Like #43297 (closed), a regression from non-persistent background scripts causing race conditions on startup / updates.
Again the connection attempt is not going anywhere on the network, being to a local multicast address, but attempts are observable until the main process machinery meant to block / redirect it is ready, which on updates is not guaranteed to happen before child process frames are injected and start trying.
Probably fixable on 12.x convergence by using switching to a persistent Manifest V3 declarativeNetRequest redirecting rule, like in Chrome.
ma1changed title from Weird XHR connection attempt to multicast IPv6 ff00:::443 on NoScript update to Failing connection attempts to multicast IPv6 ff00:::443 logged during NoScript updates
changed title from Weird XHR connection attempt to multicast IPv6 ff00:::443 on NoScript update to Failing connection attempts to multicast IPv6 ff00:::443 logged during NoScript updates