Participants ============ 1. guilhem 2. Emiliano 3. cloph Agenda ====== * Feedback to Emiliano about PeerTube https://listarchives.tdf.io/i/b5auLBp6FjwNt6srGuKdLSAm + g: as presented it feels more like a solution in search of a problem (and not the opposite unlike written on the list ^^) + Emiliano: AFAIK the main point is to let Paolo reclaim storage on his instance and serve videos ourselves - g: if we want a video platform then peertube is likely the way to go, but is there really demand from the LO community for this? - serving the videos ourselves (and not from a 3rd party) is a noble cause, but it's much simpler to serve directly from backend in that case (no need for a separate platform) . see https://video.fosdem.org for an example - cloph: agreed, no objection but if we don't need a full-fledged video platform there is indeed little point of having our own instance + Emiliano: is it feasible/desired to source videos ourselves (blog, website, conference) — and with which overhead? - cloph: hetzner cloud has 20TB/month included and we'd probably not exceed that . if it does however then we could just split the backend and use a load balancer - g: community might want a video platform later down the road (eg for commenting, counters, favs etc), but having a storage backend at hand is in fact a prerequisite . so these are not conflicting / incompatible proposals; it's more like one is an extension of the other + space needs is unclear at the moment - cloph: don't think it'd be too much of a problem budget-wise anyway, 16TB for 80€/month would cover all our needs . g: agreed, but iirc storage box is suboptimal for public storage backend (connections are capped) * PostgreSQL backups (Brett absent) * Fallback/handover during cloph's vacation * cloph: not much handover to the infra team (releases are handed by Xisco) + guilhem to publish releases and push to mirrors (incl portableapps) * AI guilhem: Upcoming upgrades (between xmas and new year, potentially disruptive) + Matomo (OS + 4.4 → 4.6 — downtime due to schema upgrade) + gerrit (OS + 3.3 → 3.4) - already tested on the stage instance, but not the replication logic - possible regression for replication over SSH (untested), in the worst case we need to be prepared for an emergency downgrade + crashreport (OS, downtime due to PostgreSQL upgrade via dump+reimport, should take ~24h) - cloph: no problem + jenkins OS + {OpenGrok (JGit backend would be a welcome change but ≥1.6 requires Tomcat10…) - g: tomcat's security track record isn't so pretty, feel more comfortable sticking to what's provided by the OS proper and let its security team take care of security fixes - that means we're stuck to tomcat9 for now, hence to {OpenGrok <1.6 - can reconsider later if people badly want {OpenGrok 1.6 or later * Feedback to the SUSE folks regarding gandalf: + g, cloph: no objection + guilhem to reply * Next call: Jan 18 at 17:30 CET + happy holidays! :-]
-- Guilhem. -- To unsubscribe e-mail to: website+unsubscr...@global.libreoffice.org Problems? https://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/ Posting guidelines + more: https://wiki.documentfoundation.org/Netiquette List archive: https://listarchives.libreoffice.org/global/website/ Privacy Policy: https://www.documentfoundation.org/privacy
