Other than SQL queries being slow, there hasn’t been much. I’m trying to make as much noise as I can about actual data missing between servers and that ‘nginx 500’ errors are happening on the major Lemmy instances on a consistent basis. https://lemmy.ml/post/1453121
Should the front end be waiting on a response from the server in this case? Seems like an interaction like this should be fire/forget and just automatically updated on the UI side.
I mainly use jerboa and it seems like this is how it’s done as likes feel instant
Should the front end be waiting on a response from the server in this case?
yha, it is unusual. But I like it at this point, as it exposes the general problem. It isn’t trivial to populate the database (there is no current back-fill procedure for new instances) and user traffic to match the big sites (Beehaw, Lemmy.world, Lemmy.ml) - so at least I can get a sense of what ‘standard performance’ is considered.
It seems slow to me too. I haven’t seen enough discussion on why…
Is there an issue up on github for this? I’ll check later today
Other than SQL queries being slow, there hasn’t been much. I’m trying to make as much noise as I can about actual data missing between servers and that ‘nginx 500’ errors are happening on the major Lemmy instances on a consistent basis.
https://lemmy.ml/post/1453121
Should the front end be waiting on a response from the server in this case? Seems like an interaction like this should be fire/forget and just automatically updated on the UI side.
I mainly use jerboa and it seems like this is how it’s done as likes feel instant
yha, it is unusual. But I like it at this point, as it exposes the general problem. It isn’t trivial to populate the database (there is no current back-fill procedure for new instances) and user traffic to match the big sites (Beehaw, Lemmy.world, Lemmy.ml) - so at least I can get a sense of what ‘standard performance’ is considered.