libnetwork/pasta: always pass --quiet

[Luap99]

We want to log pasta warnings/errors even if pasta did not fail to start, however by default it prints a bit to noisy for us. With --quiet we get the things users should care about.

Note right now --quiet defaults to errors only but it should also do warnings based on the discussion with Stefano[1]. However this change is still safe to do right now.

[1] https://archives.passt.top/passt-dev/20240216114304.7234a83f@elisabeth/T/#m42652824644973674e84baf9e0bf1d0e88104450

[openshift-ci]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: Luap99

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [Luap99]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[Luap99]

@mheon PTAL
cc @sbrivio-rh

[Luap99]

/hold
Please do not merge, I am testing this right now but there is a problem with it.

[sbrivio-rh]

I thought you'd just pass --quiet, and we'd fix --quiet to log messages up to LOG_WARN (instead of LOG_ERR), and for the rest you'd leave everything as it is, instead of catching stdout.

[Luap99]

I thought you'd just pass --quiet, and we'd fix --quiet to log messages up to LOG_WARN (instead of LOG_ERR), and for the rest you'd leave everything as it is, instead of catching stdout.

Yes that is my goal but do you write anything do stdout? I don't think so and IMO it is easier that way to code here. I can of course make the code only read stderr if that makes a real difference.

ps: I am in the process to send a patch to fix pasta.

[sbrivio-rh]

I thought you'd just pass --quiet, and we'd fix --quiet to log messages up to LOG_WARN (instead of LOG_ERR), and for the rest you'd leave everything as it is, instead of catching stdout.

Yes that is my goal but do you write anything do stdout?

No, just for -h / --help (https://bugs.passt.top/show_bug.cgi?id=52) and --version.

I don't think so and IMO it is easier that way to code here.

Sure.

I can of course make the code only read stderr if that makes a real difference.

No no, it shouldn't -- I was just wondering.

ps: I am in the process to send a patch to fix pasta.

Oh, thank you.

[mheon]

LGTM

[Luap99]

OK this should be good right now, logging this on warning right now is not possible as it is still to noisy.
For the first problem I posted https://archives.passt.top/passt-dev/20240222171739.141623-2-pholzing@redhat.com/
but I also see a dns warning Couldn't get any nameserver address which I need to investigate further.

[rhatdan]

/lgtm
/hold

[rhatdan]

/unhold