lwIP - A Lightweight TCP/IP stack - Bugs: bug #51138, RTT measurement after...
You are not allowed to post comments on this tracker with your current authentication level.
bug #51138: RTT measurement after retransmission
Submitter: | Ambroz Bizjak <abizjak> | ||
Submitted: | Mon 29 May 2017 05:37:41 PM UTC | ||
Category: | TCP | Severity: | 3 - Normal |
Item Group: | Faulty Behaviour | Status: | None |
Privacy: | Public | Assigned to: | None |
Open/Closed: | Open | Planned Release: | None |
lwIP version: | git head |
Thu 01 Jun 2017 07:36:22 PM UTC, comment #3: |
Simon Goldschmidt <goldsimon> |
Mon 29 May 2017 06:02:14 PM UTC, comment #2: (sorry for the multiple posts)
|
Ambroz Bizjak <abizjak> |
Mon 29 May 2017 05:51:17 PM UTC, comment #1: ... actually I see that even my proposed solution (invalidate any RTT measurement on any retransmission) does not always prevent RTT over-estimation. Especially with fast retransmission. If a new segment is sent for the first time before the retransmitted segment is acknowledged, an RTT measurement would be started which could be over-estimated due to ACK being delayed due to the lost segment.
|
Ambroz Bizjak <abizjak> |
Mon 29 May 2017 05:37:41 PM UTC, original submission:
Hi,
|
Ambroz Bizjak <abizjak> |
No files currently attached
Depends on the following items: None found
Items that depend on this one: None found
There are 0 votes so far. Votes easily highlight which items people would like to see resolved in priority, independently of the priority of the item set by tracker managers.
Powered by Savane 3.13-4448.
Corresponding source code
Hmm, from reading your posts here, you might be right. I don't have an immediate answer right now though. I simply don't know which direction makes it worse (measuring RTT too slow or possibly not measuring RTT at all due to too many retransmissions)...