Every time I try to access this community, ther’s some kind of problem with the server. If you have a look at the status page, it’s almost all orange/red. The problem aren’t DDoS attack since the server is behind Cloudflare protection. Admin/mods, why don’t you move this community to a different server instance? I’m not accusing anybody, I know that maintain a server can be a challenging sometimes, I just want to enjoy this community!

Please!

@Loki

  • static09@lemmy.world
    link
    fedilink
    English
    arrow-up
    50
    ·
    1 year ago

    ADMIN, isn’t it time to move from lemmy.world?

    They said, from their lemmy.world account.

    • peregus@lemmy.worldOP
      link
      fedilink
      English
      arrow-up
      7
      ·
      1 year ago

      What’s wrong with that? I’ve started here some months ago and I mainly follow this community.

      • Dr. Jenkem@lemmy.blugatch.tube
        link
        fedilink
        English
        arrow-up
        19
        ·
        1 year ago

        The problem is it’s the largest and is being attacked. Use a different instance and you’ll have no problem accessing this community. If it weren’t for people posting about the outages, I’d never even know.

          • ninjan@lemmy.mildgrim.com
            link
            fedilink
            English
            arrow-up
            6
            ·
            1 year ago

            I honestly haven’t noticed. For me to see the posts they simply need to get pushed, which is pretty lightweight. And if Lemmy.world crashes after and stays down for a few hours it won’t impact my ability to view the post. For a lemmy.world user the instance needs to be up for them to view any content.

          • Dr. Jenkem@lemmy.blugatch.tube
            link
            fedilink
            English
            arrow-up
            2
            ·
            1 year ago

            In a way yes, when I post a comment, no one else will be able to see it while the site is down. But it eventually will go through when the site is up. And because I’m browsing from a different instance, I can see all the content even if lemmy.world is offline.

        • peregus@lemmy.worldOP
          link
          fedilink
          English
          arrow-up
          5
          ·
          1 year ago

          I actually didn’t think about it.Icm on Lemmy, but I’m not a Lemmy specialist so I didn’t know that I could go to a different instance and access this communit indirectly. Thanks

          P.s. I didn’t get the agenda part.

      • xebix@lemmy.srv0.lol
        link
        fedilink
        English
        arrow-up
        7
        ·
        1 year ago

        They explained that the attacks are in the form of requests that tax the database server, not the website itself.

        • peregus@lemmy.worldOP
          link
          fedilink
          English
          arrow-up
          1
          ·
          1 year ago

          The database can’t be accessed directly, all the requests needs to be done from the website or API.

          • xebix@lemmy.srv0.lol
            link
            fedilink
            English
            arrow-up
            8
            ·
            1 year ago

            Right, but it’s possible to execute those API requests to trigger those expensive database requests in a way that wouldn’t necessarily trigger cloudflare’s DDoS protection.

      • gabriele97@lemmy.g97.top
        link
        fedilink
        English
        arrow-up
        6
        ·
        1 year ago

        Yep Cloudflare protects against classic DDoS (like many clients doing a lot of small requests). Here attacks are performed presumibly by users that know very well how the Lemmy’s backend works and where bottlenecks are, so that with a small number of well made requests they are able to mess up the backend and Cloudflare doesn’t notice it

      • dditty@lemmy.world
        link
        fedilink
        English
        arrow-up
        4
        ·
        1 year ago

        Cloudflare DDos protection isn’t a silver bullet; the attacks are distributed and come from shifting source IPs, and are sophisticated in that they exploit resource intensive queries specifically designed to overload a Lemmy instance. If lemmy.world were to pivot to some other instance, who’s to say the culprits wouldn’t just resume their efforts pointed at the new location? There are theories these may be carried out by the recently-defederated fringe hate communities

      • Pankkake@lemmy.world
        link
        fedilink
        English
        arrow-up
        3
        ·
        1 year ago

        From what I understand, Cloudflare can block some DDoS attacks, but not all of them.

        The attacks on Lemmy have to do with poorly optimized SQL requests; these are requests that shouldn’t take long to execute, but do due to some oversight. By spamming these requests, the attackers can bring Lemmy on it’s knees.

        Actually, wouldn’t this attack better be categorized as a DoS attack ? What’s so distributed about it ?

      • habitualTartare@lemmy.world
        link
        fedilink
        English
        arrow-up
        3
        ·
        1 year ago

        Cloudflare has DDoS protection but it can’t stop everything 100% of the time. According to the admins, the attackers are very familiar with how lemmy works and are using this knowledge to overwhelm resources. This isn’t just a simple script kiddy or bonnet for hire but likely points to someone that has worked within the lemmy community.

        https://www.cloudflare.com/learning/ddos/what-is-a-ddos-attack/

        https://www.cloudflare.com/learning/ddos/glossary/web-application-firewall-waf/

      • FARTYSHARTBLAST@kbin.social
        link
        fedilink
        arrow-up
        1
        ·
        edit-2
        1 year ago

        It can help, but is not perfect. There’s kind of been an arms race between services like Cloudflare and script kiddies/hackers DDOSing: Their methods became a little more sophisticated to the point that they keep the traffic shifting from address to address so it’s harder to track/block.

        They’ll do other things to mess with the host too like spamming “white noise” pictures to fill up server storage space, so while DDOS attacks play a role, there are other issues at play on top of that.

        For what it’s worth, I’ve been using Kbin.social and sh.itjust.works as well and they have also had some issues here and there. No host will be perfect and invulnerable from every attack, and when an instance becomes more popular, it becomes increasingly likely to be targeted by attackers.

    • GreyBeard@lemmy.one
      link
      fedilink
      English
      arrow-up
      7
      ·
      1 year ago

      There’s a difference between self hosting and running a public service. There’s a lot more overhead running a pubic service.

    • peregus@lemmy.worldOP
      link
      fedilink
      English
      arrow-up
      4
      ·
      edit-2
      1 year ago

      What’s wrong with that? I’ve started here some months ago and I mainly follow this community and I don’t blame anyone.

  • Droolio@feddit.uk
    link
    fedilink
    English
    arrow-up
    10
    ·
    1 year ago

    Well your account is on lemmy.world so how d’ya know the issue isn’t with your own access to the front end?

    Many don’t interact with the lemmy.world directly, so we might only see delays in post propogation (if there is such an issue on the backend - I don’t see any but could be wrong).

    I agree picking the biggest instances isn’t great from a scaling perspective, but s’gonna be hard to move any community once established.

  • boyi@lemmy.sdf.org
    link
    fedilink
    English
    arrow-up
    4
    ·
    edit-2
    1 year ago

    The problem is not the server. One of the reasons is because lemmy world is being DDoS attacked by certain party with grudges. Because lemmy is not used to this kind of attack, which is made difficult by way it is designed and open source nature where the attacker can easily uncover vulnerabilities, it took some times and learning to mitigate them. You can open new account in other instances or move if that’s too much inconvenience for you. No one is stopping you.

  • Johannes Jacobs@lemmy.jhjacobs.nl
    link
    fedilink
    English
    arrow-up
    3
    ·
    1 year ago

    Have you donated money so they can?

    Another server (and i assume you mean one with better specs) means more money. Its not just challenging, its also costly.

    Alternative you could setup your own lemmy instance, you can still join the community’s on lemmy.world or lemmy.ml or whereever :) and its fun to do.

  • PeachMan@lemmy.one
    link
    fedilink
    English
    arrow-up
    2
    ·
    1 year ago

    What makes you think other servers aren’t having similar problems?

    (spoiler alert: they all are, lemmy is an immature platform)

  • krayj@sh.itjust.works
    link
    fedilink
    English
    arrow-up
    2
    ·
    edit-2
    1 year ago

    By claiming that the problem isn’t DDOS, you’re just advertising your ignorance. Cloudflair is outstanding for protecting static web content against DDOS, and Lemmy.world is well protected against that. The problem is certain dynamic pages and api calls that can only be rendered from costly realtime dynamic database operations…those are the url that the DDOS attackers are focusing on and those are the kinds of content that cannot be easily protected by cloudflair.

    Your premise, though, is still accidently correct. The way to mitigate instances being targeted by DDOS is to spread the user base and community hosting across a vast number of instances so that no one instance is such a rewarding target for DDOS attack.

    • peregus@lemmy.worldOP
      link
      fedilink
      English
      arrow-up
      2
      ·
      1 year ago

      You’re right about my ignorance about Lemmy, I’m a user on this federated …thing and I know nothing ahout the Lemmy server. Being in a selfhosted community your answer is what I was expecting (maybe with less attack). The API are used only by the federated instance or also by the smartphone apps? For what I see, it seems to be the former, and, if it really is so, the API calls could be allowed only by those server and blocked from everyone else, Cloidflare WAF can do this. I know that the servers are a lot, but it could be possible to insert in the WAF all the IPs of the federated instances. …or not?

  • solberg
    link
    fedilink
    English
    arrow-up
    1
    ·
    1 year ago

    Is it possible to migrate a community from one instance to another?

    • krayj@sh.itjust.works
      link
      fedilink
      English
      arrow-up
      1
      ·
      1 year ago

      No, not programmatically. The only way to ‘move’ an instance is for a community of users to create a new community elsewhere and arbitrarily start using it as their preferred community. There is no way to force it and there is no mechanic for moving.