Horton

RE: Timeout parameters

Nagaura, Ryohei <nagaura.ryohei@jp.fujitsu.com>

View thread View on postgr.es
From: "Nagaura, Ryohei" <nagaura.ryohei@jp.fujitsu.com>
To: 'Fabien COELHO' <coelho@cri.ensmp.fr>
Cc: "Jamison, Kirk" <k.jamison@jp.fujitsu.com>, "Tsunakawa, Takayuki" <tsunakawa.takay@jp.fujitsu.com>, 'Robert Haas' <robertmhaas@gmail.com>, "MikalaiKeida@ibagroup.eu" <MikalaiKeida@ibagroup.eu>, "AYahorau@ibagroup.eu" <AYahorau@ibagroup.eu>, Michael Paquier <michael@paquier.xyz>, "pgsql-hackers@postgresql.org" <pgsql-hackers@postgresql.org>
Date: 2019-03-28T01:11:23Z
Lists: pgsql-hackers

Attachments

Hello, Fabien-san.

> From: Fabien COELHO <coelho@cri.ensmp.fr>
> About the backend v11 patch.
> No space or newline before ";". Same comment about the libpq_ timeout.
Fixed.

> There is an error in the code, I think it should be < 0 to detect errors.
Yes, thanks!

> If the parameter has no effect on Windows, I do not see why its value should be
> constrained to zero, it should just have no effect. Idem libpq_ timeout.
I had a misunderstanding.
Indeed, it doesn't need to be zero. Removed.

> Documentation:
> ISTM this is not about libpq connections but about TCP connections. There can be
> non libpq implementations client side.
Then, where do you think the correct place?
I thought that this parameter should be explained here because the communication
will be made through the library libpq same as keepalive features.


Best regards,
---------------------
Ryohei Nagaura

Commits

  1. Add support TCP user timeout in libpq and the backend server