It’s usually the cache after the dns making you think it was the cache all along when it’s just still hanging onto messed up dns data
Just some IT guy
It’s usually the cache after the dns making you think it was the cache all along when it’s just still hanging onto messed up dns data
Just my two cents but if you decide to go for the self hosted GitLab approach I think Forgejo might be a better fit. It’s not as resource intensive as GitLab is but has all of the essential features you’d need from a forge.
Tempo is a really good Navidrome Client for Android imo
Navidrome is a subsonic server, feom the cursory research I did before setting it up it is also among the best supported/developed ones available.
The developer working on federation plans to merge the changes into forgejo first and then from there into gitea but I’m not sure in how far the recent changes to gitea’s CLA have affected those plans.
Forgejo is a drop in replacement (they are committed to keeping it that way for as long as possible) so, as far as I know, simply changing the gitea image to the forgejo image is all you would need to do.
They did start a cloud service for hosting Gitea which introduces a direct incentive for them to make Gitea less hosting friendly by, for example, making newly added configuration options less comfortable to set up. And more recently some changes to code contributions that are not exactly community friendly (as a result forgejo will be unable to upstream some of their changes)
What lead to Forgejo, as far as I am aware, was less a problem that is already there and more the set of problems that have a very high chance of eventually manifesting, at which point forking the project would be too late.
This is absolutely awesome, I love it already. Substreamer had some real annoying quirks but it was the least worst option I found so far, this is better in almost every way for my use case
Gitea is managed by a for profit which is now offering a hosting service. That alone is already a conflict of interest because one of Giteas core features is the easy self hosting.
Then the contribution guidelines have been made stricter, anyone contributing now has to give up their copyright to the gitea management, meaning they could change the opensource license to a stricter one down the line without requiring community consent.
The concern is that as time passes features will be locked behind a premium tier for self-hosters or the self-hosting itself will be made more difficult in an effort to push their cloud service.
Due to some concerns about Gitea’s future I would recommend Forgejo instead. It’s a drop-in replacement with less concerning contribution policies and management structure.
You need some paragraphs my dude
I had to constantly skip back and forth to figure out where one word starts and another ends. Very painful to read, I’ve found new appreciation for whitespace.
The meme is the how the packages are managed to begin with.
Worked with Maven once, I’d rather saw my legs off than do it again.
Honest question: What is/are your problem(s) with Svelte? So far it seems a lot easier to use than react to me but I wouldn’t consider myself experienced so there might be unwelcome surprises waiting.
yes but most isp’s (at least where I live) change the IP in the middle of nobody-should-be-awake-anyway o’clock
Plus the connection drops only for a minute or two at most unless ypu set a really high ttl on the domain dns record
wait Jira is a Java website?
That explains so much…
On that note: Why is it every single time a piece of Software I find runs or feels like absolute Garbage it turns out it was written in Java? (Ok all things being fair occasionally the ancient PHP App is in the mix as well but it’s 95% Java Apps being shit)
As someone with a beef server: Nope, performance stays unsatisfactory. Redis helps a lot but only if the page is cached which tbh just makes the experience worse if the page isn’t cached
Edit: I’m using the AIO installer though, as discussed elsewhere in this post that might be the root cause of the poor performance, will check on the weekend by installing nextcloud manually in a fresh vm
Sounds easy enough to implement tbh, will maybe take a look later how much work that’d be because it’s a must have for me as well
A lot of smaller things, at least for me. The biggest grievance I have with it is the garbage tier UX between hitting “Deploy” on a stack and getting ito to do so. Error messages in the notification bubble get cut off, are unhelpful amd/or disappear too fast. That and the L9g Voewer sucking ass are my main problems with it and why I’ll definitely check this out.
if their service runs as poorly as their website I’ll give that a pass
The lemmy docs are all a mess. Try writing something that uses the lemmy api and you start crying because looking up the endpoints in the code tells you what it does faster than their ‘documentation’