From: | Bharath Rupireddy <bharath(dot)rupireddyforpostgres(at)gmail(dot)com> |
---|---|
To: | Michael Paquier <michael(at)paquier(dot)xyz> |
Cc: | Ronan Dunklau <ronan(dot)dunklau(at)aiven(dot)io>, PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org> |
Subject: | Re: improve pg_receivewal code |
Date: | 2021-09-17 06:16:33 |
Message-ID: | CALj2ACXV9EjvTUKrwaMBGMic7DiaaHwR2-yMWMKKGp7a6G=+jw@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Thu, Sep 16, 2021 at 9:31 AM Michael Paquier <michael(at)paquier(dot)xyz> wrote:
>
> > Here's the v2 with above modifications.
>
> I was looking at this patch, and I agree that checking for the system
> ID and the timeline by setting sysidentifier beforehand looks like an
> improvement.
>
> The extra IDENTIFY_SYSTEM done at the beginning of StreamLog() is not
> a performance bottleneck as we run it only once for each loop. I
> don't really get the argument of a server replacing another one on the
> same port requiring to rely only on the first system ID fetched before
> starting the loops of StreamLog() calls. So I would leave main()
> alone, but fill in the system ID from RunIdentifySystem() in
> StreamLog().
Thanks. I changed the code that way. PSA v3 patch.
Regards,
Bharath Rupireddy.
Attachment | Content-Type | Size |
---|---|---|
v3-0001-improve-pg_receivewal-code.patch | application/octet-stream | 4.5 KB |
From | Date | Subject | |
---|---|---|---|
Next Message | Amit Kapila | 2021-09-17 06:18:16 | Re: POC: Cleaning up orphaned files using undo logs |
Previous Message | houzj.fnst@fujitsu.com | 2021-09-17 06:06:02 | RE: [BUG] Unexpected action when publishing partition tables |