#metabrainz

zas: good morning. a last small cleanup before we tag the release today: https://github.com/metabrainz/picard/pull/2108

outsidecontext: moooin, coffee (first) and I'll have a look

top, coffee is important :D

outsidecontext: I think you can merge and tag

yvanzo: i started a full reindex with that branch yesterday night, should complete soon. lets merge after that?

zas: will do. we do a rc2, right? And final a week later?

+1

yvanzo: went fine, i'll update the mbdata commit messages and if needed the commit sha in sir and then fine to merge.

moooin!

lucifer: did you see the LB support@ email about "Mismatched redirect URI" when logging into LB? Possibly for a new accounts?

mayhem: yes i did. investigating.

logins work, I checked that.

checking sentry logs now. nothing obvious

outsidecontext: there are some issues with builds -> https://github.com/metabrainz/picard/runs/64023...

Happy Birthday reosarevok 🎊

zas: fixed it

mayhem: i just created a new account and it went fine.

lucifer: I think I will respond and ask for screenshots.

mayhem: thanks! was going to do the same.

happy birthday reosarevok, 🎉🎉!

zas: morning!

mayhem: hey

can you please check the support@ emails -- there is a suspected IP blocked address that might be causing us issues. can you please check to see if that is the case?

lol, hey!

Thanks!

which support emails?

ah, they didn't copy to support@ on their response. sending now.

ah thanks

zas: https://github.com/metabrainz/picard/releases/t...

mayhem: the email doesn't provide exact IPs, but it doesn't seem any is blocked in the /24 blocks he provided. Also the error says "Mismatched redirect URI" which seems to indicate an OAuth issue. Is there something indicating it may be related to a blocked IP?

not from our side, no, but the person suggested it might be so. lucifer and I have probed this and we can login just fine and find no errors in the logs for LB.

mayhem: first screenshot, they are using api.listenbrainz.org instead of listenbrainz.org

outsidecontext: did you see the error in https://github.com/metabrainz/picard/runs/64026... ?

we should disable webpages except api there to avoid such issues i guess.

aahhh, well spottted.

yes, we should do that, for sure.

I'll respond.

thanks!

zas: hi! can you please configure gateways to throw a error when someone tries to access api.listenbrainz.org on a url not prefixed with a number. basically opposite of this check: https://github.com/metabrainz/openresty-gateway...

this check is on listenbrainz.org to prevent access of api using it. conversely we want to prevent access of normal webpages on api.listenbrainz.org .

O’Moin

lucifer: yep, that was it.

ah nice.

lucifer: only for api.lb, right? (prod/beta/?)

zas: api.lb, beta-api.lb, test-api.lb

yvanzo: sir pr updated with new commit sha of mbdata (with feedback from your review applied). fine to merge now

moin yvanzo

https://twitter.com/qDot/status/152445099506482... heh

[listenbrainz-server] 14amCap1712 merged pull request #1987 (03master…timeduration-fix): Removing extra Z from missing MB data API and view https://github.com/metabrainz/listenbrainz-serv...

zas: don't know why it failed, retrying this uploaded the file to pypi successfuly

hullo

I kind of want to go and buy an ipod now

zas: website I sent a PR. are the files already copied to FTP?

well, it looks like Napster will beat us to the web3 fun... whatever that might be.

Oh no

I can't help but notice that this Napster article isn't an april's fool

(april fools?)

it will be april fools for them when they realize that they pivoted towards pure bullshit.

Lollll

Let's go straight to web8

not sure if that was in response to the ^ napster article, or just coincidence

yvanzo: so, mbdata/sir hopefully ready now?

!m lucifer

You're doing good work, lucifer!

lucifer: Which kind of database did you use for the full reindex ?

nice

yvanzo: the musicbrainz-docker db running on wolf currently

fwiw, the musicbrainz-docker on wolf is currently still setup that way for further testing if needed. the local sir copy is in /home/musicbrainz/sir

lucifer: So it's running with a schema 26 MB database?

yes

lucifer: I opened a PR to your mbdata fork :)

lucifer: Is https://github.com/metabrainz/sir/pull/130 ready for merge?

yvanzo: yup.

[sir] 14yvanzo merged pull request #130 (03master…schema-change-2022-q2): Update SIR for schema change 2022 q2 https://github.com/metabrainz/sir/pull/130

yvanzo: i can try to setup a schema version 27 db and try an index again if its helpful

We can try migrating test.mb.o

cool, sounds good

I'd like to try that now, yes

I think it's time

yvanzo: will you run the migration?

I can try too if you prefer, but happy to let you do it

reosarevok, lucifer: Just have to make a few changes to sir's SQL scripts beforehand.

Ok :)

outsidecontext: yes, files are on FTP

alastairp: I've updated the failing tests in my BU PR's. Please have a look. If it looks good, I'll open another PR to modify get_release_groups_for_label in a similar way.

Ansh: I saw the updates, thanks. they look really good so far, I was in fact just doing some final testing on them now, I'll let you know soon

Ok :)

mayhem: zas has updated gateways with config to block the pages on api sites. i tested it works fine. want to double check?

!m zas

You're doing good work, zas!

I'll need to update some checks though

oh great

oh the tg alert!

on it

how does this currently work? both api.lb.org and lb.org on the gateways point to the same instance, right?

yes afaik.

ah, just found the openresty commit

perfect

yeah, I was just wondering what we would use to identify if a request was API, version number looks good

do we have the opposite in place too? (https://listenbrainz.org/1/whatever) ?