You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This feels clunky, but probably gets the job done, with 1 proposed change:
Currently QuicVersion is defined as : hexstring and I would define it as : hexstring / "none" to support the mismatch case explicitly in that event as well (currently you'd omit the chosen_version field to signal the same).
Another change we could do is to add ? supported_versions: [+ QuicVersion] to the server_listening event to make it more explicit which versions a server supports at startup (currently no real way to indicate that). Not sure this is crucial though.
One of the main points however is that I'm not sure this:
allows logging of all the possible flows with the VersionNegotiation extension . Maybe someone who's implemented that logic can give some insight? CC @huitema
should even be an issue for the qlog documents (as we're leaning towards having separate qlog documents for separate QUIC features/extensions). CC @LPardue
The text was updated successfully, but these errors were encountered:
As noted in #138 (comment), version mismatch problems causing a connection close are logged in 3 ways currently:
version_mismatch
trigger value inconnection_closed
packet_sent
andpacket_received
.supported_versions
version_information
event grouping both external and internal (see below)This feels clunky, but probably gets the job done, with 1 proposed change:
Currently QuicVersion is defined as
: hexstring
and I would define it as: hexstring / "none"
to support the mismatch case explicitly in that event as well (currently you'd omit thechosen_version
field to signal the same).Another change we could do is to add
? supported_versions: [+ QuicVersion]
to theserver_listening
event to make it more explicit which versions a server supports at startup (currently no real way to indicate that). Not sure this is crucial though.One of the main points however is that I'm not sure this:
allows logging of all the possible flows with the VersionNegotiation extension . Maybe someone who's implemented that logic can give some insight? CC @huitema
should even be an issue for the qlog documents (as we're leaning towards having separate qlog documents for separate QUIC features/extensions). CC @LPardue
The text was updated successfully, but these errors were encountered: