RE: Timeout parameters

Hi Nagaura-san,

On Monday, March 25, 2019 2:26 PM (GMT+9), Ryohei Nagaura wrote:

>Yes, I want to commit TCP_USER_TIMEOUT patches in PG12.
>Also I'd like to continue to discuss about socket_timeout after this CF.
Ok. So I'd only take a look at TCP_USER_TIMEOUT parameter for now (this CommitFest),
and maybe we could resume the discussion on socket_timeout in the future.

> About TCP_USER_TIMEOUT:
> 1) Documentation and checking on UNIX system As far as I surveyed
> (solaris and BSD family), these UNIX OS don't have "TCP_USER_TIMEOUT" parameter.
> Accordingly I have not checked on real machine.
> Also, I modified documentations to remove "equivalent to socket option"

Your patch applies, however in TCP_backend_v10 patch,
your documentation is missing a closing tag </varlistentry>
so it could not be tested.
When that's fixed, it passes the make check.

Also regarding docs,
what's the reason to emphasize Windows not supported in a separate paragraph?
How about simplifying and combining the 2 paragraphs such as below?

Specifies the number of milliseconds after which a TCP connection can be
aborted by the operating system due to network problems when sending or
receiving data through this connection. A value of zero uses the system default.
For connections made via a Unix-domain socket, this parameter is ignored. This
parameter is supported only on systems that support <literal>TCP_USER_TIMEOUT</literal>;
on other systems such as Windows, it has no effect and must be zero.

Or if you really want to emphasize which systems are supported and which are not,
you may remove the last sentence above, then indicate that in the note tag.
Example:
This parameter is supported only on systems that support
<literal>TCP_USER_TIMEOUT</literal> such as Linux version 2.6.37 or later;
on other systems such as Windows, it has no effect and must be zero.

(Note: I haven't checked which Linux versions are supported,
I got it from your previous patch version.)

Regards,
Kirk Jamison