Index | Thread | Search

From:
Job Snijders <job@openbsd.org>
Subject:
Re: rpki-client: expose Manifest sequence number gaps in log & telemetry
To:
Theo Buehler <tb@theobuehler.org>, tech@openbsd.org
Date:
Sat, 2 Nov 2024 20:07:59 +0000

Download raw body.

Thread
On Sat, Nov 02, 2024 at 09:01:09PM +0100, Sebastian Benoit wrote:
> Job Snijders(job@openbsd.org) on 2024.11.02 12:28:06 +0000:
> > On Sat, Nov 02, 2024 at 01:07:20PM +0100, Theo Buehler wrote:
> > > On Sat, Nov 02, 2024 at 11:56:00AM +0000, Job Snijders wrote:
> > > > The below patch emits a warning per manifest, adds metrics to the
> > > > openmetrics output, and displays a summary at the end of the run.
> > > 
> > > I do wonder if we don't want to hide the warning under -v. But as you
> > > say, perhaps that encourages continuous frequent running of
> > > rpki-client.  We can reconsider this after seeing how it works in
> > > practice.
> > 
> > Yup, I started out hiding it under verbose, but then theorized the gaps
> > perhaps don't happen so often that it is too noisey... And in order to
> > do correlation it perhaps is nice to know *which* Manifest had a gap. :)
> > 
> > let's see what happens in practise.
> 
> yeah, i wonder how difficult it will be to find the reason for the gaps.
> I do hope people give enough time for their manifests to propagate, but i
> fear that DNS demonstrates they wont.

If you'd like a longer read on possible reasons:
https://mailarchive.ietf.org/arch/msg/sidrops/h4Gk1G8HeFXg43g2LiNJ2tv_9jo/

In short - it depends on your role in the ecosystem whether you can
meaningfully discover the reason

Kind regards,

Job