qlog
qlog copied to clipboard
quicwg
The IETF I-D documents for the qlog format
When looking at a qlog at this time, if there is a certain event type missing, you don't know if it's because those events never occurred or if the implementation...
Depending on some TLS parameters, things might behave strangely on a QUIC level. For example, TLS key shares can apparently use different "group types", e.g., p-256 or x25519 (this is...
@marten-seemann asked how to best approach logging from a server's perspective, given that things like version negotiation and stateless retry are not inherently tied to a single connection. We should...
The spec is not clear on when to log key updates. I'm trying to debug a failure of the key update interop runner test case, and qlog just doesn't give...
On a production system you probably don't want to log TLS secrets, even if you qlog (some of the) connections. The `new` field in the `key_updated` event therefore should not...
Currently, we have few options for logging a lot of TLS related data (e.g., cipher suites, supported groups, (E)SNI, ...). Need to figure out a generic form for TLS (extensions)...
The following is a 'bag' of general comments ... perhaps not so actionable but I did not find a mail list to send these kind of general observations. **scope** Some...
If qlog is used on a proxy, the question becomes what vantage_point it should use for its traces, since it's now simultaneously accepting and opening connections, and thus behaves as...
The current text lists about 8 reasons for dropping packets. Microsoft's implementation lists 60+ individual reasons (via @nibanks) Some of those: > LogDrop("Retry sent to server"); LogDrop("Already received server response");...
This is just a start. While these are editorial changes they will definitely require discussion for how it's best to change each Note:.
