07:02 <MrCooper> jenatali eric_engestrom: that job merely copied the tag from the main registry using skopeo, didn't rebuild it from scratch

07:03 <MrCooper> surprising that took over 20 minutes though

09:17 <emersion> anyone knows what's up with these google meet emails from Mario?

09:17 <emersion> i've sent an email asking them to stop

09:19 <daniels> emersion: they're just Google Calendar - I suspect he's been creating a calendar event from the email which is then helpfully inviting everyone CCed

09:20 <emersion> the participants are dozens of mailing lists :/

09:20 <daniels> indeed

09:20 <daniels> 'helpfully'

11:03 <MrCooper> he did cancel the event again, not sure there's anything else he can do now

11:30 <__tim> would it be possible to add the group gstreamer/meson-ports also to the ci-ok group please? :)

12:34 <emersion> they know about the issue, it sounds like Google somehow auto-added the participants and sent out emails even if "notify participants" was off

12:34 <emersion> i thought it happened again this morning but maybe it was just some emails getting delayed and getting to me via yet another ML

13:18 <MrCooper> yeah, that was most likely it

17:24 <bentiss> __tim: done

17:24 <__tim> thanks!

17:24 <bentiss> nw

17:26 <daniels> bentiss: hm, I'm seeing a fair few failures like https://gitlab.freedesktop.org/wayland/weston/-/jobs/40268861 - weirdly all on equinix-arm-7 for some reason - anyway the fact it fails after printing that first message suggests that maybe curl's getting HTTP errors back and we probably need to add retries to it?

18:07 <eric_engestrom> daniels: over the last few weeks there's been several docker daemon failures on that specific machine, bentiss had to restart at least twice that I know of

18:08 <eric_engestrom> so my guess is that it's not a network failure but something's wrong with docker

18:12 <daniels> eric_engestrom: it's not docker - that script should never stop printing after 'Checking if user is allowed'

18:12 <daniels> it'll either go on to print that the user is allowed, or print through the next steps including checking the group and/or denying access with another larger print

18:13 <eric_engestrom> ah, you're right

18:13 <eric_engestrom> I was remembering the script wrong

18:14 <eric_engestrom> (that greet_and_exit didn't exist when I reviewed the script, and I haven't been paying enough attention since)