Mark WebTransport streams keep alive #1389

MayyaSunil · 2022-12-01T11:39:15Z

This PR is intended to resolve #1369

martinthomson

See my comments on the issue.

martinthomson

I don't think you have the right places for these calls.

You need one for the server and one for the client.

I think that the client one goes in webtransport_create_session and the server one goes in webtransport_accept_session.

The points you have identified are about streams created within the session, not the session stream.

Finally, I would like to see test cases for this, showing that keep-alives (PING) are sent if the connection is left to go idle for too long, with only a webtransport session active. There are test cases for keep-alive that you can base this on.

larseggert · 2024-02-07T07:51:34Z

@MayyaSunil are you still intending to land this change?

MayyaSunil · 2024-02-07T12:25:06Z

@larseggert, thanks for checking.
I would like to land this.
If you/anyone is planning to submit a patch for the bug, please feel free to do so, as I won't be able to work on this until my current assignments are completed (by the end of March)

larseggert · 2024-04-09T11:36:18Z

@MayyaSunil any update on when you might get to this?

MayyaSunil · 2024-04-11T08:56:51Z

@MayyaSunil any update on when you might get to this?

I will resume this early next week and hope to complete by the end of the week

larseggert · 2024-04-15T05:48:45Z

OK. Making this a draft PR until it's ready for review, please change it back then.

github-actions · 2024-04-17T23:06:30Z

Benchmark results

Performance differences relative to 2463618.

drain a timer quickly time: [311.82 ns 319.15 ns 325.94 ns]
change: [-29.772% -28.465% -27.032%] (p = 0.00 < 0.05)
💚 Performance has improved.
coalesce_acked_from_zero 1+1 entries
time: [202.76 ns 203.28 ns 203.81 ns]
change: [+2.4616% +2.8914% +3.3435%] (p = 0.00 < 0.05)
💔 Performance has regressed.
coalesce_acked_from_zero 3+1 entries
time: [248.40 ns 248.99 ns 249.61 ns]
change: [+3.3741% +3.7299% +4.0540%] (p = 0.00 < 0.05)
💔 Performance has regressed.
coalesce_acked_from_zero 10+1 entries
time: [247.08 ns 247.76 ns 248.59 ns]
change: [+3.0582% +3.6074% +4.1948%] (p = 0.00 < 0.05)
💔 Performance has regressed.
coalesce_acked_from_zero 1000+1 entries
time: [226.01 ns 226.33 ns 226.76 ns]
change: [-20.439% -6.8145% +3.0254%] (p = 0.56 > 0.05)
No change in performance detected.
RxStreamOrderer::inbound_frame()
time: [119.26 ms 119.33 ms 119.40 ms]
change: [+0.0014% +0.2059% +0.3459%] (p = 0.01 < 0.05)
Change within noise threshold.
transfer/Run multiple transfers with varying seeds
time: [120.41 ms 120.73 ms 121.04 ms]
thrpt: [33.048 MiB/s 33.132 MiB/s 33.219 MiB/s]
change:
time: [+0.6807% +1.0546% +1.4080%] (p = 0.00 < 0.05)
thrpt: [-1.3885% -1.0435% -0.6761%]
Change within noise threshold.
transfer/Run multiple transfers with the same seed
time: [120.79 ms 120.99 ms 121.21 ms]
thrpt: [33.000 MiB/s 33.061 MiB/s 33.117 MiB/s]
change:
time: [-0.0877% +0.1479% +0.4062%] (p = 0.25 > 0.05)
thrpt: [-0.4045% -0.1477% +0.0877%]
No change in performance detected.
1-conn/1-100mb-resp (aka. Download)/client
time: [1.1127 s 1.1202 s 1.1281 s]
thrpt: [88.646 MiB/s 89.272 MiB/s 89.875 MiB/s]
change:
time: [-3.4496% -2.1281% -0.9430%] (p = 0.00 < 0.05)
thrpt: [+0.9520% +2.1743% +3.5729%]
Change within noise threshold.
1-conn/10_000-parallel-1b-resp (aka. RPS)/client
time: [429.92 ms 432.15 ms 434.43 ms]
thrpt: [23.019 Kelem/s 23.140 Kelem/s 23.260 Kelem/s]
change:
time: [-0.3675% +0.3151% +1.0375%] (p = 0.39 > 0.05)
thrpt: [-1.0269% -0.3141% +0.3689%]
No change in performance detected.
1-conn/1-1b-resp (aka. HPS)/client
time: [50.554 ms 50.876 ms 51.220 ms]
thrpt: [19.524 elem/s 19.656 elem/s 19.781 elem/s]
change:
time: [-0.7285% +0.2310% +1.2101%] (p = 0.64 > 0.05)
thrpt: [-1.1956% -0.2305% +0.7339%]
No change in performance detected.

Client/server transfer results

Transfer of 134217728 bytes over loopback.

Client	Server	CC	Pacing	Mean [ms]	Min [ms]	Max [ms]	Relative
msquic	msquic			644.6 ± 254.5	398.7	1254.7	1.00
neqo	msquic	reno	on	1065.4 ± 240.3	762.4	1414.0	1.00
neqo	msquic	reno		994.7 ± 274.8	763.2	1539.2	1.00
neqo	msquic	cubic	on	954.2 ± 234.3	784.7	1390.4	1.00
neqo	msquic	cubic		923.7 ± 252.5	754.8	1435.4	1.00
msquic	neqo	reno	on	4495.0 ± 233.5	4243.6	4915.4	1.00
msquic	neqo	reno		4454.3 ± 226.4	4104.9	4931.4	1.00
msquic	neqo	cubic	on	4481.1 ± 182.1	4220.8	4763.1	1.00
msquic	neqo	cubic		4401.3 ± 194.7	4131.6	4626.2	1.00
neqo	neqo	reno	on	4185.6 ± 320.6	3906.2	4849.4	1.00
neqo	neqo	reno		3737.2 ± 343.4	3379.3	4470.1	1.00
neqo	neqo	cubic	on	4426.0 ± 215.8	4085.8	4849.5	1.00
neqo	neqo	cubic		4381.3 ± 183.8	4108.5	4703.5	1.00

⬇️ Download logs

larseggert · 2024-04-19T12:44:50Z

@MayyaSunil is this good to go functionality-wise? Does it still need a test?

MayyaSunil · 2024-04-19T12:54:23Z

@larseggert Its not okay functionality wise and needs a test

larseggert · 2024-12-03T13:45:48Z

CC @jesup

github-actions · 2024-12-03T15:57:55Z

Failed Interop Tests

QUIC Interop Runner, client vs. server, differences relative to 9143d73.

neqo-latest as client

neqo-latest as server

msquic vs. neqo-latest: Z U
mvfst vs. neqo-latest: Z A L1 C1
ngtcp2 vs. neqo-latest: ⚠️C20
quic-go vs. neqo-latest: run cancelled after 20 min
quinn vs. neqo-latest: V2
xquic vs. neqo-latest: M

All results

Succeeded Interop Tests

QUIC Interop Runner, client vs. server

neqo-latest as client

neqo-latest vs. aioquic: H DC LR C20 M S R 3 B U A L1 L2 C1 C2 6 V2
neqo-latest vs. go-x-net: H DC LR M B U A L2 C2 6
neqo-latest vs. haproxy: H DC LR C20 M S R Z 3 B U A L1 L2 C1 C2 6 V2
neqo-latest vs. kwik: H DC LR C20 M S R 🚀Z 3 B U A L1 L2 C1 C2 6 V2
neqo-latest vs. lsquic: H DC LR C20 M S R Z 3 B U E A L1 L2 C1 C2 6 V2
neqo-latest vs. msquic: H DC LR C20 M S R 🚀Z B U L1 L2 C1 C2 6 V2
neqo-latest vs. mvfst: H DC LR M R Z 3 B U L2 C2 6
neqo-latest vs. neqo: H DC LR C20 M S R Z 3 B U E A L1 L2 C1 C2 6 V2
neqo-latest vs. neqo-latest: H DC LR C20 M S R Z 3 B U E A L1 L2 C1 C2 6 V2
neqo-latest vs. nginx: H DC LR C20 M S R Z 3 B U A L1 L2 C1 C2 6
neqo-latest vs. ngtcp2: H DC LR C20 M S R Z 3 B U E A L1 L2 C1 C2 6 V2
neqo-latest vs. picoquic: H DC LR C20 M S R Z 3 B U E A L1 L2 C1 C2 6 V2
neqo-latest vs. quic-go: H DC LR C20 M S R Z 3 B U A L1 L2 C1 C2 6
neqo-latest vs. quiche: H DC LR C20 M S R Z 3 B U A L1 L2 C1 C2 6
neqo-latest vs. quinn: H DC LR C20 M S R Z 3 B U E A L1 L2 C1 C2 6
neqo-latest vs. s2n-quic: H DC LR C20 M S R 3 B U E A ⚠️L1 L2 C1 C2 6
neqo-latest vs. xquic: H DC LR C20 M R Z 3 B U L1 L2 C1 C2 6

neqo-latest as server

aioquic vs. neqo-latest: H DC LR C20 M S R Z 3 B A L1 L2 C1 C2 6 V2
chrome vs. neqo-latest: 3
go-x-net vs. neqo-latest: H DC LR M B U A L2 C2 6
kwik vs. neqo-latest: H DC LR C20 M S R Z 3 B U A L1 L2 C1 C2 6 V2
lsquic vs. neqo-latest: H DC LR M S R 3 B E A L1 L2 C1 C2 6 V2
msquic vs. neqo-latest: H DC LR C20 M S R B A L1 L2 C1 C2 6 V2
mvfst vs. neqo-latest: H DC LR M 3 B L2 C2 6
neqo vs. neqo-latest: H DC LR C20 M S R Z 3 B U E A L1 L2 C1 C2 6 V2
ngtcp2 vs. neqo-latest: ⚠️H DC LR M S R Z 3 B U E A L1 L2 C1 C2 6 V2
picoquic vs. neqo-latest: H DC LR C20 M S R Z 3 B U E A L1 L2 C1 C2 6 V2
quiche vs. neqo-latest: H DC LR M S R Z 3 B A L1 L2 C1 C2 6
quinn vs. neqo-latest: H DC LR C20 M S R Z 3 B U E A L1 L2 C1 C2 6
s2n-quic vs. neqo-latest: H DC LR M S R 3 B E A L1 L2 C1 C2 6
xquic vs. neqo-latest: H DC LR C20 S R Z 3 B U A L1 L2 C1 C2 6

Unsupported Interop Tests

QUIC Interop Runner, client vs. server

neqo-latest as client

neqo-latest vs. aioquic: E
neqo-latest vs. go-x-net: C20 S R Z 3 E L1 C1 V2
neqo-latest vs. haproxy: E
neqo-latest vs. kwik: E
neqo-latest vs. msquic: 3 E
neqo-latest vs. mvfst: C20 S E V2
neqo-latest vs. nginx: E V2
neqo-latest vs. quic-go: E V2
neqo-latest vs. quiche: E V2
neqo-latest vs. quinn: V2
neqo-latest vs. s2n-quic: Z V2
neqo-latest vs. xquic: S E V2

neqo-latest as server

aioquic vs. neqo-latest: U E
chrome vs. neqo-latest: H DC LR C20 M S R Z B U E A L1 L2 C1 C2 6 V2
go-x-net vs. neqo-latest: C20 S R Z 3 E L1 C1 V2
kwik vs. neqo-latest: E
lsquic vs. neqo-latest: C20 Z U
msquic vs. neqo-latest: 3 E
mvfst vs. neqo-latest: C20 S R U E V2
quiche vs. neqo-latest: C20 U E V2
s2n-quic vs. neqo-latest: C20 Z U V2
xquic vs. neqo-latest: E V2

larseggert · 2024-12-03T16:22:25Z

Webtransport newbie here. Trying to help wrap this up.

I think that the client one goes in webtransport_create_session

The call to create_bidi_transport_stream in that function already enabled keepalives. Hence, are we done for the client?

and the server one goes in webtransport_accept_session.

I've made that addition, and added a test. WDYT?

codecov · 2024-12-03T19:03:27Z

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 95.39%. Comparing base (9143d73) to head (183237d).
Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #1389   +/-   ##
=======================================
  Coverage   95.39%   95.39%           
=======================================
  Files         113      113           
  Lines       36683    36682    -1     
=======================================
  Hits        34993    34993           
+ Misses       1690     1689    -1

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

MayyaSunil requested a review from KershawChang as a code owner December 1, 2022 11:39

martinthomson requested changes Dec 1, 2022

View reviewed changes

MayyaSunil force-pushed the web_transport_issues branch from 7b6b9b1 to 355fdde Compare December 7, 2022 16:28

MayyaSunil requested a review from martinthomson December 7, 2022 16:30

martinthomson requested changes Dec 7, 2022

View reviewed changes

larseggert marked this pull request as draft April 15, 2024 05:48

Mark WebTransport streams keep alive

09d431e

MayyaSunil force-pushed the web_transport_issues branch from 355fdde to 09d431e Compare April 17, 2024 22:21

Merge branch 'main' into web_transport_issues

2c31d30

Implement suggestions from @martinthomson

183237d

larseggert marked this pull request as ready for review December 3, 2024 16:22

larseggert requested review from larseggert and mxinden as code owners December 3, 2024 16:22

larseggert requested a review from martinthomson December 3, 2024 16:23

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Mark WebTransport streams keep alive #1389

Mark WebTransport streams keep alive #1389

MayyaSunil commented Dec 1, 2022

martinthomson left a comment

martinthomson left a comment

larseggert commented Feb 7, 2024

MayyaSunil commented Feb 7, 2024

larseggert commented Apr 9, 2024

MayyaSunil commented Apr 11, 2024

larseggert commented Apr 15, 2024

github-actions bot commented Apr 17, 2024

larseggert commented Apr 19, 2024

MayyaSunil commented Apr 19, 2024

larseggert commented Dec 3, 2024

github-actions bot commented Dec 3, 2024 •

edited

Loading

Succeeded Interop Tests

neqo-latest as client

neqo-latest as server

Unsupported Interop Tests

neqo-latest as client

neqo-latest as server

larseggert commented Dec 3, 2024

codecov bot commented Dec 3, 2024

Mark WebTransport streams keep alive #1389

Are you sure you want to change the base?

Mark WebTransport streams keep alive #1389

Conversation

MayyaSunil commented Dec 1, 2022

martinthomson left a comment

Choose a reason for hiding this comment

martinthomson left a comment

Choose a reason for hiding this comment

larseggert commented Feb 7, 2024

MayyaSunil commented Feb 7, 2024

larseggert commented Apr 9, 2024

MayyaSunil commented Apr 11, 2024

larseggert commented Apr 15, 2024

github-actions bot commented Apr 17, 2024

Benchmark results

Client/server transfer results

larseggert commented Apr 19, 2024

MayyaSunil commented Apr 19, 2024

larseggert commented Dec 3, 2024

github-actions bot commented Dec 3, 2024 • edited Loading

Failed Interop Tests

neqo-latest as client

neqo-latest as server

Succeeded Interop Tests

neqo-latest as client

neqo-latest as server

Unsupported Interop Tests

neqo-latest as client

neqo-latest as server

larseggert commented Dec 3, 2024

codecov bot commented Dec 3, 2024

Codecov Report

github-actions bot commented Dec 3, 2024 •

edited

Loading