i don’t think so, but this is part of the lemmy source code, not configurable. i might take a look at changing that for LW though.

hi, this should be fixed now. see also https://lemmy.world/post/36187021 for some additional context.

I believe this should be fixed now.

@Monster96@lemmy.world

the issue lies with cloudflare and not lemmy. wold, correct?

the issue seems to stem from a change done by Cloudflare not too long ago, but that doesn’t mean that we’re unable to work around it. i’m currently implementing some changes that should help with this.

any info or links about this cloudfare issue known atm?

the link was included in my previous comment

please translate this into user experience

for the most part, any “read” operations (looking at a post, comment, user, etc) should not be affected by this. for any “write” operations (login, posting, commenting, editing posts/comments, etc.) there seems to be a higher error rate currently related to this. in those cases the action would simply not be completed, and depending on the client (app, default web interface, alt ui) this could range from and endless loading indicator to an error being displayed to nothing being displayed.

if this has resulted into more strict automod functions

no, and this was not an automod removal. you can see the reason why it was removed in the modlog: https://lemmy.world/modlog?postId=36129581

you should also have received a message from our automod informing you about the removal, but automod is only informing you about what happened, not the one taking action in this case.

seems like we’re not the only ones trouble by some changes clouflare did to their OWASP ruleset: https://community.cloudflare.com/t/owasp-ruleset-unexpectedly-has-a-high-false-positive-rate/814544/19

this is a change cloudflare did to the detections they have for http requests to our server. this could affect any API calls from any clients towards our API backend, including automod actions. at most however this would prevent some automod actions from being executed, it couldn’t cause e.g. false positives.

thanks for sharing this.

it seems that cloudflare’s managed rules for the OWASP are hitting a bit much lately. this request has been blocked as “Inbound Anomaly Score Exceeded” with an “OWASP score” of 49, hitting a bunch of rules in the managed OWASP ruleset, many of them relating to SQL injections. i’m gonna have a look at what we can do to tune this down.

as @Blaze@lemmy.zip mentioned, we don’t care about usage of multiple accounts generally.

we do prohibit the same person voting on content (their own or others’) with multiple accounts, and obviously using multiple accounts for ban evasion is not allowed. spamming a community with multiple accounts pretending to be different persons should also be avoided unless the community explicitly allows it, which seems to how you want to do it based on your description.

you could save some storage with this, but i don’t think it’s a good idea. a lot of people expect the threadiverse to have a lot more permanence than e.g. mastodon, similar to reddit. being able to find old posts/comments about a certain topic is one of the things that made reddit as useful as it used to be, especially when searching for tech related issues in my experience. old doesn’t necessarily mean obsolete, and whether this would be suitable would be highly dependent on the community. most communities are not intended to be for ephemeral content only.

what would that have to do with this?

not sure if that is implemented there, but you can always go to https://lemmy.world/settings in your web browser

we do have some limits in place, but lemmy only allows rate limits per ip, and those are counted in each backend process independently. I’m currently working on implementing better rate limits in our load balancer.

due to rate limits historically not working at all or not working properly, there are still various instances without decent limits. additionally, these rare limits only apply to local users. federated activities are not limited within lemmy. we recent added some fairly high limits to our automod to catch some of these cases and it’s been working alright so far.

removed from all 3

we don’t ban individual IP addresses, but we restrict posting/commenting/media uploads from some VPN/data center IPs.

afaik the decrease in users has mostly been on mastodon.world, where the suspected reason is to a large degree requiring registration applications to prevent spam bots. this seems to scare off a lot of users unfortunately. i’m not involved in the mastodon stuff, but afaik there is currently no system similar to what we have on lemmy or piefed to automatically accept applications after certain checks were successful.

they’re referring to Lemmy developers, not someone using/running .world

I’m assuming that you’re referring to the Lemmy developers. FHF supports Lemmy development with a monthly donation of €50, same as we do for Mastodon development.

If it makes you feel better, there are enough other people donating who don’t mind some or all of their donation amount going towards Lemmy development to offset your donation not being used for Lemmy development contributions.

It’s not worth the effort to set up independent purpose-bound donation pools to separate donations from people unwilling to support certain people/organizations, as the end result would be the same, except a lot more maintenance/accounting overhead. Calculating with the current expenses of €1,700/month, this is less than 3% of the total donations we’d need to cover our bills.

that’s not a thing on lemmy. it’s probably just the user setting to hide bot accounts.

which client are you using?

The account was already banned 10 days ago, if you look at https://lemmy.world/u/NoMoreTrds@toast.ooo you’ll see it’s gone with no content remaining.

Looking at the account on toast.ooo, https://toast.ooo/u/NoMoreTrds, it seems that the home instance ban may not have removed all content, that’s something @grant@toast.ooo should be able to fix by banning the user again with content removal selected.

Additionally, the bans/removals don’t seem to have properly federated to reddthat.com as seen on https://reddthat.com/u/NoMoreTrds@toast.ooo. This should either get fixed once the user gets properly banned from toast.ooo or can be fixed by a ban from reddthat.com e.g. by @ticoombs@reddthat.com, @example@reddthat.com or @Nankeru@reddthat.com.

if they’ve posted or commented in the community, visit the post/comment and it’ll be in the menu, same place where you’d be able to report/remove/etc.

if they haven’t posted or commented, it’s not possible in the default lemmy ui, but you can use e.g. tesseract or photon to do so in the community settings.