From: | Антон Пацев <patsev(dot)anton(at)gmail(dot)com> |
---|---|
To: | Justin Pryzby <pryzby(at)telsasoft(dot)com> |
Cc: | pgsql-hackers(at)lists(dot)postgresql(dot)org |
Subject: | Re: Proposal: remove string "contains errors; unaffected changes were applied" |
Date: | 2020-06-02 09:33:33 |
Message-ID: | CAOx4deoOSkY6B-ta1zyW8jWAf1Xni0b+WNjZfsUqO3rgeePy6w@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
No. I think the message be sent to the log.
i created diagram.
May be check postgresql.conf of valid or invalid?
[image: contains errors unaffected changes were applied.png]
вс, 31 мая 2020 г. в 23:31, Justin Pryzby <pryzby(at)telsasoft(dot)com>:
> On Sun, May 31, 2020 at 01:43:45PM +0600, Антон Пацев wrote:
> > Hello!
> > When parameter cannot be changed without restarting the server postgresql
> > write:
> > "LOG: configuration file "/var/lib/postgresql/data/postgresql.auto.conf"
> > contains errors; unaffected changes were applied"
> > May be not write this string to LOG?
> >
> > This string confuses people. If all log send to ELK, then
> > administrator think that postgresql have error. But postgresql do not
> have
> > error.
>
> I think you're suggesting that the message should be sent to the client,
> but
> not to the log.
>
> But I think it *should* go to the log; otherwise, a bad change might cause
> the
> server to later refuse to start. Any admins or monitoring system watching
> the
> log should have the ability to see that a change isn't effective. That's
> also
> why (in my mind) we have pg_settings.pending_restart.
>
> --
> Justin
>
--
С уважением, Антон Пацев.
Best regards, Anton Patsev.
From | Date | Subject | |
---|---|---|---|
Next Message | Amit Kapila | 2020-06-02 09:34:39 | Re: Resetting spilled txn statistics in pg_stat_replication |
Previous Message | Andy Fan | 2020-06-02 09:04:50 | Re: Hybrid Hash/Nested Loop joins and caching results from subplans |