• a [[server]].
  • as of [[2022-05-15]] essentially runs [[social coop]].
  • this mostly means running [[mastodon]] for the community: https://social.coop
  • to get to a shell in the mastodon web container, run: sudo docker exec -it docker_web_1 /bin/bash
  • as per https://git.coop/social.coop/tech/operations/-/wikis/Elasticsearch, thanks [[red oak]]
  • #ip 95.216.13.24
  • #admin [[twg]]

• [[flancian@social.coop]] https://social.coop/@flancian/109373340023561205 pull
• [[flancian@social.coop]] https://social.coop/@flancian/109575671146638912 pull

• [[2022-12-25 18:18:49+00:00]] @[[flancian]]: https://twitter.com/flancian/status/1607078425767854081 pull

• [[2022-11-20 13:46:03]] [[@flancian:matrix.org]] (link):
  • I’d be happy to experiment setting up a new wiki in [[hypha]] using new ansible playbooks; this way we can learn the setup without risking production (which is in [[runko]])

• [[2022-11-20 00:38:44+00:00]] @[[flancian@social.coop]] https://social.coop/@flancian/109373340023561205:
  • @ntnsndr @Matt_Noyes @edsu @protean for the first month yes, then we’ll revisit. I think an extra server can give enough value to be worth keeping around 24/7 if we use it right.About discretionary budget: sounds great! €50/mo would be sufficient to run this experiment and also scale it up as needed. Having €100 pre-approved to use one-off at some point (if €100/mo is too much for discretionary) would also let us react quickly if [[runko]] dies (new lease + one off setup fee < €100).
• [[2022-12-25 18:14:12+00:00]] @[[flancian@social.coop]] https://social.coop/@flancian/109575671146638912:
  • Then I saw the social.coop instance was having a hiccup and took a look. It may be OOM related based on what people in the Tech Working Group saw so I added some swap while we prepare the server move from [[runko]] to [[rhizome]].Happy to have been maybe useful, let’s see if it works :)

• [[2022-12-25 18:18:49+00:00]] @[[flancian]]: https://twitter.com/flancian/status/1607078425767854081 pull

Then I saw the https://t.co/lG1n70SJ9T instance was having a hiccup and took a look. It may be OOM related based on what people in the Tech Working Group saw so I added some swap while we prepare the server move from [[runko]] to [[rhizome]].

Happy to have been maybe (1/2)