We really need to set a sunset date and retire smolt. I was hoping the new census project would be up and running before then, but we can't keep smolt really limping along forever.

Looks really cool, congrats to all that made it happen! Will there be banners for when maintenance is going to happen/in progress so that if there's an issue and we go there it will let us know if it's an unplanned outage or whether it's part of planned maintenance?

Also it would be great if we could have a second section at the bottom for secondary arch koji too :)

On Fri, Aug 31, 2012 at 9:38 AM, Kevin Fenzi <kevin(a)scrye.com&gt; wrote: > On Fri, 31 Aug 2012 09:45:49 +0100 > Peter Robinson <pbrobinson(a)gmail.com&gt; wrote: > >> Looks really cool, congrats to all that made it happen! >> >> Will there be banners for when maintenance is going to happen/in >> progress so that if there's an issue and we go there it will let us >> know if it's an unplanned outage or whether it's part of planned >> maintenance? > > Yep. Planned outages are different appearing than a unplanned > outage. ;) > >> Also it would be great if we could have a second section at the >> bottom for secondary arch koji too :) > > We could look at setting secondary folks up with a similar instance? > > The code is all available in the git repo, you just need to setup a > openshift instance to run it and and point dns to it. I might have missed something, but are there plans to add bugzilla to this?

>> I might have missed something, but are there plans to add bugzilla >> to this? > > It was suggested, but we don't actually run bugzilla.redhat.com or > have anything to do with it. So it would be a bit difficult to try > and report status on it when all we could see is 'it's down' or 'it > seems to be up now' That by itself would be enough for me. And for anyone who sees some failure and has no idea what's going on. Even with a not saying "maintained by RH, not Fedora Inf", or somesuch.

devel-bounces(a)lists.fedoraproject.org wrote on 08/31/2012 12:46:31: > From: Kevin Fenzi <kevin(a)scrye.com&gt; > To: devel(a)lists.fedoraproject.org > Date: 08/31/2012 12:47 > Subject: Re: Fedora Infrastructure announces status.fedoraproject.org > Sent by: devel-bounces(a)lists.fedoraproject.org > > On Fri, 31 Aug 2012 10:21:26 -0500 > Jon Ciesla <limburgher(a)gmail.com&gt; wrote: > > > >> I might have missed something, but are there plans to add bugzilla > > >> to this? > > > > > > It was suggested, but we don't actually run bugzilla.redhat.com or > > > have anything to do with it. So it would be a bit difficult to try > > > and report status on it when all we could see is 'it's down' or 'it > > > seems to be up now' > > > > That by itself would be enough for me. And for anyone who sees some > > failure and has no idea what's going on. Even with a not saying > > "maintained by RH, not Fedora Inf", or somesuch. > > I'd really like to avoid that, because it would lead to: > > "when is it going to be up?" -> no idea. > "It works for me, but not for otheruser, is there a network problem?" > -> no idea. > "You are saying it's down, but it's up for me" (mark it up, site goes > down again because of some ongoing problem) "Now it's down, please mark > it down again" -> If we managed it, we would know the ongoing issue and > could update when it was approprate/fully back up. > > And cases where we mark it down, when it's not, or only for some users, > etc. > > It also opens the door for "Oh, can you monitor my fedora related site > and report on it there?" which I REALLY don't want to do. The updates > are a manual process, so it's already another step someone in > Infrastructure has to do when there is an outage. > > Lets keep it to sites that Fedora Infrastructure directly manages. > > kevin > [attachment "signature.asc" deleted by John Florian/EngMOp/MAS/DCC] -- > devel mailing list > devel(a)lists.fedoraproject.org > https://admin.fedoraproject.org/mailman/listinfo/devel That makes lots of sense, but I suspect you'll get just as many "why isn't bugzilla shown here?" questions ... unless maybe the page had some hint that it only monitors sites directly managed by the Fedora Project, which is why sites like bugzilla, etc., etc. are not shown here.

-- John Florian -- devel mailing list devel(a)lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel

On Fri, 31 Aug 2012 12:51:22 -0500 Jon Ciesla <limburgher(a)gmail.com&gt; wrote: > > That makes lots of sense, but I suspect you'll get just as many > > "why isn't bugzilla shown here?" questions ... unless maybe the > > page had some hint that it only monitors sites directly managed by > > the Fedora Project, which is why sites like bugzilla, etc., etc. > > are not shown here. We could add something to that effect perhaps.

Another issue is that we don't monitor bugzilla at all. With our other services if something goes down, we are paged and the first thing the person handling the outage can do is set the status. With no monitoring we would depend on someone telling us when things are down and someone who can update status being around to act on it. This would lead to it's status being out of sync a great deal more than services we monitor and manage. ;) > I see your point, but I'm swayed by Kevin's logic. :)

kevin -- devel mailing list devel(a)lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel

We could add something to that effect perhaps.

> Kevin Fenzi wrote: > > We could add something to that effect perhaps. > > "We will not and cannot monitor the following Fedora-related > services: -Red Hat Bugzilla > -RPMFusion > -Fedora blogs > -Fedora news sites > -Your Grandma's Fedora laptop" Exactly! Great examples BTW and I really like the order they're presented in, starting with those everyone is likely to ask about first down to those that some will wonder about.

> From: Kevin Fenzi <kevin(a)scrye.com&gt; > I'd prefer to avoid a list... how about: > > "This site provides status only for services managed directly by Fedora > Infrastructure" That's short and simple and drives the point home clear enough, I think.