As one Admin to another, I'm sad to hear that we're losing a good instance. And, I'm sorry that people ruined your fun.

Thanks for all you've done. All the best to you.

The key word here is "connected" instances. You don't find hashtag-related posts from all the servers, just the servers that your server is connected to and also only posts that your server has seen from other servers.

This is why, on my instance, I relay with over 300 other instances, some of them are the larger ones. All the posts they see on their server are duplicated onto my smaller server. The impact is that while my server is small, it has the same messages as one of the large servers which makes for a better experience. Note that Mastodon.social does not relay with any other instance. Not that most instances could possibly hold all their posts and accounts anyway.

When I search for hashtags on my instance, I do get a lot of hits.

Many smaller servers do not relay as much as I do, or at all, because it requires disk space to store all the posts, images, and accounts that are streaming in constantly from all these other servers. I don't have these concerns because I have unlimited capacity. So, the experience will vary depending on which server your account is on.

You have to use hashtags on Mastodon if you want your messages to be seen because there is no AI to fill timelines. People follow hashtags and search for them. Then you'll get likes and boosts.

And, you control your own timeline by following hashtags, following people, and searching for hashtags. This way, unlike Twitter, your timeline only has posts that you are interested in. Once you get this set up, it's quite nice to only see posts that you like reading, without advertising and without posts meant to enrage you.

Also you can use https://fedi.directory to find great people to follow.

OMG. Thanks for posting. I run a Mastodon server as well and, oh my, the Mastodon Developers would never have been willing to help me. A wild tale. Glad the fix ended up in the next release. Thanks for posting!

There were two screen shots. I saw them. At least one of them was doctored. That is all anybody knows. How are some people so certain which Admin doctored the screen shot?

@Chozo Interesting thought

@SamXavia I think your friend is a looney conspiracy theorist

Yep. I am now pre-emptively limiting threads.net on my Mastodon instance. No current plans to block but I have a feeling I'll never lift the limit.

Mine won't nor will any of my other 3 Fediverse servers.

I know of a number of Mastodon servers that have already de-federated or limited threads.net, even though it does not yet connect to the Fediverse. Some are even limiting or suspending connections to servers that refuse to de-federate from threads.net and are trying to pressure other servers to do the same.

An Admin has no right to force their personal agendas onto all the people who are on their servers. People are competent enough to make their own decisions and can individually decide to block or limit Threads. I block servers on my server to protect members from hateful people.

I will limit threads.net if their moderation is inadequate, just as I do now for a number of Mastodon servers that don't do much to keep hatred and offensive content off their servers. This won't prevent anyone from following someone or being followed by someone, on threads.net. It just means that people on my server need to approve being followed and that posts from threads.net won't show up in the public timeline.

At this point, I haven't heard of any Kbin instances planning to de-federate but there's a ton of yacking about it on Mastodon. I finally muted the "threads" hashtag to get some peace from it.

Love it!

@stackPeek You are currently on kbin.social so I assume kbin.social has been suspended (a.k.a. de-federated or blocked) from that server. On Mastodon, "Suspended" means totally blocked. You cannot follow anyone, they cannot follow you. All communication, including API requests, are blocked.

A Mastodon instance blocking kbin.social, is, to me, irrational behavior by the Admin and an extreme disservice to those on the server.

Can you share the name of the Mastodon instance?

@1chemistdown ActivityPub is a public protocol just like every other Internet protocol. Individual server Admins, whether they run KBIN, Lemmy, Mastodon, Calckey, PixelFed, PeerTube servers, are free to block any other server that connects to ActivityPub. You have to ask the Admin of the server you're on or interested in joining.

My last edit removes the markdown interpreter from messing up the text. I'm not sure if you saw it before I added the escape to make it more readable.

Anyway, I hope it goes well for you!

• I ignored the thing about kbin.localhost and everything on https://feddit.online is working well, also in my test environment. So, you can probably skip it too.

• The "var" folder will not be found in the kbin folder so the chown will not work. What I did, and from other comments that I've read, I think this is what it should have been, "sudo chown 82:82 /var"

Here's my .env file without the secrets.
Where I mention a long secret, you can use this Linux command to generate a strong 32-character secret
openssl rand -hex 32

kbin variables

SERVER_NAME="feddit.online"
KBIN_DOMAIN=feddit.online
KBIN_TITLE=/kbin
KBIN_DEFAULT_LANG=en
KBIN_FEDERATION_ENABLED=true
KBIN_CONTACT_EMAIL=admin@feddit.online
KBIN_SENDER_EMAIL=notifications@feddit.online
KBIN_JS_ENABLED=true
KBIN_REGISTRATIONS_ENABLED=true
KBIN_API_ITEMS_PER_PAGE=25
#Frankly unsure if the next line does anything
KBIN_STORAGE_URL=https://media.feddit.online
KBIN_META_TITLE="Kbin Lab"
KBIN_META_DESCRIPTION="content aggregator and micro-blogging platform for the fediverse"
KBIN_META_KEYWORDS="kbin, content agregator, open source, fediverse"
KBIN_HEADER_LOGO=false
KBIN_CAPTCHA_ENABLED=true

Redis

REDIS_PASSWORD=password_here_that_you_make_up
REDIS_DNS=redis://${REDIS_PASSWORD}@redis

S3 storage (optional) THIS DOES NOTHING. TURNS OUT IT ONLY WORKS WITH AWS S3 AND I WANT TO USE WASAB

SO THESE SETTINGS ARE NOT USED

S3_KEY=<key>
S3_SECRET=<secret>
S3_BUCKET=media.feddit.online
S3_REGION=us-east-1
S3_VERSION=latest

oAuth (optional)

OAUTH_FACEBOOK_ID=
OAUTH_FACEBOOK_SECRET=
OAUTH_GOOGLE_ID=
OAUTH_GOOGLE_SECRET=
OAUTH_GITHUB_ID=
OAUTH_GITHUB_SECRET=

###> symfony/framework-bundle ###
APP_ENV=prod
APP_SECRET=427f5e2940e5b2472c1b44b2d06e0525
###< symfony/framework-bundle ###

###> doctrine/doctrine-bundle ###

Format described at https://www.doctrine-project.org/projects/doctrine-dbal/en/latest/reference/configuration.html#connecting-using-a-url

IMPORTANT: You MUST configure your server version, either here or in config/packages/doctrine.yaml

==========

POSTGRES_DB=kbin
POSTGRES_USER=kbin
POSTGRES_PASSWORD=password_here_that_you_make_up
POSTGRES_VERSION=13
DATABASE_URL="postgresql://${POSTGRES_USER}:${POSTGRES_PASSWORD}@127.0.0.1:5432/${POSTGRES_DB}?serverVersion=${POSTGRES_VERSION}&charset=utf8"
###< doctrine/doctrine-bundle ###

###> symfony/messenger ###

Choose one of the transports below

RABBITMQ_PASSWORD=password_here_that_you_make_up
MESSENGER_TRANSPORT_DSN=amqp://kbin:${RABBITMQ_PASSWORD}@rabbitmq:5672/%2f/messages
#MESSENGER_TRANSPORT_DSN=doctrine://default
#MESSENGER_TRANSPORT_DSN=redis://${REDIS_PASSWORD}@redis:6379/messages
###< symfony/messenger ###

###> symfony/mailer ###
#MAILER_DSN=smtp://localhost
###< symfony/mailer ###

###> symfony/mailgun-mailer ###

MAILER_DSN=mailgun://KEY:DOMAIN@default?region=us

#MAILER_DSN=mailgun+smtp://postmaster@sandboxxx.mailgun.org:key@default?region=us
MAILER_DSN=smtp://notifications@feddit.online:<password here>@smtp.dreamhost.com:587
###< symfony/mailgun-mailer ###

###> symfony/mercure-bundle ###

See https://symfony.com/doc/current/mercure.html#configuration

The URL of the Mercure hub, used by the app to publish updates (can be a local URL)

MERCURE_URL=https://feddit.online/.well-known/mercure

The public URL of the Mercure hub, used by the browser to connect

MERCURE_PUBLIC_URL=https://feddit.online/.well-known/mercure

The secret used to sign the JWTs

MERCURE_JWT_SECRET=long_password_here_that_you_make_up

Docker

CADDY_MERCURE_URL=https://feddit.online/.well-known/mercure
CADDY_MERCURE_JWT_SECRET=long_password_here_that_you_make_up
###< symfony/mercure-bundle ###

###> nelmio/cors-bundle ###
CORS_ALLOW_ORIGIN='^https?://(kbin.localhost|127.0.0.1)(:[0-9]+)?$'
###< nelmio/cors-bundle ###

###> symfony/lock ###

Choose one of the stores below

postgresql+advisory://db_user:db_password@localhost/db_name

LOCK_DSN=flock
###< symfony/lock ###

###> lexik/jwt-authentication-bundle ###
JWT_SECRET_KEY=%kernel.project_dir%/config/jwt/private.pem
JWT_PUBLIC_KEY=%kernel.project_dir%/config/jwt/public.pem
JWT_PASSPHRASE=
###< lexik/jwt-authentication-bundle ###

###> meteo-concept/hcaptcha-bundle ###
HCAPTCHA_SITE_KEY=<key>
HCAPTCHA_SECRET=<secret>
###< meteo-concept/hcaptcha-bundle ###

I agree with all that you've said

For others reading this, there are 2 others that I know well: https://feddit.online and https://fedia.io

You can find bunches here: https://kbin.pub/en

RE: it seems like the only way to have a reasonable chance of getting decent results for hashtag searches is to be on the biggest server

Well, yes, and no. You can relay with other Mastodon instances and they will share there traffic with you so that people on your instance will have the same content as on their instances PLUS the traffic from all the other instances that are relaying with them as well. So, your server will see a ton of stuff. No shortage of stuff coming in

There's a number of good relays. I recommend these two at a minimum:

https://relay.infosec.exchange/inbox
https://bigrelay.social/inbox

In the Administration section, there's a Relay page where you can put these in.

For the first one, @Jerry is the person to contact if it doesn't enable for you. Not sure how many instances relay with him, but it must be a large number.

For the second one, which currently has 328 instances connected together, you can get more information at https://bigrelay.social/

Hope this helps

It's a closed instance and all their postings are done by their bots. I'm guessing they've chosen not to Federate and that they don't want anyone posting to their communities. Just a guess.

kbin.social must still be having severe problems. Ernest posted this a day ago:

    ernest@kbin.social

    We're experiencing massive delays on the queues today. We're actively working on a solution,
    and I think we're getting really close. Just wanted to let you know ;)

One thing you might try doing is switching to a different KBIN server. The smaller ones are probably Federating just fine with Lemmy and the other KBIN instances, although everyone will have trouble Federating with kbin.social.

I wonder why Ernest doesn't turn off new sign ups on kbin.social and ask people who want to join KBIN to do so on a different instance so the problems there aren't exasperated?

1. Before joining an instance you can go to their /terms page and see if they will tolerate you because you're just making stuff up and your ignorance will hurt and endanger people. Most KBIN and Lemmy servers will, fortunately, not willingly host you because of this.
2. I won't tolerate you, so don't join my instance. So, 1 down.