14:00 PM
bitmap
killed the rest of the daily script, we can restart the subscriptions process once we're done
2020-05-20 14105, 2020
14:01 PM
alastairp
jmp_music: essentia shouldn't be needed.
2020-05-20 14128, 2020
14:01 PM
alastairp
the current way that process works is that users download a binary which generates .json files
2020-05-20 14131, 2020
14:01 PM
bitmap
ruaok: yvanzo: ready to proceed?
2020-05-20 14145, 2020
14:01 PM
ruaok
ready. shall we do some tweeting?
2020-05-20 14150, 2020
14:01 PM
ruaok
(ready to do that)
2020-05-20 14153, 2020
14:01 PM
bitmap
sure
2020-05-20 14118, 2020
14:02 PM
alastairp
the training process and doing data->model->prediction act only on these .json files
2020-05-20 14125, 2020
14:02 PM
jmp_music
So I 'll handle only the .json files (low level) and start the processing
2020-05-20 14139, 2020
14:02 PM
jmp_music
Ok that's clear too
2020-05-20 14141, 2020
14:02 PM
alastairp
we only have essentia in the webserver to build the `streaming_extractor_music_svm` tool, which does predictions
2020-05-20 14157, 2020
14:02 PM
alastairp
with your tool, we will be able to remove essentia from the acousticbrainz-server repository too
2020-05-20 14109, 2020
14:03 PM
alastairp
because we can replace it with sklearn inference
2020-05-20 14134, 2020
14:03 PM
ruaok
tweeted. let's give it a minute.
2020-05-20 14114, 2020
14:04 PM
ruaok
are we going to take the sites down via the gateway?
2020-05-20 14114, 2020
14:04 PM
Mr_Monkey
Same. ready to take BB down
2020-05-20 14102, 2020
14:05 PM
jmp_music
Ok! Now I understand. I was a little bit confused because i was checking the gaia's code and it uses and other kinds of files (e.g. .sig)
2020-05-20 14117, 2020
14:05 PM
bitmap fixed an issue with musicbrainz_ro permissions on the genre table making MB ISE
2020-05-20 14136, 2020
14:05 PM
bitmap
ruaok: we could do that, git2consul seems to work now
2020-05-20 14148, 2020
14:05 PM
ruaok
up to you.
2020-05-20 14155, 2020
14:05 PM
ruaok
I am ready in any case.
2020-05-20 14107, 2020
14:06 PM
Lotheric joined the channel
2020-05-20 14113, 2020
14:06 PM
Mr_Monkey
Seemed to work without it yesterday no?
2020-05-20 14126, 2020
14:06 PM
alastairp
right. gaia has existed since before json was a standard. .sig is just a yaml file
2020-05-20 14137, 2020
14:06 PM
ruaok
we took the sites down on monday.
2020-05-20 14139, 2020
14:06 PM
alastairp
internally, gaia converts json to yaml before loading it
2020-05-20 14149, 2020
14:06 PM
ruaok
bitmap: let us know when you want to start.
2020-05-20 14152, 2020
14:06 PM
Mr_Monkey
Right, the other yesterday.
2020-05-20 14154, 2020
14:06 PM
Lotheric__ has quit
2020-05-20 14156, 2020
14:06 PM
alastairp
(we won't have to do this)
2020-05-20 14159, 2020
14:06 PM
jmp_music
alastairp: OK. That's all for now! Thanks a lot for clearing up the mess in my head :) .
2020-05-20 14110, 2020
14:07 PM
bitmap
ruaok: ok, I've starting putting MB in read-only already
2020-05-20 14121, 2020
14:07 PM
alastairp
jmp_music: great, no problem. Feel free to ask these kinds of questions at any time
2020-05-20 14125, 2020
14:07 PM
ruaok
ok, I'll take CB, LB and MeB down.
2020-05-20 14125, 2020
14:07 PM
bitmap
seems to be live already
2020-05-20 14150, 2020
14:07 PM
Mr_Monkey
BB is down
2020-05-20 14157, 2020
14:07 PM
ruaok
LB down
2020-05-20 14116, 2020
14:08 PM
ruaok
CB down
2020-05-20 14133, 2020
14:08 PM
jmp_music
alastairp: So you think it is a good idea to handle directly the json file and not transform it to yaml.
2020-05-20 14134, 2020
14:08 PM
bitmap
cool
2020-05-20 14137, 2020
14:08 PM
ruaok
MeB down.
2020-05-20 14151, 2020
14:08 PM
bitmap
does MSB need to come down?
2020-05-20 14158, 2020
14:08 PM
ruaok
is down
2020-05-20 14100, 2020
14:09 PM
ruaok
sorry
2020-05-20 14125, 2020
14:09 PM
alastairp
jmp_music: yes, absolutely
2020-05-20 14135, 2020
14:09 PM
jmp_music
alastairp: just loading it to a Pandas dataframe and continuing the post-process there.
2020-05-20 14146, 2020
14:09 PM
bitmap
aight, putting pg into maintenance mode in consul
2020-05-20 14157, 2020
14:10 PM
bitmap
I'll wait a min for that to propagate
2020-05-20 14122, 2020
14:11 PM
bitmap
queries are dropping already
2020-05-20 14110, 2020
14:12 PM
bitmap
ok let's see if it stops
2020-05-20 14125, 2020
14:12 PM
bitmap
pg stopped
2020-05-20 14141, 2020
14:12 PM
bitmap
restarting container
2020-05-20 14113, 2020
14:13 PM
bitmap checks ShmSize on container
2020-05-20 14116, 2020
14:13 PM
bitmap
looks to be set
2020-05-20 14151, 2020
14:13 PM
yvanzo
docker container inspect foo | jq '.[0].HostConfig.ShmSize'
2020-05-20 14156, 2020
14:13 PM
ruaok
restart?
2020-05-20 14108, 2020
14:14 PM
bitmap
pg back up, pg bouncer too
2020-05-20 14131, 2020
14:14 PM
bitmap
consul service out of maint mode
2020-05-20 14106, 2020
14:15 PM
ruaok
MeB up.
2020-05-20 14118, 2020
14:15 PM
ruaok
CB up
2020-05-20 14132, 2020
14:15 PM
ruaok
LB
2020-05-20 14134, 2020
14:15 PM
ruaok
up
2020-05-20 14154, 2020
14:15 PM
bitmap
pushed DB_READ_ONLY -> 0 for MB, waiting for that to propagate
2020-05-20 14156, 2020
14:15 PM
Mr_Monkey
BB back up
2020-05-20 14105, 2020
14:16 PM
ruaok
MSB up.
2020-05-20 14115, 2020
14:16 PM
jmp_music has quit
2020-05-20 14105, 2020
14:17 PM
bitmap
looks good
2020-05-20 14123, 2020
14:17 PM
bitmap removes banner
2020-05-20 14120, 2020
14:19 PM
bitmap
thanks ruaok yvanzo Mr_Monkey :)
2020-05-20 14126, 2020
14:19 PM
ruaok
np
2020-05-20 14153, 2020
14:19 PM
Mr_Monkey
š
2020-05-20 14100, 2020
14:20 PM
yvanzo
thank you!
2020-05-20 14145, 2020
14:20 PM
bitmap
I'll watch the logs for pgbouncer errors
2020-05-20 14142, 2020
14:24 PM
bitmap
running ProcessSubscriptions again too to get the rest of the emails out
2020-05-20 14103, 2020
14:26 PM
bitmap
I'll do the stats and reports too
2020-05-20 14104, 2020
14:27 PM
bitmap
this might send out new emails for people who already got them though. should've stopped this before it started :\
2020-05-20 14139, 2020
14:27 PM
bitmap
if they're subscriptions changed again in the past 30 mins
2020-05-20 14142, 2020
14:27 PM
bitmap
their*
2020-05-20 14102, 2020
14:28 PM
yvanzo
better sending it two times than zero
2020-05-20 14113, 2020
14:28 PM
bitmap nods
2020-05-20 14131, 2020
14:31 PM
v6lur has quit
2020-05-20 14125, 2020
14:35 PM
bitmap
2020-05-20 14136, 2020
14:35 PM
Mr_Monkey
Nice.
2020-05-20 14149, 2020
14:51 PM
bitmap
2020-05-20 14157, 2020
14:51 PM
bitmap
so 1GB is not enough to run some of these reports
2020-05-20 14113, 2020
14:52 PM
BrainzGit
2020-05-20 14114, 2020
14:52 PM
BrainzBot
2020-05-20 14122, 2020
14:55 PM
reosarevok
bitmap: how much did we give it before? I'm kind of amazed all these are failing now but not earlier :/
2020-05-20 14138, 2020
14:56 PM
bitmap
I think this is related to the new parallel query support in 9.6+, which use more shared memory
2020-05-20 14129, 2020
14:57 PM
bitmap
it was just using the docker default before, 64M, but I guess it didn't need much in 9.5
2020-05-20 14105, 2020
14:58 PM
reosarevok
That's kind of insane
2020-05-20 14135, 2020
14:58 PM
reosarevok
So is the idea that it's running a lot more stuff in parallel and as such it should be done faster if it manages to run them at all?
2020-05-20 14110, 2020
15:00 PM
BrainzGit
2020-05-20 14110, 2020
15:00 PM
BrainzGit
2020-05-20 14111, 2020
15:00 PM
BrainzBot
2020-05-20 14132, 2020
15:00 PM
bitmap
right
2020-05-20 14155, 2020
15:00 PM
bitmap
it will try to execute separate parts of the plan for a single query in parallel
2020-05-20 14113, 2020
15:01 PM
reosarevok
But there's no way to tell it "hey, maybe don't run *this* in parallel" if we know it won't work?
2020-05-20 14124, 2020
15:02 PM
bitmap
not that I know of
2020-05-20 14150, 2020
15:02 PM
bitmap
it's probably better if we just give it the memory it needs
2020-05-20 14152, 2020
15:02 PM
shivam-kapila
ruaok: Please take a look at the JSON structure I defined in this PR. The API will work on this standard. I have added a `submission_type` to validate the submission as type feedback. Please let me know if its not required.
2020-05-20 14140, 2020
15:03 PM
bitmap
these reports scan / join entire tables though so they are kinda extreme cases
2020-05-20 14151, 2020
15:03 PM
reosarevok
I mean, as long as we can give it enough, sure I guess :)
2020-05-20 14120, 2020
15:04 PM
reosarevok
Just worried at some point we'll get a query that would have worked fine (if slowly) earlier and doesn't work at all now
2020-05-20 14135, 2020
15:04 PM
BrainzGit
2020-05-20 14136, 2020
15:04 PM
BrainzBot
2020-05-20 14143, 2020
15:06 PM
BrainzGit
2020-05-20 14103, 2020
15:07 PM
yvanzo
2020-05-20 14105, 2020
15:09 PM
ruaok
shivam-kapila: the plural of `feedback` is `feedback`. stupid english.
2020-05-20 14119, 2020
15:09 PM
ruaok
`feedbacks` is not a valid english word.
2020-05-20 14156, 2020
15:09 PM
shivam-kapila
Sorry š
2020-05-20 14108, 2020
15:10 PM
ruaok
#stopapologizing
2020-05-20 14111, 2020
15:10 PM
ruaok
english is stupid.
2020-05-20 14122, 2020
15:10 PM
ruaok
while of course
2020-05-20 14138, 2020
15:10 PM
ruaok
#stopapologising is also valid
2020-05-20 14145, 2020
15:10 PM
ruaok
#englishisreallystupid
2020-05-20 14120, 2020
15:11 PM
ruaok
more seriously... given that we have a unique endpoint, I don't think we need to have the `payload` indirection.
2020-05-20 14154, 2020
15:11 PM
ruaok
submission_type is also not needed, since it uses its own endpoint.
2020-05-20 14155, 2020
15:12 PM
ruaok
iliekcomputers: do you have thoughts on the JSON for the feedback? in particular, I am not certain that POSTing a JSON document that could easily be done with two arguments is wise.
2020-05-20 14149, 2020
15:13 PM
ruaok
POST to /api/feedback/user?msid=[]&score=-1 with an empty body would be easier to do for a client.
2020-05-20 14109, 2020
15:14 PM
ruaok
well /api/feedback/user/dumbass?msid=[]&score=-1
2020-05-20 14101, 2020
15:16 PM
shivam-kapila
I felt so. Such a simple API can be done with POST request params
2020-05-20 14123, 2020
15:16 PM
iliekcomputers
ruaok: I think it'd probably be easier to give a json for post requests (specially with python `requests`), but i don't have any specific preferences either way
2020-05-20 14102, 2020
15:17 PM
ruaok
I'm not convinced either way, TBH.
2020-05-20 14119, 2020
15:17 PM
ruaok
I do, however, feel that we don't need payload and subission type in fetching the feedback
2020-05-20 14128, 2020
15:17 PM
ruaok
2020-05-20 14122, 2020
15:24 PM
shivam-kapila
Added payload just to be flexible in case we extend the API for bulk imports. submission_type will go for sure
2020-05-20 14122, 2020
15:24 PM
iliekcomputers
is the score variable?
2020-05-20 14140, 2020
15:24 PM
iliekcomputers
if not can't we just have two endpoints /like /dislike or something
2020-05-20 14158, 2020
15:24 PM
ruaok
score can be +1 or -1
2020-05-20 14123, 2020
15:26 PM
iliekcomputers
should we have a constraint on that in the db as well?
2020-05-20 14148, 2020
15:26 PM
ruaok
seems overkill, but sure.
2020-05-20 14104, 2020
15:27 PM
shivam-kapila
I was thinking to check it API level to prevent unnecessary DB access
2020-05-20 14131, 2020
15:27 PM
ruaok
I think that is sufficient.
2020-05-20 14107, 2020
15:29 PM
shivam-kapila
> if not can't we just have two endpoints /like /dislike or something
2020-05-20 14107, 2020
15:29 PM
shivam-kapila
score an go with +1,-1,0. +1 for love, -1 for hate, 0 in cases when user unmarks the track as loved/hated.
2020-05-20 14134, 2020
15:29 PM
shivam-kapila
+1/-1 ==> store in DB
2020-05-20 14134, 2020
15:29 PM
shivam-kapila
0 ==> Remove entry from DB
2020-05-20 14159, 2020
15:29 PM
iliekcomputers
i would probably have built 3 endpoints for them, unless we're planning on extending them.
2020-05-20 14130, 2020
15:30 PM
ruaok
also seems overkill to me.
2020-05-20 14158, 2020
15:31 PM
ruaok
do you feel strongly about having 3 endpoints?
2020-05-20 14118, 2020
15:32 PM
iliekcomputers
nah, not really
2020-05-20 14127, 2020
15:33 PM
iliekcomputers
i'd say keep the json though if it's a post request.
2020-05-20 14137, 2020
15:33 PM
ruaok
ok, fair.
2020-05-20 14149, 2020
15:33 PM
iliekcomputers
most request libraries support it well out of the gate.
2020-05-20 14151, 2020
15:33 PM
ruaok
one endpoint, POST json, ditch the payload/type.
2020-05-20 14102, 2020
15:34 PM
ruaok
that ok?
2020-05-20 14107, 2020
15:34 PM
iliekcomputers
sounds good to me
2020-05-20 14111, 2020
15:34 PM
ruaok
shivam-kapila: you?
2020-05-20 14120, 2020
15:34 PM
shivam-kapila
Absolutely
