Bug 439174 – rc-update (and other rc utilities) need to be quiet when --quiet is specified

• Product: Gentoo Hosted Projects
• Status: UNCONFIRMED ---

[Description Alon Bar-Lev 2012-10-21 15:24:27 UTC]

example:
---
# rc-update -q add apache2 && echo ok
 * rc-update: apache2 already installed in runlevel `default'; skipping
ok
---

In quiet mode the warning should not be presented.

Thanks!

[Comment 1 William Hubbs 2012-11-30 22:53:34 UTC]

This bug is here as a result of bug #380073. The idea of that bug was
that warnings are important enough that you want to see them, so they
should not be suppressed, even in quiet mode. Info messages (e.g. einfo)
are still suppressed however. So, I suppose the question is, should
these messages be downgraded from warning to info?

[Comment 2 Maxim Kammerer 2012-11-30 23:14:39 UTC]

I think that there should be a switch to suppress warnings as well, or at least certain types of warnings. E.g, "start-stop-daemon -K" complains when a daemon has already terminated, and this is not always desirable.

[Comment 3 William Hubbs 2012-12-01 03:42:54 UTC]

(In reply to comment #2)
> I think that there should be a switch to suppress warnings as well, or at
> least certain types of warnings. E.g, "start-stop-daemon -K" complains when
> a daemon has already terminated, and this is not always desirable.

Can we reclassify this message though as purely informational instead of being a warning? e.g. there is no special action you need to take because of it, we are just letting you know that the daemon is already terminated.

[Comment 4 Alon Bar-Lev 2012-12-01 07:17:22 UTC]

Hi,

There must be a way to suppress *ALL* warning for interface and interact with scripts.

The only result should be exit code as in any other quite mode.

It can be -qqqqqq or anything you like, but this mode is important.

Thanks,
Alon

[Comment 5 Maxim Kammerer 2012-12-01 12:46:04 UTC]

(In reply to comment #3)
> Can we reclassify this message though as purely informational instead of
> being a warning? e.g. there is no special action you need to take because of
> it, we are just letting you know that the daemon is already terminated.

Not sure about that. As it is now, the message says that the daemon either terminated prematurely by itself (no error), or crashed (something that should be investigated). I see no way that OpenRC can distinguish between the two situations.