Kia ora zas! I posted on the socials about our AI-crawler slowdowns, and again have had questions about why we don’t use Anubis, Cloudflare, etc (people are trying to be helpful, I guess). Perhaps you have a simple reply for them? I assume it’s because it blocks too much legit traffic, but I don’t think I’ve actually asked before
davic joined the channel
zas[m]
Actually we are currently pondering about various "solutions", none is perfect, and fitting them in our infrastructure isn't always simple. The ones you cited are candidates of course.
aerozol[m]1
Thanks for the info! I'll reply
Kladky joined the channel
zas[m] uploaded an image: (61KiB) < https://matrix.chatbrainz.org/_matrix/media/v3/download/matrix.org/wbBoXiuFKUWuaNksPxDQdutb/image.png >
bitmap: tests pass in https://github.com/metabrainz/musicbrainz-serve... - if you can take a look and it seems sensible, I'd put it on beta for a while for people to catch any AC weirdnesses we might have missed :)
yvanzo: looking at MBS-12729 that you shared yesterday :)
"By default, search queries are sent to search.musicbrainz.org and are rate limited." we should drop that line. But I guess given MBS-12730 is not done at the moment, the whole "Rate limiting" section in the install docs is basically outdated now
I guess we could rewrite it to something like "the server doesn't rate limit by default, if you are getting 503s check if the owner of the server you are using in SEARCH_SERVER has set specific rate limits"
Or actually just rename this section to "Searches"
reosarevok[m] sent a code block: https://matrix.chatbrainz.org/_matrix/media/v3/download/chatbrainz.org/jJTqbQwFpPXVfBkJfDkdeyTM
I can't get the metabrainz-prod container on isaac to start.
Jul 4 10:43:18 810cea9ddcd8 consul-template[30]: (view) kv.get(docker-server-configs/MeB/config.prod.json/payments/production): Unexpected response code: 429 (Your IP is issuing too many concurrent connections, please rate limit your calls) (retry attempt 6 after "8s")... (full message at <https://matrix.chatbrainz.org/_matrix/media/v3/...>)
these were the errors I was getting, but now the container is silent and not starting.
<suvid[m]> "Hi monkey..." <- suvid: Yes regarding the /settings/import/ page, and to opening a separate PR (based on your working branch if you want to have the functionality as well).
For the mockup, generally it looks fairly good, however I think you'll want to put yourself in the shoes of a user: you probably want to select the music service first (so that dropdown first), especially if it is required you don't want to place it after optional fields.
Also, two thing not currently shown in the mockup: 1. is the status component (reusing existing ImportStatus component) instead of the warning message about taking hours, and 2. is how you will display previous imports
petitminion has quit
bitmap[m] has quit
bitmap[m] joined the channel
bitmap[m]
<zas[m]> "bitmap: when you are around..." <- I'm not sure exactly what happened earlier, but pgbouncer seems okay now. haproxy logs are being spammed (since a while) because jimmy is still in the config
I can restart these at least
zas[m]
I think everything is ok now, but better double check, not sure what happened, I stopped 3 containers, one of them is likely spamming consul for some reason
bitmap[m]
when you have time later, would it be possible to update the "HaProxy Postgresql" dashboard to reference cesaria/isaac instead of jimmy/hendrix? (not urgent)
bitmap[m] is not exactly sure how to do this himself
