I’ve once again noticed a delay of 10-12 hours with Lemmy.world. I just wanted to make sure someone’s aware.
Glory to your houses, admins, and thanks for maintaining this instance.
As ValueSubtracted said, as far as we can tell, the problem is on Lemmy.world’s side as we are not seeing any major disruptions with content from other instances. But we are looking into options on our end.
EDIT: We’ve made some adjustments that appear to be closing the gap
EDIE 2: The issue has been resolved
We’re looking in to it, but I’m not sure when a solution may be found. It’s actually worse than you know when it comes to lemmy.world content making it to us - think days, not hours.
Hey just want to update you on this, we made some changes that appear to be working. We’ve gone from 1.47 weeks behind to 1.42. You can see the recent change here.
I appreciate the transparency, but fucking hell, are we pariahs?
Recently I wanted to make a comment, but I wasn’t logged in (I clicked a link that took me to a different instance and did not realize it).
I came back here , searched for the community, I sort by new to find the post… and it’s gone.
The irony is that I wanted to reply to some guy dunking on how our instance is a ghost town (was gonna reply with a pic of the Q Continum and joke that not all ghost towns are as they seem)
But since that moment a few days ago I am noticing it constantly.
Sucks feeling like Ro and LaForge being out of phase with the
enterprisemajority of the fediverse. It’s really taken the wind out of my sails the last few days.Crestfallen rant aside, I’m a coder by day I would like to know if I can offer any generic technical help (I don’t know lemmy, but maybe I can help debug the issue 🤷♀️)
Again thanks, sorry if I’m coming off angry I’m just disappointed and using your comment to get it out.
dunking on how our instance is a ghost town
Not really your point, but I can’t help but have some thoughts on this:
-
Ultimately, the instance is exactly as active as its users want it to be, for better or worse.
-
If we were overly concerned about size, we’d have stayed on Reddit.
Ya you’re right. Just seeing it in chart format triggered an some kinda emotional response and I just sorta vomited a reply. Sorry about that.
I just came for a good time and I’m having one, even if it’s not perfect.
For sure, I get it. I wish .world federation was more reliable, but we seem to run into challenges a couple times per year. We’re always looking to improve stuff like that.
-
What was the community/post? It’s possible we removed it on our side (Also we are up to 4 days behind now, progress!).
Personally I don’t prefer the .world versions of most communities so I haven’t noticed much difference.
It’s not important, I just allowed myself get worked up over an annoyance. Sorry to show up with a sloppy rant, I’m grateful to be a guest here ✌️
No worries at all, we want to make sure we’re providing a fast and reliable (well, as fast and reliable as is possible with beta software at least!) experience. Like I said I don’t notice .world missing very much but that doesn’t mean we don’t realize the value their communities provide to the overall ecosystem.
I just want to provide a brief update on this.
It’s still happening. As near as we can tell, there’s nothing we can do on our end. Federation from .world is just ~1.5 weeks behind, but it’s still happening.
Is this because their Lemmy version is ancient and they refuse to update?
I’m not really interested in casting blame - the Fediverse is complicated, and can be affected by any number of things.
All I can say is that we haven’t been able to find an issue on our end, and at the same time, federation doesn’t seem completely broken - just about 10 days slow, and only in the one direction.
I’m not really interested in casting blame
A question about version numbers isn’t really blame, though. Maybe it’ll be better when Lemmy 1.0 is out and LW finally updates.
Looking at our metrics, there’s a chance things are starting to improve now. It’s too early to know for sure, though.
It’s been a long road, but this issue is finally resolved.