Streaming replication

On 31/03/15 12:45, Tatsuo Ishii wrote:

In the doc:

25.2.5. Streaming Replication
:
The standby connects to the primary, which streams WAL records to the
standby as they're generated, without waiting for the WAL file to be
filled.

This seems to claim that walsender sends WAL files which has not been
fsync'd yet. However, in walsender.c:

/*
* Streaming the current timeline on a master.
*
* Attempt to send all data that's already been written out and
* fsync'd to disk. We cannot go further than what's been written out
* given the current implementation of XLogRead(). And in any case
* it's unsafe to send WAL that is not securely down to disk on the
* master: if the master subsequently crashes and restarts, slaves
* must not have applied any WAL that gets lost on the master.
*/

This one says walsender sends WAL records as long as there are
fsync'd.

Am I missing something?

I think the docs are trying to say that streaming replication doesn't
wait for a (16MB) WAL *file* to be filled. but sends each (fsync'd) WAL
record. I had to reread it several times too :-)

Cheers

Mark

--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers

On 31/03/15 12:45, Tatsuo Ishii wrote:

In the doc:

25.2.5. Streaming Replication
:
The standby connects to the primary, which streams WAL records to the
standby as they're generated, without waiting for the WAL file to be
filled.

This seems to claim that walsender sends WAL files which has not been
fsync'd yet. However, in walsender.c:

/*
* Streaming the current timeline on a master.
*
* Attempt to send all data that's already been written out and
* fsync'd to disk. We cannot go further than what's been written out
* given the current implementation of XLogRead(). And in any case
* it's unsafe to send WAL that is not securely down to disk on the
* master: if the master subsequently crashes and restarts, slaves
* must not have applied any WAL that gets lost on the master.
*/

This one says walsender sends WAL records as long as there are
fsync'd.

Am I missing something?

I think the docs are trying to say that streaming replication doesn't
wait for a (16MB) WAL *file* to be filled. but sends each (fsync'd) WAL
record. I had to reread it several times too :-)

Thanks for the explanation. That makes sense. The docs definitely has
a room for improvement.

Best regards,
--
Tatsuo Ishii
SRA OSS, Inc. Japan
English: http://www.sraoss.co.jp/index_en.php
Japanese:http://www.sraoss.co.jp

--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers

The key word you're misunderstanding is "filled". It means it doesn't wait
for the 16MB file to be completely filled with records. I.e. what would
happen in the file shipping form of replication.