Usually when I reach the end of the feed, it just loads more. But recently, I started getting a bad request error instead. Not every time I reach the end, but once I get it it doesn’t go away when I tap on retry.

Is anyone else seeing this happening? Is there anything I can do to fix it?

    • idunnololzM
      3·
      4 days ago

      This might have to do with the user agent issue. I’m planning to do a release tomorrow to fix this.

  • rescue_toaster@lemmy.zip
    1·
    4 days ago

    Yeah this has been occurring for me as well. It’s getting frustrating. Would hate to switch to another lemmy app as otherwise I really enjoy summit.

  • Bubs@lemmy.zip
    1·
    6 days ago

    I’ve been getting the same for maybe 2/3 weeks now. Quite frustrating because I’m missing a lot of posts and keep getting repeat posts all the time. Plus, sometimes the app just refuses to download images to my phone.

    I would hate to leave summit over this, but it’s really making Lemmy unusable.

  • limerod@reddthat.com
    1·
    6 days ago

    I have been getting 400 bad request error in comments, in URL preview of opened posts. I guess something is on summits end. Because, these errors are less frequent in Voyager and its very quick I operation.

  • Dessalines@lemmy.ml
    3·
    7 days ago

    This is a DB attack vector, which is why lemmy long ago switched to cursor pagination.

    You need to open up an issue in the summit repo asking the dev to switch away from the deprecated page/limit, and to cursor pagination.

    • flamingos-cant (hopepunk arc)@feddit.ukEnglish
      5·
      8 days ago

      I can recreate this, the error I’m seeing on the server side is Page limit is > 10. I wish client side gave a better error message here than counldnt_get_posts.

      This seems to have been a deliberately changed in this PR to fix comment list sorting DDOS exploit. How this affects clients depends on if they use page or page_cursor, which has no limit.

      For now I’ll just patch our Lemmy to increase to increase the limit to something more reasonable, like 100. Probably need to upstream this as well as it’ll affect every instance on 0.19.14.

    • Da Bald Eagul@feddit.nl
      3·
      8 days ago

      Likely is. I’m experiencing something similar with timeouts instead, but it happens across clients (and less often since a server upgrade). So probably a change in configuration for your instance, or an uptick in usage/growing userbase?

      Either way I’d check to see if this happens on other clients too, first.