fix: Don't encode large RTT guesses in tickets #2114
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Because under lossy conditions (e.g., QNS `handshakeloss` test), the guess can be multiple times the actual RTT, which when encoded in the resumption ticket will cause an extremely slow second handshake, often causing the test to time out. Broken out of mozilla#1998 Fixes mozilla#2088
larseggert
requested review from
KershawChang,
martinthomson and
mxinden
as code owners
Failed Interop TestsQUIC Interop Runner, client vs. server neqo-latest as client
neqo-latest as server
All resultsSucceeded Interop TestsQUIC Interop Runner, client vs. server neqo-latest as client
neqo-latest as server
Unsupported Interop TestsQUIC Interop Runner, client vs. server neqo-latest as client
neqo-latest as server
|
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #2114 +/- ##
=======================================
Coverage 95.35% 95.35%
=======================================
Files 112 112
Lines 36316 36334 +18
=======================================
+ Hits 34628 34647 +19
+ Misses 1688 1687 -1 ☔ View full report in Codecov by Sentry. |
Benchmark resultscoalesce_acked_from_zero 1+1 entries:time: [99.188 ns 99.537 ns 99.890 ns] Found 13 outliers among 100 measurements (13.00%) 10 (10.00%) high mild 3 (3.00%) high severe coalesce_acked_from_zero 3+1 entries:time: [116.94 ns 117.30 ns 117.68 ns] Found 15 outliers among 100 measurements (15.00%) 3 (3.00%) low mild 12 (12.00%) high severe coalesce_acked_from_zero 10+1 entries:time: [116.54 ns 117.73 ns 119.75 ns] Found 14 outliers among 100 measurements (14.00%) 4 (4.00%) low severe 1 (1.00%) low mild 2 (2.00%) high mild 7 (7.00%) high severe coalesce_acked_from_zero 1000+1 entries:time: [97.464 ns 97.579 ns 97.710 ns] Found 15 outliers among 100 measurements (15.00%) 9 (9.00%) high mild 6 (6.00%) high severe RxStreamOrderer::inbound_frame():time: [110.92 ms 111.06 ms 111.28 ms] Found 8 outliers among 100 measurements (8.00%) 7 (7.00%) low mild 1 (1.00%) high severe transfer/pacing-false/varying-seeds:time: [26.854 ms 27.948 ms 29.038 ms] Found 1 outliers among 100 measurements (1.00%) 1 (1.00%) low mild transfer/pacing-true/varying-seeds:time: [33.862 ms 35.583 ms 37.332 ms] Found 2 outliers among 100 measurements (2.00%) 2 (2.00%) high mild transfer/pacing-false/same-seed:time: [32.313 ms 33.059 ms 33.798 ms] transfer/pacing-true/same-seed:time: [40.343 ms 42.721 ms 45.125 ms] 1-conn/1-100mb-resp (aka. Download)/client: time: [114.10 ms 114.42 ms 114.74 ms]
thrpt: [871.52 MiB/s 873.96 MiB/s 876.43 MiB/s]
Found 1 outliers among 100 measurements (1.00%)
1 (1.00%) low mild1-conn/10_000-parallel-1b-resp (aka. RPS)/client: time: [313.76 ms 317.20 ms 320.61 ms]
thrpt: [31.190 Kelem/s 31.525 Kelem/s 31.871 Kelem/s]
Found 1 outliers among 100 measurements (1.00%)
1 (1.00%) low mild1-conn/1-1b-resp (aka. HPS)/client: time: [33.876 ms 34.118 ms 34.381 ms]
thrpt: [29.086 elem/s 29.310 elem/s 29.520 elem/s]
Found 9 outliers among 100 measurements (9.00%)
6 (6.00%) high mild
3 (3.00%) high severeClient/server transfer resultsTransfer of 33554432 bytes over loopback.
|
|
This needs more work. I thought the code here neqo/neqo-transport/src/path.rs Lines 971 to 990 in 75372c2 was an optimization, but the idle_timeout_crazy_rtt test hangs when I remove it.
@martinthomson do you remember why this was put in? This kind of RTT guessing isn't using packet number and so becomes very wrong very quickly when there is loss. |
mxinden
reviewed
Sep 16, 2024
mxinden
approved these changes
Sep 17, 2024
github-merge-queue
bot
removed this pull request from the merge queue due to failed status checks
github-merge-queue
bot
removed this pull request from the merge queue due to failed status checks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Because under lossy conditions (e.g., QNS
handshakelosstest), the guess can be multiple times the actual RTT, which when encoded in the resumption ticket will cause an extremely slow second handshake, often causing the test to time out.Broken out of #1998
Fixes #2088