#metabrainz

reosarevok I am very sorry, and I hope that you doing well

No worries, I'll let you know when we're done

Ok, not around :) Thanks for the reviews!

MohamedAli00949 (IRC): Shouldn’t be long.

So, schema change should happen in two months. We should pick whether on May 13 or May 20

yvanzo[m], bitmap, mayhem: ^

any reason not to choose the 13th?

We're planning a very small one AFAICT so probably not super important when to pick, but if we pick 13 I think we should announce super soon

I assumed it would be the 13th, which is fine by me

like by the 13th, yes.

So the only reason would be if we need more time to put the announcement together I guess

like tonight or tomorrow

If not, then 13 seems good to me (that was also my first guess but then I saw we've done 17 and the like before, so wanted to check with you)

yvanzo[m]: does 13 seem good?

the drafted blog post looked nearly ready to me (thanks reosarevok) so there shouldn't be any real delay needed to announce

Also a question to zas, would it be reasonable to announce Solr 9 upgrade at the same time?

(with the same release date in production, not just beta)

reosarevok (IRC): 13, sure.

yvanzo[m]: I am new here and i will chang it as soon as posible

Ok, I think we have a decision then :)

Thanks!

To zas & atj actually ^

yvanzo[m]: can that be talked post-meeting? (as in, is that this topic or do we close?)

yvanzo[m]: are we ready concerning SOLR for the 13th? It looks very short to me

13th May

Ah! ok then :l)

May should be OK

Ok, now we're done I think

So thanks everyone for coming!

It may be OK

See you in a week!

Is May an orangutan? If so, Ook.

</BANG>

Thanks everyone!

MohamedAli00949: now we're done :)

For BookBrainz issues, if that's what you're looking into, there's #bookbrainz too, but not sure if monkey[m] prefers helping people here, there, or it's the same to him

I answered in private and on #bookbrainz:moviebrainz.org

MohamedAli00949 (IRC): I just meant that the meeting shouldn’t be long for you to wait :) Now it is finished.

yvanzo[m], bitmap: available to talk schema change for a bit if you want :)

cool

I self-assigned MBS-13361, MBS-13337, MBS-12371 (the latter two because they seem related)

MBS-13361: Upgrade the required version of PostgreSQL to 16 https://tickets.metabrainz.org/browse/MBS-13361

MBS-13337: Query standby Postgres instances for background tasks when possible https://tickets.metabrainz.org/browse/MBS-13337

MBS-12371: Allow using musicbrainz_ro user in PROD_STANDBY database https://tickets.metabrainz.org/browse/MBS-12371

For the Perl upgrade, basically we will have to install it with perlbrew which can take a while, so I will go for making a pre-built base-image with Perl (and other requirements) installed.

I can push it a bit further to have pre-built images for Docker Compose too.

yvanzo: that would be nice

I don't expect the PG upgrade to require any other changes to the MBS images

(outside of the test images and upgrading psql, but we shall see)

there's one unassigned ticket, MBS-13360

MBS-13360: Detect which indexes need to be rebuilt for collations https://tickets.metabrainz.org/browse/MBS-13360

If we want to have a Solr 9 upgrade at the same time, it means to have it deployed in production the previous week at least, and in beta a few more weeks before that.

For that ticket, yvanzo seems to know exactly what to do so maybe he can take it and it'd be quick? :)

It should be mostly transparent to MBS, but will take some time to rebuild on mirrors.

omg im so sorry, I missed the meeting :___(

sure, if you feel there is enough time, we can prioritize any code review/testing period needed

rimskii[m]: it's ok :) if you post your review now, I'm sure aerozol[m] can at least add it to the forum notes :)

sure thank you!

re: MBS-13360, one thing that's unclear to me is where this detection will happen?

MBS-13360: Detect which indexes need to be rebuilt for collations https://tickets.metabrainz.org/browse/MBS-13360

I have no idea, I copied the information from IRC, not mine.

monkey[m]: Yeah, I have seen it Thanks again, and apologies for this noise

For this week, I actually havent done anything significant regarding coding... (full message at <https://matrix.moviebrainz.org/_matrix/media/v3...>)

bitmap: “some patches were merged for PG14 that improve version tracking”

ok, if it was copied from something I wrote, then I don't recall what the idea was :)

jasje (IRC): akshaaatt (IRC) aerozol can we please have a gmeet Tom or whenever u guys r free.. jasje (IRC) had some feedback reg my GSoC proposal and I thought it would be better if everyone involved could also join the meet. That would be awesome. Please let me know. Thanks!

aerozol[m]: Damn. That’s very sad. 😔

bitmap, reosarevok: Also I checked that we cannot stay with Perl 5.34 as it won’t have any further security updates starting from the 20th of May, this year.

Very sad! But Lucy is not in that region 😮‍💨

Glad Lucy is ok!

yvanzo: good timing then

aerozol[m]: so, please add rimskii[m]'s review above to your meeting notes when you make them, thanks <3

(but no rush with that)

thanks!

yvanzo: agreed re: solr, that if you are willing to prod us a bit to review stuff we can certainly do that, assuming you think it's not going to be very last minute :)

pranav[m]: count me in!

"Upgrade everything" seems like a very good alternative to "change a lot of sql schema stuff we need to still implement as well" anyway :D

pranav: jasje: akshaaatt (IRC): I can do this time tomorrow? But feedback really should be public in the first place imo. Let us know what the meetings about so me and akshaaatt aren't going in blind 👍

Or now 😊

Which reminds me, bitmap, yvanzo, if you find a bit of time to look at the event form react conversion, I can spend some time converting other forms as well after that one in between testing schema change things

bitmap, reosarevok: I don’t see anything related in https://www.postgresql.org/docs/release/14.0/

I think the relevant commit is https://git.postgresql.org/gitweb/?p=postgresql...

which may have landed in version 15

bitmap: Indeed: https://www.postgresql.org/docs/release/15.0/

bitmap, reosarevok: re Solr: Yes we can also have search improvements if we have beta early enough.

so, I do think it will be possible to query this information, but unclear where to perform this step: perhaps a script that is executed on container start

Hi reosarevok, I've been with the BookBrainz community since December 23rd. The Monday meeting looks super cool. Can I also be added in list for next week, please? =D

since the collation version of libc or icu should not change unless a new image is deployed

Tarun_0x0: sure! :) Added you now

yvanzo: since you will be quite busy with search the the Perl upgrade, I can look into it

So I'll call you on Monday

thanks !!

(I don't anticipate the PG upgrade to require as much work)

so, all tickets are assigned now

bitmap: If you aren’t too sure, we can also not announce it for now.

(And still do it if feasible in time.)

It isn’t a schema change ticket anyway.

right, let's strike it from the blog post

but we can keep it in the fix version & I'll look into it

Thanks!

hi @yvanzo. If you got time, I would like to discuss on IA integration

hi yellowhatpro, sure

akshaaatt: sorry if you got notified about this already, but there was a reply to you on the forums here, that doesn't tag you directly: https://community.metabrainz.org/t/golden-quote...

At first glance, It seems a bit similar to coverart-archive project, right?

We are creating a new service for fetching metadata from IA collections

Other tasks like full db dumbs and re-archiving links seem like a script running in background as a cron job

yvanzo, bitmap: so are we happy with where we stand? I think there's nothing for me other than any needed changes on the stuff I've already put up for PRs, right?

I think so. I just made some small adjustments to the blog draft

But I am not sure of this particular problem statement:

The Wayback Machine is able to take and render a snapshot of public webpages at any given time. It can be used for URL relationships added to the MusicBrainz database, and for references given in edit notes and annotations.

How do I proceed with this one?

We have to announce Solr 9 upgrade too in the same post.

right. I'm happy with the rest of it if you can add that :)

yellowhatpro: you can ask the Archive to save a specific page, so I expect that's what it means?

(or alternatively, you can use the Archive to check if a version of a now gone page is still there, but that wouldn't work unless you know it's gone to begin with...)

Wait, this isn't about an idea?

Now I'm extra confused :D

yeah but I was wondering how it is to be done with the rust binded project I was going on

Ooh, it's for the rust project, sorry

haha np reo

Well, one thing someone did in the past as a one-off was to ask the IA to save a snapshot of every URL we link to (which are a lot)

It’s the same in Rust, Go, Perl or .bf

Dunno if that's what yvanzo is thinking, but that seems like something that would be nice to do

yellowhatpro: Basically, extracting URLs from edit data and edit notes, for submitting to the Wayback Machine.

Oh I see

Make sure to skip those in edits submitted by users marked as spammers :D

Yes, there are some details like that

cool, so all I got to do is come up with some nice sql queries for filtering, and write rust scripts on top of these queries

There is no need to submit the same URL twice in the same day either.

There are probably some other things to check as well.

kind of what we are doing with reports

Other things like?

The SQL part is negligible for this task.

There is more work for connecting to the IA APIs, caching recently archived URLs…

Other things you will find.

Cool, I should play around with IA apis a bit then

since all the tasks seem to be running on background, should I consider it as a separate project, detached from the musicbrainz-server?

yes

aight, lesgo rustbrainz