Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:112493
Date: Sun, 13 Dec 2020 15:01:48 +0000
User-Agent: K-9 Mail for Android
In-Reply-To: <CAH5C8xW981Fn91HsHmZfbLseXwSyOtM1X5u1zsAL4ofXhka=nw@mail.gmail.com>
References: <CAH5C8xW981Fn91HsHmZfbLseXwSyOtM1X5u1zsAL4ofXhka=nw@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain;
 charset=utf-8
Content-Transfer-Encoding: quoted-printable
To: PHP Internals List <internals@lists.php.net>
Message-ID: <A4EB5880-5866-4501-A7A3-C411B23C4166@gmail.com>
Subject: Re: [PHP-DEV] [RFC] [Discussion] Measuring maximum execution time based on wall-time
From: rowan.collins@gmail.com (Rowan Tommins)

Hi M=C3=A1t=C3=A9,

On 13 December 2020 00:25:10 GMT+00:00, "M=C3=A1t=C3=A9 Kocsis" <kocsismat=
e90@gmail=2Ecom> wrote:
>RFC: https://wiki=2Ephp=2Enet/rfc/max_execution_wall_time

I wonder if you could give some more detailed examples of what you would u=
se this for=2E

You write:

> [The current functionality] can have serious consequences for distribute=
d systems with high traffic, where finishing a request in a timely manner i=
s essential for avoiding cascading failures=2E

It feels like "finishing in a timely manner" is rather different from "bei=
ng forcefully killed after a fixed time limit"=2E

I'm struggling to picture when I'd want a hard wall-time limit, rather tha=
n:

- checking the script's duration at key points where I know it can gracefu=
lly exit, ensure a consistent state, and return an appropriate message to t=
he caller
- enforcing a network timeout on the calling end so that I don't need to r=
ely on all services cooperatively exiting in good time


Regards,

--=20
Rowan Tommins
[IMSoP]