This repository has been archived by the owner on Apr 26, 2024. It is now read-only.
joining a large room completely hosed a homeserver #3527
Labels
A-Federated-Join
joins over federation generally suck
A-Performance
Performance, both client-facing and admin-facing
O-Uncommon
Most users are unlikely to come across this or unexpected workflow
S-Major
Major functionality / product severely impaired, no satisfactory workaround.
T-Defect
Bugs, crashes, hangs, security vulnerabilities, or other reported issues.
Comments
|
we fixed this by kicking the user back out of the room and restarting the joining server. |
|
Once the get_missing_events fails (as it will, thanks to #3444), we end up doing a /state_ids/ request instead, and then hundreds of thousands of /event requests. |
neilisfragile
added
p1
z-major
erikjohnston
added
A-Federated-Join
kittykat
added
the
T-Defect
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
somewhat related to #3495, but seen from the other end. The joining server thought the join failed, but other servers in the room think it joined successfully, and are now sending it events. The joining server, however, is missing a bunch of important stuff, so is sending off millions of
get_missing_eventsrequests, which are slow (cf #3526), end up getting queued (cf #3524), and 429ed, and federation basically ends up breaking downThe text was updated successfully, but these errors were encountered: