From: Vladimir Davydov <vdavydov.dev@gmail.com>
To: Georgy Kirichenko <georgy@tarantool.org>
Cc: tarantool-patches@freelists.org
Subject: Re: [tarantool-patches] [PATCH 3/7] Replication: Relay does not rely on xlog boundaries
Date: Wed, 21 Aug 2019 14:35:11 +0300 [thread overview]
Message-ID: <20190821113511.GA13834@esperanza> (raw)
In-Reply-To: <cf8b2e805c90c38843edb1133dcb14d89bc9809a.1565676868.git.georgy@tarantool.org>
On Tue, Aug 13, 2019 at 09:27:41AM +0300, Georgy Kirichenko wrote:
> When relay got an ACK then it matches received vclock against xlog file
> boundaries detected using on_close_log trigger and send a consumer
> advance message. However, for each ACK relay send a status update
> message to the tx cord which could be used for gc purposes.
> This patch removes any knowledge about xlog boundaries from relay
> because there would not any xlog files in case of in-memory replication.
> As gc now tracks all xlog files then it is able to handle garbage files
> using relay status updates.
>
> Note: after parallel applier there is no more one ACK per transaction
> so it should not be too expensive to advance a consumer on each status
> update. However I think it could be improved, for instance with tracking
> the next wal file vclock.
I don't see a way to preserve the old behavior after switching to
replication from memory so I guess this patch is okay.
next prev parent reply other threads:[~2019-08-21 11:35 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-08-13 6:27 [tarantool-patches] [PATCH 0/7] Replication: In-memory replication Georgy Kirichenko
2019-08-13 6:27 ` [tarantool-patches] [PATCH 1/7] Refactoring: wal writer fiber and queue Georgy Kirichenko
2019-08-16 13:53 ` [tarantool-patches] " Konstantin Osipov
2019-08-20 10:57 ` Георгий Кириченко
2019-08-21 10:18 ` [tarantool-patches] " Vladimir Davydov
2019-08-13 6:27 ` [tarantool-patches] [PATCH 2/7] Refactoring: Track wal files using gc state Georgy Kirichenko
2019-08-21 10:44 ` Vladimir Davydov
2019-08-13 6:27 ` [tarantool-patches] [PATCH 3/7] Replication: Relay does not rely on xlog boundaries Georgy Kirichenko
2019-08-21 11:35 ` Vladimir Davydov [this message]
2019-08-13 6:27 ` [tarantool-patches] [PATCH 4/7] Replication: wal memory buffer Georgy Kirichenko
2019-08-21 11:57 ` Vladimir Davydov
2019-08-13 6:27 ` [tarantool-patches] [PATCH 5/7] Replication: in memory replication Georgy Kirichenko
2019-08-21 13:52 ` Vladimir Davydov
2019-08-13 6:27 ` [tarantool-patches] [PATCH 6/7] Refactoring: remove wal_watcher routines Georgy Kirichenko
2019-08-21 13:52 ` Vladimir Davydov
2019-08-13 6:27 ` [tarantool-patches] [PATCH 7/7] Refactoring: get rid of on_close_log Georgy Kirichenko
2019-08-21 13:52 ` Vladimir Davydov
2019-08-16 13:47 ` [tarantool-patches] Re: [PATCH 0/7] Replication: In-memory replication Konstantin Osipov
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20190821113511.GA13834@esperanza \
--to=vdavydov.dev@gmail.com \
--cc=georgy@tarantool.org \
--cc=tarantool-patches@freelists.org \
--subject='Re: [tarantool-patches] [PATCH 3/7] Replication: Relay does not rely on xlog boundaries' \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox