it makes sense, but we need to manage the user's expectations. because the canonical data is a derived data set that isn't as stable as the other data returned by the API.
Ok, not something we'll probably work on super fast anyway, but good to know it might make sense
monkey: LB-1647 got marked as fixed but is still open, how do you usually cycle through states for LB? :) (trying to understand it in general, maybe it's correct)
reosarevok, bitmap: Starting the release processā¦
mara42 has quit
mara42 joined the channel
s-crypt has quit
s-crypt joined the channel
zas[m]
atj: new server finally arrived, I did first steps (DNS, etc.) and open a PR for adding it to ansible. First bootstrap check shows an issue with missing binutils (24.04), so I guess you can take it from there. Current disk setup is software RAID1 + ext4. Not sure we need to bother with zfs on this one, but that's as you wish.
jasje[m]
lucifer: is it possible to get last few failed api calls from my account to lb server?
atj[m] joined the channel
atj[m]
zas: what was the error with missing binutils?
jasje[m]
jasje[m]: Ive had few crashes on release versions of the app, cannot export logs as it was not my device. Errors were related to feed primarily.
(that link doesn't work for me but I can check the actual post in the editor)
"Both the Cover Art Archive and the Event Art Archive were hosted by the Internet Archive" are, hopefully :)
yvanzo[m]
It isnāt available at the moment.
reosarevok[m]
Well sure, but they're still hosted there :) Unless they lost them š
What about something like
BrainzGit
[musicbrainz-server] 14JadedBlueEyes opened pull request #3390 (03masterā¦MBS-13719-1): MBS-13719: Enable sending contact emails through new service https://github.com/metabrainz/musicbrainz-serve...
reosarevok[m]
As it has been mentioned in the banner for the MusicBrainz website all of last week, images for events and releases are currently unavailable due to the Internet Archive (IA) needing to recover from a DDoS attack. See the IAās Mastodon page for follow-up. Both the Cover Art Archive and the Event Art Archive are hosted by the IA, and since they are doing their best to restore services safely, they are not rushing and it might
still take several days for the images to come back. Therefore, todayās server update temporarily disables editing images, since it's impossible to vote properly on edits to images the editors cannot see.
In any case I'd add something like that last clarification, because while it should be obvious why people shouldn't edit images... it also should be obvious we don't have control on when they'll be back and people keep asking, so let's be super clear :D
Should we also mention we will make the editing available as soon as possible once the services are back safely?
Should also be obvious but š¤·āāļø
yvanzo[m]
Itās mentioned it is temporary
I would just post it as-is.
reosarevok[m]
Yes, my point was more that I assume we'll hotfix the server to bring them back if needed, not make people wait two weeks
But that's not clear from the post :)
yvanzo[m]
"that will take as many days as necessary"
reosarevok[m]
Yes, of course, but that just says the downtime at the IA will take as long as needed. Once that stops the images will be back, but editing won't, for that we need to hotfix or wait for a new release. I expect we'll hotfix, but we could also wait for the release and make sure that it's consistently back I guess.
Just trying to understand what the expected plan is (and maybe let users know)
bitmap[m]
I'd +1 changing "were" to "are" in case people misinterpret that, but otherwise it looks good to go
yvanzo[m]
āuntil our submission access to the Internet Archive is restoredā?
bitmap: That was intentional to underline that these archives are not currently available.
"were provided"?
"were available from"?
I'll be back later onā¦
bitmap[m]
yeah, I get what you meant but I guess there is a chance it could be misinterpreted. "were available from" sounds good to me
Jade[m] joined the channel
Jade[m]
<BrainzGit> "[musicbrainz-server] JadedBlueEy..." <- Btw I've made this, I didn't see any feature flag system but let me know if I should add a toggle that turns back the old code
bitmap[m]
Jade[m]: thank you! we normally just add an option to DBDefs.pm for feature flags, though I think this is fine without one
mara42 has quit
mara42 joined the channel
pite joined the channel
mara42 has quit
mara42 joined the channel
yvanzo[m]
Back!
reosarevok, bitmap: I made a few changes and additions, does it look good now?
bitmap: Should we support disabling images editing through server configuration?
bitmap[m]
yvanzo: possibly yes, I was expecting we would just revert the code and never have to use it again, but that might be optimistic
Maxr1998_ has quit
Maxr1998 joined the channel
mara42 has quit
yvanzo[m]
bitmap: Does the maxima for workers/memory in seldaās PR seem appropriate (for an AX102) to you?
reosarevok[m]
<BANG>
Welcome to another MetaBrainz Monday Meeting!
I have one mailed in review this week, by outsidecontext:
'''
I'm on a business trip the entire week and hardly available. Last week not much did happen. I researched a bit about code signing certificates and did some code review.
mayhem, please ignore the SSL.com warnings about the certificate for now. Nothing bad will happen, we just need a new one for the next Picard release. I'll get back to you once I've returned from my trip and can concentrate on MB again.
'''
On my list for today: JadedBlueEyes, kellnerd, pranav, MonkeyPython, monkey, ansh, atj, akshaaatt, yellowhatpro, lucifer, derat, yvanzo, bitmap, theflash_, zas, rimskii, reosarevok, mayhem, jasje, ericd
