Re: [PATCHES] Forcing current WAL file to be archived - Mailing list pgsql-hackers

From Hannu Krosing
Subject Re: [PATCHES] Forcing current WAL file to be archived
Date
Msg-id 1155419978.5899.42.camel@localhost.localdomain
Whole thread Raw
In response to Re: [PATCHES] Forcing current WAL file to be archived  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
Ühel kenal päeval, L, 2006-08-12 kell 10:59, kirjutas Tom Lane:
> Hannu Krosing <hannu@skype.net> writes:
> > Ühel kenal päeval, K, 2006-08-09 kell 10:57, kirjutas Tom Lane:
> >> Insert points to the next byte to be written within the internal WAL
> >> buffers.  The byte(s) preceding it haven't necessarily gotten out of
> >> those buffers yet.  Write points to the end of what we've actually
> >> written to the kernel,
>
> > I assume that it also points to the byte after what is written to
> > kernel, or is it tha last byte written ?
>
> Right, it's really first-unwritten-byte for all three pointers.
> The two newly added functions to convert WAL locations to filenames
> use XLByteToPrevSeg(), so they should do the right thing here
> (see comments in src/include/access/xlog_internal.h).

How do they behave exactly at the file boundary ?

That is will it point 1 byte past end of old file, or byte 0 of the new
one ?

>             regards, tom lane
--
----------------
Hannu Krosing
Database Architect
Skype Technologies OÜ
Akadeemia tee 21 F, Tallinn, 12618, Estonia

Skype me:  callto:hkrosing
Get Skype for free:  http://www.skype.com




pgsql-hackers by date:

Previous
From: Martijn van Oosterhout
Date:
Subject: Re: Buildfarm owners: check if your HEAD build is stuck
Next
From: Andrew Dunstan
Date:
Subject: list archives not being updated?