Conversation

Notices

1. Embed this notice
   matrix07012 :thotpatrol: (matrix@gameliberty.club)'s status on Saturday, 15-Mar-2025 00:11:37 JST matrix07012 :thotpatrol:

   • Phantasm
   • Pawlicker
   • dilbert 1

   @PurpCat @phnt @sun Isn't Mastodon older?

   • Embed this notice
     dilbert 1 (sun@shitposter.world)'s status on Saturday, 15-Mar-2025 00:11:35 JST dilbert 1

     in reply to

     • Phantasm
     • Pawlicker
     @matrix @PurpCat @phnt mastodon was built from the ground up for scaling, you can argue about how good it is but the attempt was made
   • Embed this notice
     Phantasm (phnt@fluffytail.org)'s status on Saturday, 15-Mar-2025 00:12:18 JST Phantasm

     in reply to

     • Pawlicker
     • dilbert 1
     @matrix @PurpCat @sun It is. By like a year or two depending on where you look.
   • Embed this notice
     dilbert 1 (sun@shitposter.world)'s status on Saturday, 15-Mar-2025 00:15:35 JST dilbert 1

     in reply to

     • Phantasm
     • Pawlicker
     @matrix @PurpCat @phnt honestly it's really frustrating that out of the box with no config a typescript node.js application doesn't have the same problems
   • Embed this notice
     matrix07012 :thotpatrol: (matrix@gameliberty.club)'s status on Saturday, 15-Mar-2025 00:15:36 JST matrix07012 :thotpatrol:

     in reply to

     • Phantasm
     • Pawlicker
     • dilbert 1

     @sun @PurpCat @phnt Pleroma should ideally be better at scaling because of Erlang

   • Embed this notice
     lainy (lain@lain.com)'s status on Saturday, 15-Mar-2025 00:18:51 JST lainy

     in reply to

     • Phantasm
     • Pawlicker
     • dilbert 1
     @sun @PurpCat @phnt @matrix I still have no idea what’s going on with your servers. It’s not expected performance.
   • Embed this notice
     dilbert 1 (sun@shitposter.world)'s status on Saturday, 15-Mar-2025 00:20:21 JST dilbert 1

     in reply to

     • lainy
     • Phantasm
     • Pawlicker
     @lain @PurpCat @phnt @matrix its ok right now, the current issue is a bug in pleroma when unauthenticated access to local objects is disabled it kills some federation
   • Embed this notice
     Phantasm (phnt@fluffytail.org)'s status on Saturday, 15-Mar-2025 00:40:53 JST Phantasm

     in reply to

     • Pawlicker
     • dilbert 1
     @matrix @sun @PurpCat feld had an almost working prototype for multiple nodes at some point I think. There was an issue with Cachex I think and that was it.
   • Embed this notice
     pomstan (pomstan@xn--p1abe3d.xn--80asehdb)'s status on Saturday, 15-Mar-2025 00:41:05 JST pomstan

     in reply to

     • Phantasm
     • Pawlicker
     • dilbert 1

     @matrix @sun @PurpCat @phnt this shit will never be not funny

      likes this.
   • Embed this notice
      (mint@ryona.agency)'s status on Saturday, 15-Mar-2025 00:42:26 JST 

     in reply to

     • Phantasm
     • Pawlicker
     • dilbert 1
     @phnt @PurpCat @matrix @sun Multiple nodes that still have a single bottleneck in form of pumping I/O back and forth between the node and postgres.
   • Embed this notice
     Phantasm (phnt@fluffytail.org)'s status on Saturday, 15-Mar-2025 00:48:27 JST Phantasm

     in reply to

     • 
     • Pawlicker
     • dilbert 1
     @mint @PurpCat @matrix @sun You can loadbalance multiple postgres nodes in a read scenario. And in the case of Pleroma, writing to the DB usually isn't as I/O heavy.
      likes this.
   • Embed this notice
     feld (feld@friedcheese.us)'s status on Saturday, 15-Mar-2025 00:59:31 JST feld

     in reply to

     • Phantasm
     • Pawlicker
     • dilbert 1
     @phnt @PurpCat @matrix @sun oh yeah we just need to replace Cachex and it's not hard, I just ended up moving and other priorities stacked up
     Phantasm likes this.
   • Embed this notice
     feld (feld@friedcheese.us)'s status on Saturday, 15-Mar-2025 01:04:02 JST feld

     in reply to

     • Phantasm
     • Pawlicker
     • dilbert 1
     @matrix @PurpCat @phnt @sun there is a cluster native caching mechanism available to use which has a beautifully simply syntax because you just decorate the functions you want to cache and which ones to bust the cache. Then you also configure which caches should be independent per node and which should be shared across the cluster
     
     https://hexdocs.pm/nebulex/Nebulex.Caching.Decorators.html
   • Embed this notice
     matrix07012 :thotpatrol: (matrix@gameliberty.club)'s status on Saturday, 15-Mar-2025 01:04:03 JST matrix07012 :thotpatrol:

     in reply to

     • Phantasm
     • Pawlicker
     • dilbert 1
     • feld

     @feld @PurpCat @phnt @sun why not Redis/Valkey? Is it to minimize dependencies?

   • Embed this notice
     feld (feld@friedcheese.us)'s status on Saturday, 15-Mar-2025 01:12:38 JST feld

     in reply to

     • Phantasm
     • Pawlicker
     • dilbert 1
     • feld
     @matrix @PurpCat @phnt @sun also yes the main reason for not using redis/valkey is that it's an unnecessary dependency when this type of functionality is core to the language already and it will just perform better because the OS doesn't need to context switch to access the cache
   • Embed this notice
     lainy (lain@lain.com)'s status on Saturday, 15-Mar-2025 01:14:09 JST lainy

     in reply to

     • Phantasm
     • Pawlicker
     • dilbert 1
     • feld
     @feld @PurpCat @phnt @matrix @sun tbh there are zero currently active pleroma instances that would need to run distributed on multiple nodes.
   • Embed this notice
     feld (feld@friedcheese.us)'s status on Saturday, 15-Mar-2025 01:14:40 JST feld

     in reply to

     • Phantasm
     • Pawlicker
     • dilbert 1
     @matrix @PurpCat @phnt @sun also, if you still really really wanted to use Redis then Nebulex supports using it as a backend without really needing you to make any code changes, so it would give flexibility for people
   • Embed this notice
     matrix07012 :thotpatrol: (matrix@gameliberty.club)'s status on Saturday, 15-Mar-2025 01:14:42 JST matrix07012 :thotpatrol:

     in reply to

     • Phantasm
     • Pawlicker
     • dilbert 1
     • feld

     @feld @PurpCat @phnt @sun I see, that's much better than redis

   • Embed this notice
     dilbert 1 (sun@shitposter.world)'s status on Saturday, 15-Mar-2025 01:16:45 JST dilbert 1

     in reply to

     • lainy
     • Phantasm
     • Pawlicker
     • feld
     @lain @PurpCat @phnt @feld @matrix did you know poa.st is on stock pleroma now
     lainy likes this.
   • Embed this notice
     Phantasm (phnt@fluffytail.org)'s status on Saturday, 15-Mar-2025 01:18:26 JST Phantasm

     in reply to

     • lainy
     • Pawlicker
     • dilbert 1
     • feld
     @sun @PurpCat @feld @matrix @lain I also remember graf talking about failover Pleroma (Rebased at that time) nodes he has set up. Same with Postgres. So in this one specific case, he almost does run multi-node.
     ✙ dcc :pedomustdie: :phear_slackware: likes this.
   • Embed this notice
     dilbert 1 (sun@shitposter.world)'s status on Saturday, 15-Mar-2025 01:19:11 JST dilbert 1

     in reply to

     • lainy
     • Phantasm
     • Pawlicker
     • feld
     @phnt @PurpCat @feld @matrix @lain they basically just threw huge hardware at it, I think the biggest gain they got was from dedicated db server
     Phantasm likes this.
   • Embed this notice
     Phantasm (phnt@fluffytail.org)'s status on Saturday, 15-Mar-2025 01:19:48 JST Phantasm

     in reply to

     • lainy
     • Pawlicker
     • dilbert 1
     • feld
     @sun @PurpCat @feld @matrix @lain The point of it all was self-healing I think, so graf doesn't have to touch anything most of the time.
   • Embed this notice
     feld (feld@friedcheese.us)'s status on Saturday, 15-Mar-2025 01:20:50 JST feld

     in reply to

     • lainy
     • Phantasm
     • Pawlicker
     • dilbert 1
     @sun @PurpCat @phnt @matrix @lain they all come home eventually 💅
     Phantasm likes this.
   • Embed this notice
     👺防空識別區👹 (adiz@mtl.jinxian.casa)'s status on Saturday, 15-Mar-2025 01:27:40 JST 👺防空識別區👹

     in reply to

     • lainy
     • Phantasm
     • Pawlicker
     • dilbert 1
     • feld

     @feld The same with Misskey! @PurpCat @phnt @matrix @lain @sun

     feld likes this.
   • Embed this notice
     jae (jae@fsebugoutzone.org)'s status on Saturday, 15-Mar-2025 01:58:03 JST jae

     in reply to

     • 
     • Phantasm
     • Pawlicker
     • dilbert 1
     @mint @phnt @PurpCat @matrix @sun
     
     > Multiple nodes that still have a single bottleneck in form of pumping I/O back and forth between the node and postgres.
     
     if you're evenly distributing ingress traffic from the proxy to the nodes, it should (depending on lb algorithm) spread traffic evenly(ish) across the nodes then down the io tunnel to pg. if the pg node had a low latency link (ie same network/lan) i/o should be less of an issue i would think.
     
     i recently experimented with pleroma on kubernetes with an out of band pg backend over wireguard (local, not via vps) and threw stressor on it. performance was pretty good tbh
     , ✙ dcc :pedomustdie: :phear_slackware: and Phantasm like this.