In this post I will list the known issues or possible improvements for Lemmy.world.
Please comment with any issue or area for improvement you see and I will add it here.
Remember: this instance was only started June 1st so a lot of troubleshooting and tweaking to be done.
Issues can be:
Local (lemmy.world) (also performance issues)
Lemmy software issues
Other software related (apps/Fediverse platforms etc)
Remote server related
(User error? ...)
Known issues
Websockets issues
There are some issues with the Websockets implementation used in Lemmy, which handles the streaming. Websockets will be removed in version 0.18 so let's hope these issues will be all gone then!
Top posts page gets a stream of new posts ? Websockets issue
You're suddenly in another post than you were before > Websockets issue
Your profile will briefly display another name/avatar in the top right corner
Spinning wheel issues
Error handling is not one of Lemmy's strongpoints. Sometimes something goes wrong, but instead of getting an error, the button will have a 'spinning wheel' that lasts until eternity.
These are some of the known cases:
You want to create an account but the username is already taken
You want to create an account but the username is too long (>20 characters)
You want to create an account but the password is too long
You want to create a community but the name is already taken
You want to create a community but the name is not in all lowercase letters
You want to create a post over 2000 characters
You want to post something in a language that isn't allowed in the community
Other issues
Federation not always working; Apparently not everything gets synced all the time. This needs troubleshooting.
“404: FetchError: invalid json response body at http://lemmy:8536/api/v3/site” This sometimes happens when the Lemmy app container is very busy. Needs troubleshooting
Enhancement requests
Can themes be added? > To be checked if this can be done without changing code.
Probably an idea for the future, but a status page with upcoming maintenance or ongoing issues would help quell some of these kinds of requests. I assume you probably don't have much in the way of redundancy at the moment?
Well we're in the baby phase now, so most maintenance is ad-hoc. But after that it's a good idea.
About redundancy: it's a single VPS. But when the VPS or datacenter breaks, we have a backup, and it runs in Docker so it's just restoring the backup, starting the containers and redirect the DNS