That was because of a a bug that no one pinpointed for a long time → until I disclosed it ← That’s why it’s now defused. But it led to a ban. It’s entirely possible that the administrators of apps.nextcloud.com installed a throtle mechanism instead of banning the ip for 24 hours. But it doesn’t have to come from serverinfo. I actually think that’s extremely unlikely in this case, since you have to call the info by the serverinfo api and it not hapens while a serverinfo tab is opened as it was before. But it can also happen if an update scan is done a lot - not only using this api - i.e. often in succession. But there is no real pattern discernible. I’ve already tried a lot of things by interacting with the server using the REST API.
The command line app store → nc-apps ← uses → http caching with etag ← , which ensures that only data that has changed is downloaded. This means that many requests can be made to the server within a few seconds without a ban or throttling occurring.
The server code normally uses this mechanism as well, but the bug still occurred…
But, as mentioned, it could also simply be a network bottleneck. There are millions of requests to the app store, so bottlenecks can occur that need to be balanced without any pattern being apparent to individual users.
Much and good luck,
ernolf