The great Twitter migration
-
Who else has spun up a Mastodon instance recently?
I'm still looking to extend my network, while it's easy to run your own server (especially with Cloudron), the social factor is a bit lacking when everybody has their own server instance.
-
@fbartels yep, I'm @jdaviescoates@social.coop https://social.coop/@jdaviescoates but I have also spun up https://safejust.space in the last couple of days too
where I'm also running @doughnuteconomics@safejust.space https://safejust.space/@doughnuteconomicsOut of interest, who here has already heard of Doughnut Economics/ read the book/ is actively working to humanity into the safe and just space of the doughnut?
-
I've got social.futurnumerique.com/@shane there. I've had it for awhile now though. Just deciding whether to open it up to registrations or not.
-
@jdaviescoates said in The great Twitter migration:
also spun up https://safejust.space in the last couple of days too
In case anyone tried to visit that very recently and found it down... I ran out of diskspace!
Seemingly diskspace runs out quickly when running Mastodon!
I've just unfollowed everyone I was following as the admin @doughnuteconomics@safejust.space account on there in the hope that'll slow down how quickly if fills up!
-
@jdaviescoates yes I saw your toot and it made me curious as well (we are already writing on Mastodon but repeating here for others as well).
The main culprit seems to be the cache directory. On my instance with only my user (who is following 64 accounts on other servers) this already amounts to close to 7gb of cache data. Main offender are image attachments in toots, but also profile pictures and banners.
Kind of off topic, but @girish @nebulon is there a way to exclude folders from the backup?
The good news is that the app already removes files older than seven days (although the cutoff date does not seem configurable) https://git.cloudron.io/cloudron/mastodon-app/-/blob/795983fa31c4170f2c502a3c1f73aa830c3d2e52/CloudronManifest.json#L22
-
@fbartels I remember we hit this issue on our own instance as well. Might be worth to investigate again if the upstream app has now a bit better config options to handle the media caching.
For the off-topic question, volumes could be used there, as they are not backed up together with an app, however for this then we would have to somehow make it possible to configure the caching folder.
-
@nebulon said in The great Twitter migration:
Might be worth to investigate again if the upstream app has now a bit better config options
This sadly does not seem to be the case. At least I could not find any switches to purposefully disable caching, but on the other side the current behaviour is probably what is desirable for larger instances, but just not really optimal for one user ones.
Maybe I need to externalise these to Minio.
-
@fbartels and others. If you look in the Minio topic of the forum you'll find some posts from me about using Minio with Mastodon. There is simply no other way to do this while on Cloudron, unless you have a huge VPS and run only Mastodon. If you federate with your instance, you are going to rack up a lot of data being stored. I'm going to post something about this on Mastodon for all those new users. Most I think have signed up on someone else's instance, and whoever runs those probably knows about the storage demands. But the new self-hosters are in for a shock after a few weeks of federation fun!
-
@scooke thanks. I found your instructions on https://forum.cloudron.io/post/50215
Yes, those storage heavy apps definitely benefit from blob storage like s3/minio. I just did not expect that much of storage was needed for mastodon, even if it at max caches 7 days.
-
This is just like so many similar situations of the past where people get spooked, rush to something different but ultimately revert back.
Much like the cPanel issue where so many said cPanel would die because of the pricing hikes. Never happened. Quite a few went to DirectAdmin but it was the lowend budget hosts that had to move to something cheaper.
Same thing here, people appear to switch to Mastodon but they soon realise it's not all that and most have likely already gone back to Twitter.
-
@LeeW The masses, sure. But the topics I'm interested in... those posters and their instances are going strong. It is sooooo nice to follow these accounts (whose account at a given instance reveals all the other cool accounts) and get a nice stream of interesting, pertinent info and shares.
-
Mastodon 4 just released! Excited for the update to flow through (cc: @girish ) - do note there appears to be a two step upgrade process.
Also - for folks running their own instance, please do feel free to set up to relay against my open relay I offer as a community service if you like - it will help populate your federated timeline nice and quick! (https://mastodon-relay.thedoodleproject.net).
-
@doodlemania2 That's a nice thing to do - helps a lot in the early days of a new instance
-
F fbartels referenced this topic on
-
I'm running my own and you can find me at https://social.shas.am/@shanelord
I like the idea that I'm not reliant on others to maintain the site, although it is a bit quieter with only me as a user.
Anyone have any relays they recommend that are currently open for sign-ups?
I'm using:
https://relay.berserker.town/inbox
https://relay.libranet.de/inbox
https://relay.mistli.net/inboxNot sure if they're all still open, but I'm finding it hard to find others to strenghten the federated feed.
I have also added @doodlemania2 https://mastodon-relay.thedoodleproject.net/inbox also, but it doesn't seem to have completed connection as it still says waiting for approval.
Thanks,
Shane. -
I have also added @doodlemania2 https://mastodon-relay.thedoodleproject.net/inbox also, but it doesn't seem to have completed connection as it still says waiting for approval.
This is now working.
-
@shanelord01 (waves) it auto-approves after a few minutes assuming your side is set up cleanly - I see you and welcome!
Many people have tried to relay over the past few weeks. Some don't have SSL (hard pass), some are using fake FQDN (hard pass), and then of course, we had a few bad actors come along and I had to boot em.
Be mindful of your blocks and mutes - whole instance blocking is a big sledgehammer unless it's just a whole instance of bad. Also, if you see someone on my relay and you're like, ewe, Derek should know about that, please ping me (at)derek(at)toot.thedoodleproject.net - I am watching but it's a LOT to watch and any visibility on bad instances is appreciated - let me know and I'll block em at the relay.
