a light weight and compatible social networking server alternate to is now available on
It is high-performance and can run on small devices like a


What is Pleroma?

@anmol @yunohost there's already a pleroma package (not official) for yunohost.

developed by @ilja and I forgot where is the repo

I did not find any latest repo of for 3. So I did one.
But it would be good to see another package and compare it, so that I can fix anything I have missed.

@anmol (my instance is running on this package, just wait until @ilja answer for a link :3)

Can you show your me ?

$ cat /etc/yunohost/apps/plerome/status.json

@anmol Sorry I cannot now. I'm on my phone. I will link you to the package tomorrow If I don't forget

@brunpi @marsxyz
The package looks good and well updated. It has the media cache support too. If this package passes the CI then it should be mirrored to yunohost-apps github repos page.

@anmol @marsxyz @brunpi Yup that's the one. It's not really finished tho. I'm happy with the install scripts now (untill recently it wasn't possible to install multiple instances since it used default names for a whole bunch of things). The remove works, but needs extra work. backup and restore _kinda_ works but you have to remove the app before restoring works and I don't thinks that's supposed to be the case+I don't think you'll be able to fully restore to a fresh ynh install. And the upgrade is very naive atm.

I've never done anything like this before (I've never even set up a webapp before ^^' ) which is why it's taking me so long. Feel free to clone or MR or copy or whatever tho. It's AGPL and so is yours I see.

I have done some update. The app should work as multi-instance. The install,remove,update,backup restore work. I have taken option for open registration from your repo. But I don't understand how cache works in your repo. For me it gives error when nginx restarts. I have not take scope option from your repo. Is scope important option?

@anmol Setting scope to true allows _direct_, _followers only, __unlisted_, _public_ messages be sent from Pleroma-FE. It also allows you to use subjects in your posts. By default this is set to false, which means people can't do it from the Pleroma-FE. Personally I set it to true (but left it as an option in the package). The reason why it's set to false by default is to be compatible with the GNU-social backend, but that's not an issue in this case, so I think having scopes on is better.

The cache in the nginx is something I did with trial-and-error (and eventually trial-and-succes :p ). You need to add a line in the config of the app, but also in the config of the domain. In my package that's and

I have added scope by default. I think most of user will want it.
I tried to put cache as your your repo, but nginx gives me error. I don't know what how your config is working.

I tried your repo but it gave me error while installing dependencies.

@anmol What's the error? Can you try a second time? In the past I had that installation failed the first time and it was also during dependencies. I never was able to figure out why it was, but it was only on my LIME2 and there where other things wrong on that one that had nothing to do with the package so I didn't look into it further

Here is the error message:
99773 WARNING Unable to install dependencies
99773 DEBUG libgtk2.0-bin libgtk2.0-common libnotify4 libwxbase3.0-0v5 libwxgtk3.0-0v5
99773 WARNING !!
99773 DEBUG notification-daemon
99773 WARNING pleroma_ynh's script has encountered an error. Its execution was cancelled.
99774 DEBUG Suggested packages:

I think there is problem in dependency.

I have following dependencies which work.

git build-essential postgresql postgresql-contrib openssl g++ apt-transport-https erlang-inets elixir erlang-dev erlang-parsetools erlang-xmerl erlang-tools

@anmol I have esl-erlang as well and it's directly dependent on at least some of the packages I see in your output. I also don't see it listed as a dependency on so I'm not really sure where I got it from...
I'm installing a new VM so I can try the install on a new machine. I'll try without esl-erlang as well. If it works without, I may well just drop it completely. Besides that one there aren't dependencies I have that you don't.
You do have some that I don't; openssl, g++ and erlang-inets, but I don't think that'll be the problem.

I test the repo and nginx don't give error now if I put the cache-plume.conf to /etc/nginx/conf.d/. But is the media chache working this after the nginx success? What link you have in base_url in prod.secret.exs? Should we have option for enable or disable media cache while installation?

@anmol Whoops, enabled is false on my server, I should change that πŸ˜… I have no idea how to test this tho so I don't really know if it really works. I'm also not sure if it's a good idea to leave the default value of _max_size=10g_ since many people will install it on small servers. The base_url is commented out.
Not sure if it should be an option. I think it's not a bad thing to have in general?
@marsxyz @anmol
ls /tmp/pleroma-media-cache/ Just gave me an empty folder ./temp
I set enable: true
restart Pleroma (takes a while since it compiles again) and now I have a whole bunch of stuff in /tmp/pleroma-media-cache/

Not sure if this is a good way to test (not something I want to put a lot of time in now either tbh), but it seems to do something at least. I should set it to true explicitly on install tho

Can you show your setting of pord.secret.exs with key and database setting removed?

@anmol There's a whole bunch of commented lines as well, I removed them here. The quarantined_instances, rewrite_policy and content of reject was added by me manually.
@anmol where are the source of pleroma that yunohost installed. I need to change configuration ?

The sources are in /var/www/pleroma/pleroma/. Sorry for late reply I did not saw your message.

Have you installed the from ? I can not see your comments on the network.

@anmol Yes with yunohost. I already test with osada and it works for me

Ok thanks for testing the connectivity. Maybe there is a problem on my instance.


By default each instance is also a . So to follow one, you can execute MIX_ENV=prod mix relay_follow URL.
Some instances with lots of users are:

Sign in to participate in the conversation
Datamol Mastodon is one server in the network