RE: Timeout parameters

From: "Jamison, Kirk" <k(dot)jamison(at)jp(dot)fujitsu(dot)com>
To: "Nagaura, Ryohei" <nagaura(dot)ryohei(at)jp(dot)fujitsu(dot)com>
Cc: "Tsunakawa, Takayuki" <tsunakawa(dot)takay(at)jp(dot)fujitsu(dot)com>, 'Michael Paquier' <michael(at)paquier(dot)xyz>, "AYahorau(at)ibagroup(dot)eu" <AYahorau(at)ibagroup(dot)eu>, "'Fabien COELHO'" <coelho(at)cri(dot)ensmp(dot)fr>, "'pgsql-hackers(at)postgresql(dot)org'" <pgsql-hackers(at)postgresql(dot)org>, "MikalaiKeida(at)ibagroup(dot)eu" <MikalaiKeida(at)ibagroup(dot)eu>
Subject: RE: Timeout parameters
Date: 2019-02-27 01:35:50
Message-ID: D09B13F772D2274BB348A310EE3027C645CE6B@g01jpexmbkw24
Views: Raw Message | Whole Thread | Download mbox | Resend email
Lists: pgsql-hackers

Hi Nagaura-san,

Your socket_timeout patch still does not apply either with
git or patch command. It says it's still corrupted.
I'm not sure about the workaround, because the --ignore-space-change
and --ignore-whitespace did not work for me.
Maybe it might have something to do with your editor when creating
the patch. Could you confirm?
The CFbot is also another way to check if your latest patches work.

On Tuesday, February 26, 2019 5:16PM (GMT+9), Nagaura, Ryohei wrote:
> "communication inactivity" seems to be a little extreme.
> If the communication layer is truly dead you will use keepalive.
> This use case is when socket option is not available for some reason.
> So it would be better "terminating the connection" in my thought.

About the doc. Alright.
- There's a typo in the doc: connectoin --> connection
- In the code, there's a part where between 0-2 seconds
is interpreted as 2 because it's the minimum time. I think
the comment should be improved, and use insist (force) instead
of "need". This should also be added in the documentation.

* Rounding could cause communication to fail; need at least 2 secs

* Rounding could cause communication to fail;
* insist on at least two seconds.

- added how it should be written as decimal integer, similar
to connect_timeout
- used closing instead of terminating, to be consistent with
other params
- added the minimum allowed timeout


Controls the number of second (write as a decimal integer, e.g. 10) of
client's waiting time for individual socket read/write operations
before closing the connection. This can be used both as a force global
query timeout and network problems detector. A value of zero (the
default) turns this off, which means wait indefinitely. The minimum
allowed timeout is 2 seconds, so a value of 1 is interpreted as 2.

> About TCP_USER_TIMEOUT patches, there are only miscellaneous changes:
> removing trailing spaces and making comments of parameters lower case
> as you pointed out.
Thanks for the update.
Confirmed the change.

Kirk Jamison

In response to


Browse pgsql-hackers by date

  From Date Subject
Next Message Haribabu Kommi 2019-02-27 02:06:51 Re: Pluggable Storage - Andres's take
Previous Message Jeff Janes 2019-02-27 01:35:18 Re: COPY FREEZE and setting PD_ALL_VISIBLE/visibility map bits