then I am fine to go to beta and then to improve the rest of the paths.
2017-05-30 15017, 2017
ruaok
the ingestion pipeline is what I really most concerned with.
2017-05-30 15052, 2017
alastairp
iliekcomputers: OK, that looks a bit more like what I expected!
2017-05-30 15000, 2017
alastairp
OK, so it's going in that direction, good
2017-05-30 15054, 2017
alastairp
although, that example is interesting too... I'm not sure if it makes sense to have `while True` in response to an API call, since we have no idea how long it might take for this to come back up
2017-05-30 15009, 2017
alastairp
we could instead immediately give a 5xx error to the client and ask them to try again
2017-05-30 15025, 2017
ruaok
ohh, that is a good point.
2017-05-30 15041, 2017
alastairp
right, this is what I'm saying :)
2017-05-30 15012, 2017
alastairp
if you need to wait for something to come up at startup -> dockerize; if something is not up when a user makes a request -> try once, but then fail fast
2017-05-30 15015, 2017
ruaok
got it. the semantic between a web requrest and a script need to be different.
2017-05-30 15021, 2017
ruaok
I hadn't made that distinction in my head.
2017-05-30 15027, 2017
alastairp
however, I understand that dockerize only waits for an open tcp port
2017-05-30 15044, 2017
alastairp
which means that the additional startup check at script/webserver startup is a good idea
2017-05-30 15013, 2017
alastairp
OK, my work here is done... I'll look at some more LB PRs tomorrow :)
2017-05-30 15037, 2017
ruaok
:)
2017-05-30 15038, 2017
ruaok
thanks!
2017-05-30 15042, 2017
alastairp
np
2017-05-30 15043, 2017
iliekcomputers
alastairp: the docker one on messybrainz too :)
2017-05-30 15053, 2017
alastairp
iliekcomputers: yeah, I assigned it to myself
2017-05-30 15012, 2017
alastairp
it's next in my inbox of unread messages... I'll get to it tomorrow
most important point here is "If the container is stopped and restarted, it uses the original volume that was created, even if the contents of the directory in the image have changed."
2017-05-30 15008, 2017
ruaok
yep.
2017-05-30 15012, 2017
alastairp
so if you compile css into a directory which is a volume
2017-05-30 15017, 2017
alastairp
then update the image
2017-05-30 15020, 2017
alastairp
and then run it again
2017-05-30 15023, 2017
ruaok
and since the CSS was compiled at build time, it should be fine.
2017-05-30 15036, 2017
ruaok
production does not use a volume for the -web container.
2017-05-30 15052, 2017
alastairp
I'm not 100% sure that the contents of that folder in the image will exist in the volume which already exists
2017-05-30 15059, 2017
alastairp
ah, ok. that's less of a problem then
2017-05-30 15033, 2017
ruaok
heck, let me try it and push it live. see what happens. :)
2017-05-30 15041, 2017
alastairp
what could go wrong
2017-05-30 15053, 2017
ruaok
sandbox goes boom. oh well.
2017-05-30 15003, 2017
alastairp
anyway, this took me a long time to work out... I ended up reading the docker source code to work out why stuff didn't make snse
2017-05-30 15004, 2017
ruaok
its a mess right now anyway.
2017-05-30 15015, 2017
ruaok
wow
2017-05-30 15046, 2017
iliekcomputers bookmarks gist in fear of having to read docker source code
2017-05-30 15009, 2017
ruaok did the same
2017-05-30 15025, 2017
lazka has quit
2017-05-30 15037, 2017
Freso
alastairp: I've posted meeting notes the same night before, but usually I've just been too tired in the head to do it. I hope I'll be able to do this more often going forward, now that we're finishing earlier. :)
2017-05-30 15036, 2017
yvanzo joined the channel
2017-05-30 15012, 2017
lazka joined the channel
2017-05-30 15058, 2017
D4RK-PH0ENiX has quit
2017-05-30 15053, 2017
gcilou joined the channel
2017-05-30 15014, 2017
lazka has quit
2017-05-30 15057, 2017
D4RK-PH0ENiX joined the channel
2017-05-30 15030, 2017
khan____ has quit
2017-05-30 15018, 2017
ruaok
"what could go wrong?"
2017-05-30 15028, 2017
ruaok
just a few rabbit holes, no big deal.
2017-05-30 15037, 2017
alastairp
welcome back from docker
2017-05-30 15041, 2017
ruaok
still not back. :(
2017-05-30 15028, 2017
CatQuest
thanks, yvanzo, (embarrassed about typoes.)
2017-05-30 15041, 2017
agentsim joined the channel
2017-05-30 15041, 2017
SkullTech has quit
2017-05-30 15041, 2017
drsaunders joined the channel
2017-05-30 15043, 2017
ruaok
zas: got a minute?
2017-05-30 15042, 2017
Major_Lurker joined the channel
2017-05-30 15028, 2017
MajorLurker has quit
2017-05-30 15040, 2017
drsaunders has quit
2017-05-30 15012, 2017
zas
ruaok: i have a meeting with Thelma's teachers in few minutes, what do you want to know?
2017-05-30 15036, 2017
ruaok
I need a second set of eyes to help me spot some docker problems.
2017-05-30 15039, 2017
ruaok
tomorrow, I guess.
2017-05-30 15057, 2017
zas
Or after Picard meeting tonight
2017-05-30 15009, 2017
ruaok
tomorrow. :)
2017-05-30 15013, 2017
Freso
Or, crud. I need to go read that document finally.
2017-05-30 15018, 2017
Freso
*Oh
2017-05-30 15025, 2017
zas
K ;)
2017-05-30 15043, 2017
chirlu has quit
2017-05-30 15038, 2017
TOPIC: MetaBrainz Community and Development channel | MusicBrainz non-development: #musicbrainz | Picard meeting (Tues, 18:00 UTC): Working document https://goo.gl/anENrc | MetaBrainz meeting agenda (now: 17 UTC!): reviews
[listenbrainz-server] mayhem closed pull request #183: LB-128: Install less and clean-css using npm so that compiling css works (master...clean-css) https://git.io/vHCA6
2017-05-30 15023, 2017
github has left the channel
2017-05-30 15027, 2017
Sophist_UK
Freso: Yes - however they are also often treated as a separate work in its own right - indeed probably better known that way. Which is exactly my point.
2017-05-30 15016, 2017
Freso
Sophist_UK: I'm not sure what point that is. :)
2017-05-30 15040, 2017
Sophist_UK
That we need semantic data held in MBS to say what the work parts are.
2017-05-30 15011, 2017
Freso
I'm not sure how that would do anything for this case.
2017-05-30 15037, 2017
Freso
Are "TRACK_WORK{,_LANGUAGE}" tags that are already being used in the wild?
2017-05-30 15040, 2017
Freso
I really wish I could leave comments in the document.
