Re: Include WAL in base backup

Fujii Masao <masao.fujii@gmail.com>

From: Fujii Masao <masao.fujii@gmail.com>
To: Heikki Linnakangas <heikki.linnakangas@enterprisedb.com>
Cc: Magnus Hagander <magnus@hagander.net>, Tom Lane <tgl@sss.pgh.pa.us>, Stephen Frost <sfrost@snowman.net>, PostgreSQL-development <pgsql-hackers@postgresql.org>
Date: 2011-01-29T07:10:19Z
Lists: pgsql-hackers
On Sat, Jan 29, 2011 at 6:02 AM, Heikki Linnakangas
<heikki.linnakangas@enterprisedb.com> wrote:
> On 27.01.2011 06:44, Fujii Masao wrote:
>>
>> +               XLByteToSeg(endptr, endlogid, endlogseg);
>> <snip>
>> +                       /* Have we reached our stop position yet? */
>> +                       if (logid>  endlogid ||
>> +                               (logid == endlogid&&  logseg>= endlogseg))
>> +                               break;
>>
>> What I said in upthread is wrong. We should use XLByteToPrevSeg
>> for endptr and check "logseg>  endlogseg". Otherwise, if endptr is
>> not a boundary byte, endlogid/endlogseg indicates the last
>> necessary WAL file, but it's not sent.
>
> We should use XLByteToPrevSeg, but I believe >= is still correct.
> logid/logseg is the last WAL segment we've successfully sent, and
> endlogif/endlogid is the last WAL segment we need to send. When they are the
> same, we're done.

Really? logid/logseg is incremented just before the check as follows.
So, when they are the same, the WAL file which logid/logseg indicates
has not been sent yet. Am I missing something?

+			/* Advance to the next WAL file */
+			NextLogSeg(logid, logseg);
+
+			/* Have we reached our stop position yet? */
+			if (logid > endlogid ||
+				(logid == endlogid && logseg >= endlogseg))
+				break;

Regards,

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