Fwd: Checkpoint Location Format]
Andy Shellam

2006-11-07, 7:18 pm

Further to the below e-mail, I've come across a slight problem.
The starting checkpoint location in a backup I've just taken is
"0/22A3190" (note: 7 digits after the /, not 6 as I first thought.)
However the .backup file is called <WAL_FILE>.002A3190 (ie. it only
takes the right-most 6 digits).

Can someone confirm this is the correct case?

Thanks,

Andy.


-------- Original Message --------
Subject: Checkpoint Location Format
Date: Tue, 07 Nov 2006 13:12:53 +0000
From: Andy Shellam (Mailing Lists) <andy.shellam-lists@mailnetwork.co.uk>
To: pgsql-admin@postgresql.org



Hi,

I'm writing an automated file-system level backup application for use
with WAL archiving, that will issue the pg_start_backup call, tar and
gzip the cluster data directory, issue the pg_stop_backup call, and
remove all previous un-needed WAL files from the archive.

I need to write a regular expression that will search for the WAL
filename and checkpoint location from the backup_label file, and just
want to clarify that the checkpoint location will always be of the
format: X/XXXXXX - where X is one of 0-9, A-F?

And then the WAL .backup file that is generated in the archive, has a
filename of the form:

<WAL_FILE>.00XXXXXX.backup

where <WAL_FILE> is the name of the "STARTING WAL LOCATION" directive in
the backup_label file, and XXXXXX is the last 6 digits of the checkpoint
(after the / )?

Thanks,

Andy.



---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match

Tom Lane

2006-11-07, 7:18 pm

"Andy Shellam (Mailing Lists)" <andy.shellam-lists@mailnetwork.co.uk> writes:
> Further to the below e-mail, I've come across a slight problem.
> The starting checkpoint location in a backup I've just taken is
> "0/22A3190" (note: 7 digits after the /, not 6 as I first thought.)
> However the .backup file is called <WAL_FILE>.002A3190 (ie. it only
> takes the right-most 6 digits).


Yeah, the second part of the .backup file name is the offset within the
corresponding WAL file, which for the standard file size of 16M can only
range up to 6 hex digits. The higher-order bits of the second word of
the checkpoint location go into the last part of the WAL file name.
That is, checkpoint location xxxxxxxx/yyzzzzzz should be associated
with a backup file named like ttttttttxxxxxxxx0000
00yy.00zzzzzz.backup.
(I think ... haven't had any caffeine yet this morning ...)

regards, tom lane

---------------------------(end of broadcast)---------------------------
TIP 4: Have you searched our list archives?

http://archives.postgresql.org

Sponsored Links





Also available: Server administration forum archive | Web Design forum archive | Software forum archive | Hardware reviews archive | Programming forum archive

Copyright 2008 droptable.com