supervision - discussion about system services, daemon supervision, init, runlevel management, and tools such as s6 and runit
 help / color / mirror / Atom feed
From: Alex Kiernan <alex.kiernan@gmail.com>
To: Daniel Barlow <dan@telent.net>
Cc: supervision@list.skarnet.org
Subject: Re: s6-rc how to log oneshot services
Date: Tue, 7 Mar 2023 11:53:45 +0000	[thread overview]
Message-ID: <CAO5Uq5TpTOjZHnRvULp6_PWxvmft9jbzUQM-dwzjLODuFbHBiw@mail.gmail.com> (raw)
In-Reply-To: <87edq0suut.fsf@telent.net>

On Tue, Mar 7, 2023 at 10:45 AM Daniel Barlow <dan@telent.net> wrote:
>
>
> Hi all
>
> I've adopted s6 and s6-rc for a small distro that's targeted at
> consumer wifi routers and similar low-powered devices (same target
> devices as openwrt) and so far I am appreciating the consistency and the
> warm fuzzies that I get from having a supervision tree (and simple
> readiness checks!)  instead of a folder full of pid files that may or
> may not correspond with what's actually running.
>

We're very similar... my target has a decent amount of compute, but
still an embedded device with constrained RAM and flash.

> My question is: how should I handle logging for oneshot services? I
> have oneshots doing things that might fail, like configuring network
> interfaces or inserting modules, and I'd like those failure messages to
> go somewhere useful as the longrun messages do. If I just add
> a logging service to a oneshot with producer-for/consumer-for I get
>
> s6-rc-compile: fatal: longrun wlan.module-log declares a producer wlan.module of type oneshot
>
> so I guess that's not the answer ...
>
> What am I missing? I guess I could convert them into longruns that do
> the "up" actions then call pause(2), but that seems a bit of a
> ... non-traditional approach?
>

Interested in the right answer - right now we just have an icky hack
so it just goes to uncaught-logs...

redirfd -w 1 /run/service/s6-svscan-log/fifo

I've not actually checked since we converted PID1 to s6-linux-init if
we still need that - I suspect we may not.

-- 
Alex Kiernan

  reply	other threads:[~2023-03-07 11:54 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-03-07 10:45 Daniel Barlow
2023-03-07 11:53 ` Alex Kiernan [this message]
2023-03-07 15:06   ` Laurent Bercot
2023-03-07 17:17 ` Carlos Eduardo
2023-03-07 19:00   ` Laurent Bercot

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CAO5Uq5TpTOjZHnRvULp6_PWxvmft9jbzUQM-dwzjLODuFbHBiw@mail.gmail.com \
    --to=alex.kiernan@gmail.com \
    --cc=dan@telent.net \
    --cc=supervision@list.skarnet.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).