#metabrainz

ok

then vagrant destroy, vagrant up, vagrant ssh, sudo pip install -r requirements.txt

alright. vagrant up is running. I'll let you know when I catch up

you don't need to run the last one

when `vagrant up` is done running, paste its output

like paste all the " ==> default " lines? or the stuff after?

everything

its too big to pastebin here :\

don't worry

thanks https://gist.github.com/caroline-gschwend/d0ee4...

hmm, that error

yeah

`vagrant ssh`

`sudo su;cd /vagrant`

ok

those worked fine

run these two commands https://github.com/metabrainz/metabrainz.org/bl...

"python: can't open file 'manage.py': [Errno 2] No such file or directory"

ls /vagrant

"root@metabrainz:/home/vagrant#"

does it list files from project root?

cd /vagrant

from first command^

from "python manage.py create_db"

weird

pip install -r requirements.txt

and I got the same error

is this where I'm supposed to be "root@metabrainz:/vagrant#"

oh, remove line with "requests" package from requirements.txt

yes, this is the right directory

Ok install requirements worked this time

run the manage.py commands?

try create_db again

yes

erm

that's "ok"

Ok . RUn the next one?

yep

Ok that did nothing (worked I guess)

python run.py

YES!!

zas: so wait, you made a change that should make things worse and it made things better?

Is there a reason not to just keep doing the fast thing? (as in, were we doing that for performance only or is there another reason)

zas: huh, well that's pretty surprising

Gentlecat, thank you thank you thank you

you are welcome

need to fix that issue though

reosarevok: eariler we were getting some DNS errors trying to resolve totoro.mb, I think, and the DNS cache seemed to have solved it

no idea if those are gonna come back now, hopefully not

Gentlecat: I've commented your recent comments.

yes, I get notifications about replies

Gentlecat, am I supposed to be working in metabrainz/metabrainz.org or metabrainz/metabrainz-server?

metabrainz-server is old version of the website

ignore it

so metabrainz/metabrainz.org

Ok. Freso, linked to there on the task page. Thanks

stanislas: also to https://github.com/metabrainz/critiquebrainz/pu...

see http://docs.python-guide.org/en/latest/writing/...

Python is not C

Gentlecat: I am used to coding in C, maybe that's the problem :)

CJ_: the xml responses get converted to json by running them through https://github.com/metabrainz/mb-solrquerywrite... which is an https://wiki.eclipse.org/EclipseLink/Examples/M... mapping file

might work on http://tickets.musicbrainz.org/browse/MBS-8736 today to make it easier to test the search server

bitmap, if you didn't see earlier http://chatlogs.metabrainz.org/brainzbot/metabr...

ah, there should be a pic in the summit blog you can use

Ok

JesseW: ruaok had some questions about the archiveteam thing, maybe check logs for today :)

alastairp: did we decide what to do with JAMS before? http://www.terasoft.com.tw/conf/ismir2014/proce...

I remember we discussed it

bitmap: I wonder what to do with the cache advisory locks when DB_READ_ONLY is set.

AFAIU, the locks are handled in memory, no WAL is written.

So it might be OK to have them running.

hmm, I'm not sure

OTOH, while in read-only mode, there should be no harm in caching things without locking.

right

I suppose it'd be harmless and possibly safer to check DB_READ_ONLY before acquiring any locks

(I also thought about possible future replicated servers. If we have hot standby servers that are used for web display, they can use the cache, but not cache anything because they can’t get locks.)

(Where “they” confusingly refers to the web frontends using the replica. ;-) )

Mineo, Why did you do that instead of just using the regular json flag?

"regular json flag"?

it's what the current search server uses to generate the json and moves some elements around etc.

Mineo, wt=json instead of wt=mbjson

I am wondering if the writer can be removed and regular json can be used instead.

chirlu`: I guess that if MBS is configured to connect to the master DB for certain things, we could keep the locks there?

For example: http://solr-test.musicbrainz.org:8983/solr/anno...

CJ_: wt=json is completely different, it's just a key-value mapping of field names to their values for each document

bitmap: Yes, those requests that use the master database could cache items (for use by all requests).

Mineo, What is the expected mbjson result, then?

It’s probably enough since the often-used items will hit the master database at some point, too.

CJ_: the same one as http://search.musicbrainz.org/ws/2/annotation/?... (admittedly, it's just a flat mapping for annotations as well in this case :) )

Gentlecat, are you around? I have some questions about how the metabrainz navbar works

Gentlecat: Do you have some alternative to controlling whitespace the way I am doing it currently ?

yes

if not, someone else might be able to answer. so feel free to just ask

Mineo, Do you know the difference between json and jsonnew?

stanislas: what do you mean?

no

Gentlecat: I would like to use {{sth}} without whitespace showing around

it can become hard to read in big templates

Gentlecat, ok so each of the buttons, "About", "Sponsors"..have this as the href "{{ url_for('index.about') }}" (or similar). But, I've added a team.html file to the index folder and doing { url_for('index.team') }} in the navbar does not work. What is my problem?

'index.about', for example, is a reference to a specific view

Gentlecat: I guess it can, but ... how to do replace the hyphens ?

right the, about.html file

no, not the template file

ahh I see

there's that one function in between

CJ_: http://tickets.musicbrainz.org/browse/MBS-5671

So i need to make one for team

probably

stanislas: oh, I see what you mean

chirlu`: so "none"?

I thought you were asking about {{sth}} vs {{ sth }}

Gentlecat: hahaha, no :)

It's working, thanks

reosarevok: No, unless the planned merge was already done. Otherwise, they are different (mostly regarding the entity lists, I think).

Oh, true, it just say we don't use it, not that it's been merged

ok :)

Gentlecat: Maybe my english is not very clear sometimes.

can you link part where it shows up?

Perhaps it has been merged, but I don’t know.

sure

chirlu`, Understood. Thank you. How was the new json format decided on?

Uh, no idea, actually. I assume it was approved in one of the regular meetings.

Gentlecat: I am pretty sure it showed here https://github.com/stasszczesniak/critiquebrain...

before adding the hyphens

chirlu`, I am just wondering if it may be easier to just move to regular solr json instead.

CJ_: If it changes the output format, no.

leave it if it's significant then

chirlu`, It is a different format than the jsonnew. Has the jsonnew migration been completed?

Yes. The WS only delivers jsonnew.

Hmm.

I still need pictures from zas, ruaok, and alastairp. You can email them to me at caroline.gschwend@gmail.com

LordSputnik and Leftmost, I got your pictures from the summit blog post, if you have other pictures that you would rather me use, email them to me at the above address

Who is Wieland Hoffman?

me

Mineo, Oh, okay. :)