Conversation

Notices

1. Embed this notice
   Daniel Supernault (dansup@mastodon.social)'s status on Wednesday, 12-Jun-2024 17:07:41 JST Daniel Supernault

   You may have noticed pixelfed.social is slower than usual, after some investigation, I have discovered that Lemmy is behind the sudden surge in traffic.

   While we are implementing Groups, we expected additional load from Lemmy and other Group supported projects

   However, I never anticipated how much activity we'd get from Lemmy

   To put this in perspective, lemmy.world has sent more activities to us than mastodon.social (to pixelfed.social)

   I'm overhauling the inbox again to better handle this!

   • Embed this notice
     Michael T. Richter (qqmrichter@mastodon.world)'s status on Wednesday, 12-Jun-2024 17:33:18 JST Michael T. Richter

     in reply to

     @dansup You might want to make your polling of upstream dynamically periodic. If a given upstream hits you too hard, lower the frequency of updates. If the upstream has a low volume, you can safely hit them more often.

     Think of it like a PID for network backpressure.

   • Embed this notice
     deromme (deromme@mastodon.social)'s status on Thursday, 13-Jun-2024 00:56:33 JST deromme

     in reply to

     I didn’t know Lemmy was still a thing 😅

     The internet keeps me amazed 🤯

   • Embed this notice
     Emelia 👸🏻 (thisismissem@hachyderm.io)'s status on Thursday, 13-Jun-2024 04:29:30 JST Emelia 👸🏻

     in reply to

     @dansup could it be that Lemmy just has a very inefficient federation model that could be improved? (i.e., only publishing content to remote server's inboxes when there's relationships on that server?)