From: Kirill Yukhin via Tarantool-patches <tarantool-patches@dev.tarantool.org>
To: Serge Petrenko <sergepetrenko@tarantool.org>
Cc: tarantool-patches@dev.tarantool.org, v.shpilevoy@tarantool.org
Subject: Re: [Tarantool-patches] [PATCH] relay: yield explicitly every N sent rows
Date: Fri, 26 Feb 2021 08:41:36 +0000 [thread overview]
Message-ID: <20210226084136.h7jtkw3jlysv6atj@tarantool.org> (raw)
In-Reply-To: <20210212112541.27561-1-sergepetrenko@tarantool.org>
Hello,
On 12 фев 14:25, Serge Petrenko via Tarantool-patches wrote:
> While sending a WAL, relay only yields in `coio_write_xrow`, once it
> sees the socket isn't ready for writes.
> It may happen that the socket is always ready for a long period of time,
> and relay doesn't yield at all while recovering a whole .xlog file. This
> may take well more than a minute.
> During this period of time, relay doesn't read replica's ACKs due to
> relay reader fiber not being scheduled, and once the reader is finally
> live it times out immediately, causing the replica to reconnect.
>
> The problem is amplified by the fact that replica waits for
> replication_timeout to pass prior to reconnecting, which lets master
> pile up even more ready WALs, and effectively making it impossible for
> the replica to sync.
>
> To fix the problem let's yield explicitly in relay_send_row every
> WAL_ROWS_PER_YIELD rows. The same is already done in local recovery, and
> serves the same purpose: to not block the event loop for too long.
>
> Closes #5762
> ---
> No test provided as the fix is quite obvious but rather hard to test
> automatically.
> https://github.com/tarantool/tarantool/issues/5762
> https://github.com/tarantool/tarantool/tree/sp/gh-5762-relay-yield
I've checked your patch into 2.6, 2.6 and master.
--
Regards, Kirill Yukhin
next prev parent reply other threads:[~2021-02-26 8:41 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-02-12 11:25 Serge Petrenko via Tarantool-patches
2021-02-12 11:37 ` Cyrill Gorcunov via Tarantool-patches
2021-02-12 11:46 ` Cyrill Gorcunov via Tarantool-patches
2021-02-12 12:08 ` Serge Petrenko via Tarantool-patches
2021-02-12 17:00 ` Cyrill Gorcunov via Tarantool-patches
2021-02-12 21:48 ` Vladislav Shpilevoy via Tarantool-patches
2021-02-12 22:25 ` Cyrill Gorcunov via Tarantool-patches
2021-02-15 8:45 ` Serge Petrenko via Tarantool-patches
2021-02-15 8:40 ` Serge Petrenko via Tarantool-patches
2021-02-17 21:11 ` Vladislav Shpilevoy via Tarantool-patches
2021-02-18 20:24 ` Serge Petrenko via Tarantool-patches
2021-02-23 22:30 ` Vladislav Shpilevoy via Tarantool-patches
2021-02-24 9:48 ` Serge Petrenko via Tarantool-patches
2021-02-24 10:15 ` Cyrill Gorcunov via Tarantool-patches
2021-02-24 10:35 ` Serge Petrenko via Tarantool-patches
2021-02-24 12:07 ` Cyrill Gorcunov via Tarantool-patches
2021-02-24 12:14 ` Serge Petrenko via Tarantool-patches
2021-02-24 22:20 ` Vladislav Shpilevoy via Tarantool-patches
2021-02-26 8:41 ` Kirill Yukhin via Tarantool-patches [this message]
2021-02-26 20:24 ` Vladislav Shpilevoy via Tarantool-patches
2021-03-01 11:25 ` Serge Petrenko via Tarantool-patches
2021-03-01 21:24 ` Vladislav Shpilevoy via Tarantool-patches
2021-03-02 9:52 ` Kirill Yukhin via Tarantool-patches
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=20210226084136.h7jtkw3jlysv6atj@tarantool.org \
--to=tarantool-patches@dev.tarantool.org \
--cc=kyukhin@tarantool.org \
--cc=sergepetrenko@tarantool.org \
--cc=v.shpilevoy@tarantool.org \
--subject='Re: [Tarantool-patches] [PATCH] relay: yield explicitly every N sent rows' \
/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