Kbin and Lemmy seem to not agree on time zones, there has been some mention on GitHub.
I’m trying to work on Lemmy performance and scale issues. I’m the same person as RoundSparrow @ Lemmy.ml instance.
Kbin and Lemmy seem to not agree on time zones, there has been some mention on GitHub.
Major performance problems have been fixed in Lemmy in the past 48 hours, with more pending in the next 24. The latest code on GitHub is far better than 0.17.4/0.18.0 in terms of hammering the server.
The admin of Beehaw told me to post it on Lemmy.ml despite the message being that the server was down.
Beehaw is suppressing the news about a Lemmy outage, when their /c/Technology community is about rumors in the sidebar!
removed by mod:
Lemmy 0.18 (and earlier) has no ability to save the federation queue, when the server stops it forgets what it had in retry to send. I suspect the problem is more the other servers going down causing lemmy.ml to keep crashing on outbound content it is sending. There are hundreds of small servers out there subscribed to lemmy.ml’s communities.
@lixus98 I can’t reply to your comment for some reason… here is my reply
There is a community on the enterprise server we started putting them in. The markdown rendering you linked has been mentioned: https://enterprise.lemmy.ml/post/416737
Keep in mind that the outbound comment/post sending queue is in RAM only and will be lost in 0.17.4 with a restart
An internal crash was identified in Lemmy and 0.18 should have fixed it. https://github.com/LemmyNet/lemmy/issues/3076
Good news, the new version, 0.18 is in beta testing today. Hopefully it comes soon.
I’m seeing 9 days old, it’s really stuck.
There is an open bug about internal crashes within lemmy_server: https://github.com/LemmyNet/lemmy/issues/3076
A restart of the lemmy_server service might get it fixed for one-time.
Quite many comments that were over an hour old were missing. I guess there’s a problem with the federation.
I opened an issue on Github about the replication failures of posts and comments between servers. It is not really getting much attention: https://github.com/LemmyNet/lemmy/issues/3101
deleted by creator
I think it’s just a matter of time before there are alternate-federated networks for Lemmy… different lists of federated servers that the install peers with.
I’m just curious what my fellow sh.it.heads think of this development in the fediverse, any input is appreciated!
I think performance problems with the Lemmy database lead to a lot more fragmentation… it didn’t scale well and instances were telling people to spread around. That and the discovery of remote instances being unintuitive - plus the data replication problems (see: https://github.com/LemmyNet/lemmy/issues/3101) has laid a rather chaotic foundation. People are still trying to get their server installs up and running today, there are a huge number of servers, and I expect when those severs go offline there will be a mess left over of orphan posts, comments, etc.
I don’t think anyone planned it this way, they were working for several years to create an alternative to Reddit and you find communities with mods who created them 3 years ago and stopped using Lemmy… it will be an interesting year, for sure.
Lemmy is having growing pains and there are problems with internal tasks not updating the database correctly plus I’m sure individual instances would want to change some of these parameters to define what is ‘active’ in terms of age criteria, etc.
The lemmy_server code is not scaling well, lots of database contention.
Are we blocked from editing communities that we created? I tried to revise the sidebar of https://sh.itjust.works/c/jukebox and the “save” network works.