We need to deploy a new packet format again. This will be neg type 6. The immediate cause is that the HMAC for each packet is too short. We agreed on at least 10 bytes, it is still using 4. Are there any other important changes that should go in before the new neg type is deployed?
One thing that was proposed was to record how long a packet had been delayed in coalescing for. This would allow us to get an accurate round-trip time. The downside is that if CPU usage is high packets can be delayed for longer than the maximum - so we'd still need to average both parts for purposes of timeouts etc. So maybe there isn't much point? -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 197 bytes Desc: This is a digitally signed message part. URL: <https://emu.freenetproject.org/pipermail/devl/attachments/20101231/cc1793d3/attachment.pgp>
