Ever since things came back up after the Jan 5th outage, I’ve started to encounter regular timeouts. I will scroll past a couple dozen number posts and then it will stop as of there are no more. Within a few seconds, the application produces a time out error.

I use Boost for Lemmy as my client, but I’m not convinced that the issue is the app since switching my instance allows me to continue scrolling without a problem. And to be clear, prior to January 5th, I’ve never experienced a timeout in the app.

I’m curious if I’m the only one experiencing timeouts on Lemmy.ca. If so, then I’m curious if the admins are aware of any issue.

  • GlassHalfHopefulOP
    2·
    12 days ago

    Alright, @[email protected]. Here’s some more information.

    General Findings

    • Problem occurs on the WebUI as well while logged into Lemmy.ca.
    • Problem occurs regardless of WebUI front-end used.
    • Problem occurs regardless of Andriod apps used.
      • Exception: Eternity (Nightly). I don’t know if they poll for information differently or what, but I can endlessly scroll without issue at the moment.
    • Error outputs are unfortunately inconsistent between FEs. (The standard mobile UI shows no error and sim.)
    • Problem occurs both with and without VPNs employed.

    Findings by Server

    • lemmy.ca: BE: 0.19.7
      • Problem instance and My primary.
    • lemmy.world: BE: 0.19.3-7-g527ab90b7
      • No issues observed. Logged in with Boost & Voyager using alternate account.
    • beehaw.org: BE: 0.18.4
      • No issues observed. Logged in with Boost using alternate account.
    • lemm.ee: BE: 0.19.8
      • No issues observed. Logged in Anonymously via Voyager Android App.
    • feddit.uk: BE: 0.19.7
      • Will try if/when they approve the account request.
      • Really want to try this since they are using the same BE as lemmy.ca
    • lemmy.sdf.org and another instance with BE: 0.19.3
      • Will try if/when they approve my alt accounts. If someone knows an app where I can browser sdf as anonymous, then I can do it now.

    Errors By Application Note the differences in how the error is presented. Sending timestamps and IP to @[email protected] via DM for the following:

    1. Boost (Android)

    1. Voyager (Android)

    1. Photon as Android Firefox PWA via lemmy.ca

    1. Lemmy WebUI via lemmy.ca

    simply renders empty after the timeout. this was page 2

    • ShadowMA
      2·
      12 days ago

      I suspect something might be funky with your account, rather than the network / apps. The timeout message is a lie, you’re really getting 400 / 499 errors that seem to be related to this lemmy error message:

      2025-01-20T17:30:03.366014Z  WARN Error encountered while processing the incoming HTTP request: lemmy_server::root_span_builder: NotAModOrAdmin: NotAModOrAdmin
   0: lemmy_api_common::utils::check_community_mod_of_any_or_admin_action
             at crates/api_common/src/utils.rs:102
   1: lemmy_api::local_user::report_count::report_count

      and:

      2025-01-20T20:39:43.202677Z  WARN Error encountered while processing the incoming HTTP request: lemmy_server::root_span_builder: CouldntFindPerson: Record not found
   0: lemmy_apub::fetcher::resolve_actor_identifier
             at crates/apub/src/fetcher/mod.rs:22
   1: lemmy_apub::api::read_person::read_person

      I wonder if it could be a bug related to the unicode name you’ve got going on. Did you set that recently, or has it been like that since day 1?

      I just restarted all of our docker containers, see if that helps? If not, try setting a normal name and confirm if that changes things?

      • GlassHalfHopefulOP
        1·
        8 days ago

        Well, I’ve checked out the source. (Rust is very foreign to me. Haha.) However, I can see the offending code now. I don’t really have the means to dig further, but I did learn the following:

        RE: First error

        1. the error comes from a method performing the following: “Check that a person is either a mod of any community, or an admin.”
        2. the code block where NotAModOrAdmin error is returned as had the error check was refactored and moved. but… it looks like it will still act the same in later versions of the BE.
        3. what i cannot say is WHY this being checked. i am neither, so it is_mod_of_any_or_admin == false. but… so what? i neither know why the check is occurring nor why it matters.
        4. the stack trace shows preceding calling method which is report_count(), but no idea what happens before that or why… so unfortunately i didn’t learn too much about this “error.” i wonder if it would happen still if I was the mod of any given community. it should pass, but again… what would that matter and why? i am literally only trying to load a list of posts from the front page. (seems to happen more often when looking at “All” posts)

        Re: Second Error

        1. resolve_actor_identifier make two attempts to get the actor (my user account me thinks). the call to DbActor::read_from_name_and_domain() seems to fail and so webfinger_resolve_actor() is called which also does not come back OK.
        2. it’s possible my user isn’t being found due to network latency, issues with server response time, or who knows what else.

        So the cause? Who knows. Ugh.

        Even though this started for me after the defective power supply was replaced… if that’s ALL they changed (including not adjusting any server settings at all)… then we shouldn’t see this at all. I wonder still if something happened while writing to the DB and the power cut. perhaps something related to my account is unhappy in the DB. Who knows.

        You all will be changing hardware providers in the near future. The next version of the BE, which you will eventually update, also adjusts the affected code a bit. And so… maybe it will resolve itself?

        One can hope.