Meeting started at 2025-01-10 17:02:07 UTC

The Meeting name is 'Fedora ELN SIG'

The Meeting Name is now ELN

!hi

Stephen Gallagher (sgallagh) - he / him / his

!hi

Davide Cavalca (dcavalca) - he / him / his

just pinged some more people

Michel is at a doctor appointment and will be late

ok well I guess we'll start, not sure what we can do with just us

!hi

Neil Hanlon (neil) - he / him / his

this is not a good time for meetings in general, and definitely not for me personally

I would like to see if we could find another time for this, and to make it weekly

to that end, if you haven't already, if you could please fill in https://whenisgood.net/td44ayx

any questions about this?

Do you want to make the results public?

I have voted

(I think I responded, but I don't remember for certain)

I don't see a response from you

moving along then

🥳

Johnny is a long-time CentOS contributor (understatement)

oh sweet, strong +1 from me

and will be helping with packages

Understatement indeed. Johnny and CentOS were practically synonyms for quite a few years

doesn't seem like he could make it today, but I hope that a new meeting time will also allow him to attend more regularly.

my goal here is to really use this as our issue tracker, at least until forgejo comes

!hi

Michel Lind (salimma) - he / him / his

at that point, hopefully there will be some effort to consolidate Fedora projects in one place, and we can re-assess then

going forward, at least part of each meeting I plan to spend on reviewing tickets and such

anything to discuss about tickets in general?

(please don't make me do all the talking here 😁

don't everyone speak at once...

nothing from me, between holidays and perf reviews haven't had a ton of time to work on eln lately, but hopefully that'll change soon

Sorry I had a contractor show up early...

np

I'm sitting in a car dealership while my car is getting fixed, so my attention is split.

I had a PR that I have not checked but it might be merged already

Running errands before a family trip tomorrow, sorry

ok then, let's start going through tickets, I'll start with the oldest, we can get back to that

Oh it's still open https://github.com/minimization/content-resolver-input/pull/1342

see my response there

oh yeah we still need to sort that one out

yes that would be good and ASAP, the Meta workload is broken because of that

I'll take a look today after this crop of meetings

starting with the oldest this time, please speak up at any time

The status page represented there has long ceased operation and the more updated one (built into EBS) hasn't been syncing to the public location for a while

(The internal shell host it was relying on got decommissioned)

is there any reason why EBS needs to be internal?

One possible reason, yes.

We don't have proper authn/authz functionality, so anyone could send a message to the rebuild-trigger endpoint

right

It's not known to be harmful, other than submitting some useless builds that will get canceled

But it's something we should probably protect in some way before we made the service open to the general internet

oh yeah definitely

would you mind creating a ticket for that with any relevant background?

(If we were hosting this atop a traditional web server like httpd, I could actually resolve that at the server config rather than the app code, but we're using OpenShift)

Yeah, I can do that

until then, we still should see if it's possible to set up a sync from EBS?

Yeah, I just need to find a shell host somewhere and rsync it out to my fedorapeople.org space again

maybe your space isn't the best place anymore though?

No, but it's where other stuff is linked to for now

anything else on that?

nope

I need to run; they're finishing up with my car

ok thanks!

supposedly we still don't have AWS images, apparently waiting on David Duncan to do that

I've tried pinging in the ticket, so far no response

moving on then

we've add ELN scratch builds for rawhide PRs for a while, but there is still a request to run dist-git tests too

not exactly sure what that would entail

as we're running out of time, one more I want to cover today

one issue that we're having with mass rebuilds is that it creates too many simultaneous updates and overwhelms openQA

but after that we do want to get that fixed, the only catch being that bodhi may not be able to handle an update with thousands of packages, so we'll need two code paths

anything on that?

anything else to discuss today?

in that case...

thank you all for coming, I look forward to working more with you all going forward, and hopefully we can find a better meeting time for everyone