Skip to content

http: fix incorrect headersTimeout measurement (alt) - #32329

Closed
OrKoN wants to merge 4 commits into
nodejs:masterfrom
OrKoN:fix-headers-timeout-cpp
Closed

http: fix incorrect headersTimeout measurement (alt)#32329
OrKoN wants to merge 4 commits into
nodejs:masterfrom
OrKoN:fix-headers-timeout-cpp

Conversation

@OrKoN

@OrKoNOrKoN commented Mar 17, 2020•
edited
Loading

Copy link
Copy Markdown
Contributor

For keep-alive connections, the headersTimeout may fire during
subsequent request because the measurement was reset after
a request and not before a request.

Fixes: #27363

Alternative implementation for: #30184

Checklist
  • make -j4 test (UNIX), or vcbuild test (Windows) passes
  • tests and/or benchmarks are included
  • documentation is changed or added
  • commit message follows commit guidelines

OrKoN added 2 commits March 17, 2020 18:10
@OrKoN
http: fix incorrect headersTimeout measurement
ed7b316
For keep-alive connections, the headersTimeout may fire during
subsequent request because the measurement was reset after
a request and not before a request.
Fixes: nodejs#27363
@OrKoN
Clean up
f162c42
@nodejs-github-botnodejs-github-bot added c++ Issues and PRs that require attention from people who are familiar with C++. lib / src Issues and PRs related to general changes in the lib or src directory. labels Mar 17, 2020
OrKoN
OrKoN commented Mar 17, 2020
View reviewed changes
Comment threadlib/_http_client.js
req.insecureHTTPParser === undefined ?
isLenient() : req.insecureHTTPParser);
isLenient() : req.insecureHTTPParser,
0);

@OrKoNOrKoNMar 17, 2020

Copy link
Copy Markdown
ContributorAuthor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

for the response parsing by the client the headers timeout is not relevant

OrKoN
OrKoN commented Mar 17, 2020
View reviewed changes
Comment threadlib/_http_server.js
return;
}
function onParserTimeout(server, socket) {
const serverTimeout = server.emit('timeout', socket);

@OrKoNOrKoNMar 17, 2020

Copy link
Copy Markdown
ContributorAuthor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

time tracking is moved to c++ code

OrKoN
OrKoN commented Mar 17, 2020
View reviewed changes
Comment threadsrc/node_http_parser.cc
if (!cb->IsFunction())
return;

MakeCallback(cb.As<Function>(), 0, nullptr);

@OrKoNOrKoNMar 17, 2020

Copy link
Copy Markdown
ContributorAuthor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

should I check the return value? anything else needed to mark the parsing error?

@addaleaxaddaleaxMar 29, 2020

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@OrKoN Since you’d be return;ing here either way, whether the call threw an exception or not, I don’t think so. You can wrap the call in USE() to make it explicit that you’re discarding the return value because of that.

OrKoN
OrKoN commented Mar 17, 2020
View reviewed changes
Comment threadsrc/node_http_parser.cc
auto parsing_time = (now - header_parsing_start_time_) / 1e6;

if (parsing_time > headers_timeout_) {
Local<Value> cb =

@OrKoNOrKoNMar 17, 2020

Copy link
Copy Markdown
ContributorAuthor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

is there a good way to avoid having an extra callback here?

@addaleaxaddaleaxMar 29, 2020

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@OrKoN I don’t think so. But if it only fires in the timeout case, that should be okay.

@OrKoN

OrKoN commented Mar 17, 2020

Copy link
Copy Markdown
ContributorAuthor

There is one test case failing:

=== release test-graph.http ===
Path: async-hooks/test-graph.http
--- stderr ---
'timeout:1' expected to be triggered by 'tcp:2', but was triggered by 'httpincomingmessage:1' instead.
assert.js:102
throw new AssertionError(obj);
^
AssertionError [ERR_ASSERTION]: Expected values to be strictly equal:
1 !== 0
at verifyGraph (/home/runner/work/node/node/test/async-hooks/verify-graph.js:100:10)
at process.<anonymous> (/home/runner/work/node/node/test/async-hooks/test-graph.http.js:29:3)
at process.emit (events.js:327:22) {
generatedMessage: true,
code: 'ERR_ASSERTION',
actual: 1,
expected: 0,
operator: 'strictEqual'
}
Command: out/Release/node /home/runner/work/node/node/test/async-hooks/test-graph.http.js

I am not familiar with async hooks implementation and if anyone can suggest what's wrong here, I'd really appreciate.

@OrKoNOrKoN mentioned this pull request Mar 19, 2020
Closed
3 tasks
@OrKoNOrKoN changed the title http: fix incorrect headersTimeout measurementhttp: fix incorrect headersTimeout measurement (alt)Mar 19, 2020
@OrKoN

OrKoN commented Mar 19, 2020

Copy link
Copy Markdown
ContributorAuthor

I think the difference in the async-hooks test is caused by the call to nowDate which was setting a timer and which is now removed and the internal caching timer is triggered by another call (probably, utcDate).

@mcollina

mcollina commented Mar 19, 2020

Copy link
Copy Markdown
Member

Benchmark CI: https://ci.nodejs.org/view/Node.js%20benchmark/job/benchmark-node-micro-benchmarks/543/

@mcollina

mcollina commented Mar 19, 2020

Copy link
Copy Markdown
Member

Benchmark CI looks ok to me

Details ``` confidence improvement accuracy (*) (**) (***) 23:20:26 http/bench-parser.js n=100000 len=16 * 2.15 % ±1.64% ±2.19% ±2.86% 23:20:26 http/bench-parser.js n=100000 len=32 1.51 % ±2.93% ±3.90% ±5.08% 23:20:26 http/bench-parser.js n=100000 len=4 0.95 % ±3.47% ±4.67% ±6.17% 23:20:26 http/bench-parser.js n=100000 len=8 5.72 % ±6.32% ±8.50% ±11.23% 23:20:26 http/check_invalid_header_char.js n=1000000 input='' -1.69 % ±4.17% ±5.55% ±7.23% 23:20:26 http/check_invalid_header_char.js n=1000000 input='1' -4.21 % ±4.76% ±6.37% ±8.37% 23:20:26 http/check_invalid_header_char.js n=1000000 input='20091' * -3.33 % ±3.01% ±4.00% ±5.21% 23:20:26 http/check_invalid_header_char.js n=1000000 input='close' ** -5.49 % ±3.15% ±4.21% ±5.51% 23:20:26 http/check_invalid_header_char.js n=1000000 input='en-US' -2.43 % ±4.33% ±5.77% ±7.51% 23:20:26 http/check_invalid_header_char.js n=1000000 input='foo\\nbar' 0.34 % ±4.23% ±5.63% ±7.33% 23:20:26 http/check_invalid_header_char.js n=1000000 input='group_acmeair' 1.03 % ±2.87% ±3.82% ±4.97% 23:20:26 http/check_invalid_header_char.js n=1000000 input='gzip' -2.76 % ±4.15% ±5.53% ±7.20% 23:20:26 http/check_invalid_header_char.js n=1000000 input='keep-alive' 0.12 % ±3.33% ±4.44% ±5.77% 23:20:26 http/check_invalid_header_char.js n=1000000 input='LONG_AND_INVALID' -1.62 % ±1.80% ±2.39% ±3.12% 23:20:26 http/check_invalid_header_char.js n=1000000 input='private' -3.55 % ±4.37% ±5.82% ±7.59% 23:20:26 http/check_invalid_header_char.js n=1000000 input='SAMEORIGIN' * -4.27 % ±3.66% ±4.89% ±6.39% 23:20:26 http/check_invalid_header_char.js n=1000000 input='Sat, 07 May 2016 16:54:48 GMT' -2.39 % ±2.54% ±3.38% ±4.40% 23:20:26 http/check_invalid_header_char.js n=1000000 input='text/html; charset=utf-8' 1.63 % ±3.17% ±4.24% ±5.57% 23:20:26 http/check_invalid_header_char.js n=1000000 input='text/plain' * -3.60 % ±3.36% ±4.48% ±5.86% 23:20:26 http/check_invalid_header_char.js n=1000000 input='\\t\\t\\t\\t\\t\\t\\t\\t\\t\\tFoo bar baz' * -4.19 % ±3.20% ±4.28% ±5.62% 23:20:26 http/check_invalid_header_char.js n=1000000 input='\\x7F' -1.18 % ±4.16% ±5.55% ±7.25% 23:20:26 http/check_invalid_header_char.js n=1000000 input='中文呢' 0.66 % ±4.67% ±6.22% ±8.10% 23:20:26 http/check_is_http_token.js n=1000000 key=':' -1.89 % ±6.70% ±8.92% ±11.62% 23:20:26 http/check_is_http_token.js n=1000000 key='((((())))' -2.34 % ±6.06% ±8.07% ±10.53% 23:20:26 http/check_is_http_token.js n=1000000 key='@@' -0.81 % ±6.30% ±8.39% ±10.92% 23:20:26 http/check_is_http_token.js n=1000000 key='Accept-Ranges' -0.88 % ±2.90% ±3.86% ±5.02% 23:20:26 http/check_is_http_token.js n=1000000 key=':alternate-protocol' 0.62 % ±7.26% ±9.66% ±12.58% 23:20:26 http/check_is_http_token.js n=1000000 key='alternate-protocol:' -3.34 % ±4.01% ±5.33% ±6.94% 23:20:26 http/check_is_http_token.js n=1000000 key='alternate-protocol' -2.35 % ±5.15% ±6.85% ±8.91% 23:20:26 http/check_is_http_token.js n=1000000 key='alt-svc' 0.73 % ±5.97% ±7.95% ±10.35% 23:20:26 http/check_is_http_token.js n=1000000 key='Cache-Control' 1.64 % ±3.24% ±4.32% ±5.64% 23:20:26 http/check_is_http_token.js n=1000000 key='Connection' -0.01 % ±2.94% ±3.91% ±5.09% 23:20:26 http/check_is_http_token.js n=1000000 key='Content-Encoding' -1.71 % ±4.07% ±5.42% ±7.07% 23:20:26 http/check_is_http_token.js n=1000000 key='content-length' -0.18 % ±2.90% ±3.86% ±5.03% 23:20:26 http/check_is_http_token.js n=1000000 key='Content-Location' -2.30 % ±4.64% ±6.17% ±8.03% 23:20:26 http/check_is_http_token.js n=1000000 key='content-type' 0.61 % ±3.50% ±4.66% ±6.07% 23:20:26 http/check_is_http_token.js n=1000000 key='Content-Type' 7.05 % ±8.46% ±11.31% ±14.82% 23:20:26 http/check_is_http_token.js n=1000000 key='date' -0.27 % ±2.92% ±3.88% ±5.05% 23:20:26 http/check_is_http_token.js n=1000000 key='ETag' 4.21 % ±5.85% ±7.82% ±10.24% 23:20:26 http/check_is_http_token.js n=1000000 key='Expires' 0.90 % ±3.20% ±4.26% ±5.56% 23:20:26 http/check_is_http_token.js n=1000000 key='Keep-Alive' 1.57 % ±3.52% ±4.69% ±6.10% 23:20:26 http/check_is_http_token.js n=1000000 key='Last-Modified' -1.16 % ±4.07% ±5.42% ±7.05% 23:20:26 http/check_is_http_token.js n=1000000 key='location' -3.06 % ±4.15% ±5.52% ±7.19% 23:20:26 http/check_is_http_token.js n=1000000 key='server' 2.93 % ±4.61% ±6.15% ±8.03% 23:20:26 http/check_is_http_token.js n=1000000 key='Server' 1.74 % ±3.64% ±4.85% ±6.31% 23:20:26 http/check_is_http_token.js n=1000000 key='status' 1.05 % ±3.68% ±4.89% ±6.37% 23:20:26 http/check_is_http_token.js n=1000000 key='TCN' 4.20 % ±5.12% ±6.82% ±8.88% 23:20:26 http/check_is_http_token.js n=1000000 key='Transfer-Encoding' -1.33 % ±4.85% ±6.46% ±8.41% 23:20:26 http/check_is_http_token.js n=1000000 key='Vary' -0.23 % ±4.29% ±5.71% ±7.44% 23:20:26 http/check_is_http_token.js n=1000000 key='version' -0.26 % ±3.68% ±4.90% ±6.38% 23:20:26 http/check_is_http_token.js n=1000000 key='x-frame-options' 0.57 % ±2.54% ±3.38% ±4.40% 23:20:26 http/check_is_http_token.js n=1000000 key='x-xss-protection' -1.27 % ±4.81% ±6.40% ±8.33% 23:20:26 http/check_is_http_token.js n=1000000 key='中文呢' -3.24 % ±6.48% ±8.62% ±11.24% 23:20:26 http/chunked.js duration=5 c=100 len=1 n=16 benchmarker='wrk' 0.45 % ±0.47% ±0.63% ±0.82% 23:20:26 http/chunked.js duration=5 c=100 len=1 n=1 benchmarker='wrk' 0.04 % ±0.07% ±0.09% ±0.11% 23:20:26 http/chunked.js duration=5 c=100 len=1 n=4 benchmarker='wrk' 0.12 % ±0.15% ±0.20% ±0.26% 23:20:26 http/chunked.js duration=5 c=100 len=1 n=8 benchmarker='wrk' 0.11 % ±0.23% ±0.31% ±0.41% 23:20:26 http/chunked.js duration=5 c=100 len=256 n=16 benchmarker='wrk' ** 1.07 % ±0.70% ±0.93% ±1.22% 23:20:26 http/chunked.js duration=5 c=100 len=256 n=1 benchmarker='wrk' 0.01 % ±0.05% ±0.07% ±0.09% 23:20:26 http/chunked.js duration=5 c=100 len=256 n=4 benchmarker='wrk' 0.03 % ±0.14% ±0.18% ±0.24% 23:20:26 http/chunked.js duration=5 c=100 len=256 n=8 benchmarker='wrk' 0.16 % ±0.25% ±0.33% ±0.43% 23:20:26 http/chunked.js duration=5 c=100 len=64 n=16 benchmarker='wrk' 0.78 % ±0.86% ±1.14% ±1.49% 23:20:26 http/chunked.js duration=5 c=100 len=64 n=1 benchmarker='wrk' 0.00 % ±0.08% ±0.10% ±0.13% 23:20:26 http/chunked.js duration=5 c=100 len=64 n=4 benchmarker='wrk' 0.06 % ±0.12% ±0.15% ±0.20% 23:20:26 http/chunked.js duration=5 c=100 len=64 n=8 benchmarker='wrk' -0.02 % ±0.26% ±0.35% ±0.46% 23:20:26 http/client-request-body.js method='end' len=1024 type='asc' dur=5 -0.28 % ±7.69% ±10.25% ±13.38% 23:20:26 http/client-request-body.js method='end' len=1024 type='buf' dur=5 0.63 % ±8.70% ±11.58% ±15.07% 23:20:26 http/client-request-body.js method='end' len=1024 type='utf' dur=5 -1.36 % ±7.72% ±10.28% ±13.39% 23:20:26 http/client-request-body.js method='end' len=256 type='asc' dur=5 -3.93 % ±8.14% ±10.83% ±14.11% 23:20:26 http/client-request-body.js method='end' len=256 type='buf' dur=5 4.35 % ±9.08% ±12.08% ±15.73% 23:20:26 http/client-request-body.js method='end' len=256 type='utf' dur=5 4.40 % ±8.27% ±11.00% ±14.33% 23:20:26 http/client-request-body.js method='end' len=32 type='asc' dur=5 1.28 % ±8.64% ±11.50% ±14.98% 23:20:26 http/client-request-body.js method='end' len=32 type='buf' dur=5 2.35 % ±9.70% ±12.90% ±16.79% 23:20:26 http/client-request-body.js method='end' len=32 type='utf' dur=5 0.75 % ±9.55% ±12.71% ±16.56% 23:20:26 http/client-request-body.js method='write' len=1024 type='asc' dur=5 -0.52 % ±8.16% ±10.86% ±14.14% 23:20:26 http/client-request-body.js method='write' len=1024 type='buf' dur=5 3.41 % ±8.53% ±11.35% ±14.78% 23:20:26 http/client-request-body.js method='write' len=1024 type='utf' dur=5 1.03 % ±8.23% ±10.94% ±14.24% 23:20:26 http/client-request-body.js method='write' len=256 type='asc' dur=5 2.44 % ±10.11% ±13.46% ±17.52% 23:20:26 http/client-request-body.js method='write' len=256 type='buf' dur=5 -0.57 % ±8.72% ±11.60% ±15.09% 23:20:26 http/client-request-body.js method='write' len=256 type='utf' dur=5 -1.64 % ±8.41% ±11.19% ±14.58% 23:20:26 http/client-request-body.js method='write' len=32 type='asc' dur=5 3.12 % ±8.32% ±11.07% ±14.42% 23:20:26 http/client-request-body.js method='write' len=32 type='buf' dur=5 3.76 % ±9.14% ±12.16% ±15.83% 23:20:26 http/client-request-body.js method='write' len=32 type='utf' dur=5 3.15 % ±9.44% ±12.56% ±16.34% 23:20:26 http/cluster.js duration=5 c=500 len=102400 type='buffer' benchmarker='wrk' 0.60 % ±3.13% ±4.17% ±5.43% 23:20:26 http/cluster.js duration=5 c=500 len=102400 type='bytes' benchmarker='wrk' 2.21 % ±3.22% ±4.30% ±5.61% 23:20:26 http/cluster.js duration=5 c=500 len=1024 type='buffer' benchmarker='wrk' 0.17 % ±3.23% ±4.30% ±5.59% 23:20:26 http/cluster.js duration=5 c=500 len=1024 type='bytes' benchmarker='wrk' 0.35 % ±3.28% ±4.36% ±5.68% 23:20:26 http/cluster.js duration=5 c=500 len=4 type='buffer' benchmarker='wrk' -0.60 % ±3.42% ±4.55% ±5.92% 23:20:26 http/cluster.js duration=5 c=500 len=4 type='bytes' benchmarker='wrk' 1.82 % ±2.31% ±3.08% ±4.01% 23:20:26 http/cluster.js duration=5 c=50 len=102400 type='buffer' benchmarker='wrk' * 4.94 % ±3.74% ±4.98% ±6.48% 23:20:26 http/cluster.js duration=5 c=50 len=102400 type='bytes' benchmarker='wrk' -2.31 % ±3.62% ±4.81% ±6.27% 23:20:26 http/cluster.js duration=5 c=50 len=1024 type='buffer' benchmarker='wrk' -3.97 % ±5.67% ±7.55% ±9.82% 23:20:26 http/cluster.js duration=5 c=50 len=1024 type='bytes' benchmarker='wrk' 1.26 % ±6.17% ±8.21% ±10.70% 23:20:26 http/cluster.js duration=5 c=50 len=4 type='buffer' benchmarker='wrk' -3.58 % ±5.97% ±7.94% ±10.33% 23:20:26 http/cluster.js duration=5 c=50 len=4 type='bytes' benchmarker='wrk' -0.53 % ±5.66% ±7.53% ±9.80% 23:20:26 http/create-clientrequest.js e=1 arg='options' url='idn' -0.25 % ±8.27% ±11.00% ±14.32% 23:20:26 http/create-clientrequest.js e=1 arg='options' url='long' -5.66 % ±7.02% ±9.33% ±12.15% 23:20:26 http/create-clientrequest.js e=1 arg='options' url='wpt' -2.20 % ±7.11% ±9.46% ±12.31% 23:20:26 http/create-clientrequest.js e=1 arg='string' url='idn' 4.12 % ±6.83% ±9.11% ±11.89% 23:20:26 http/create-clientrequest.js e=1 arg='string' url='long' -3.20 % ±9.09% ±12.09% ±15.75% 23:20:26 http/create-clientrequest.js e=1 arg='string' url='wpt' -0.84 % ±5.77% ±7.68% ±9.99% 23:20:26 http/create-clientrequest.js e=1 arg='URL' url='idn' 1.91 % ±6.31% ±8.40% ±10.96% 23:20:26 http/create-clientrequest.js e=1 arg='URL' url='long' 0.03 % ±7.48% ±9.95% ±12.96% 23:20:26 http/create-clientrequest.js e=1 arg='URL' url='wpt' 1.19 % ±6.10% ±8.11% ±10.56% 23:20:26 http/end-vs-write-end.js duration=5 method='end' c=100 len=1048576 type='asc' benchmarker='wrk' -0.70 % ±2.56% ±3.41% ±4.45% 23:20:26 http/end-vs-write-end.js duration=5 method='end' c=100 len=1048576 type='buf' benchmarker='wrk' -2.56 % ±8.07% ±10.73% ±13.97% 23:20:26 http/end-vs-write-end.js duration=5 method='end' c=100 len=1048576 type='utf' benchmarker='wrk' 0.14 % ±1.98% ±2.64% ±3.44% 23:20:26 http/end-vs-write-end.js duration=5 method='end' c=100 len=131072 type='asc' benchmarker='wrk' -0.81 % ±2.10% ±2.80% ±3.66% 23:20:26 http/end-vs-write-end.js duration=5 method='end' c=100 len=131072 type='buf' benchmarker='wrk' -0.03 % ±2.37% ±3.15% ±4.10% 23:20:26 http/end-vs-write-end.js duration=5 method='end' c=100 len=131072 type='utf' benchmarker='wrk' 0.45 % ±1.49% ±1.99% ±2.59% 23:20:26 http/end-vs-write-end.js duration=5 method='end' c=100 len=262144 type='asc' benchmarker='wrk' -0.52 % ±1.53% ±2.04% ±2.66% 23:20:26 http/end-vs-write-end.js duration=5 method='end' c=100 len=262144 type='buf' benchmarker='wrk' * 5.79 % ±4.40% ±5.86% ±7.64% 23:20:26 http/end-vs-write-end.js duration=5 method='end' c=100 len=262144 type='utf' benchmarker='wrk' -0.01 % ±2.32% ±3.08% ±4.01% 23:20:26 http/end-vs-write-end.js duration=5 method='end' c=100 len=65536 type='asc' benchmarker='wrk' 0.23 % ±2.33% ±3.10% ±4.04% 23:20:26 http/end-vs-write-end.js duration=5 method='end' c=100 len=65536 type='buf' benchmarker='wrk' 1.37 % ±2.84% ±3.78% ±4.92% 23:20:26 http/end-vs-write-end.js duration=5 method='end' c=100 len=65536 type='utf' benchmarker='wrk' * -1.82 % ±1.80% ±2.40% ±3.12% 23:20:26 http/end-vs-write-end.js duration=5 method='write' c=100 len=1048576 type='asc' benchmarker='wrk' 0.10 % ±5.13% ±6.83% ±8.89% 23:20:26 http/end-vs-write-end.js duration=5 method='write' c=100 len=1048576 type='buf' benchmarker='wrk' 2.24 % ±7.30% ±9.72% ±12.66% 23:20:26 http/end-vs-write-end.js duration=5 method='write' c=100 len=1048576 type='utf' benchmarker='wrk' -0.21 % ±4.83% ±6.43% ±8.37% 23:20:26 http/end-vs-write-end.js duration=5 method='write' c=100 len=131072 type='asc' benchmarker='wrk' -0.11 % ±3.13% ±4.16% ±5.41% 23:20:26 http/end-vs-write-end.js duration=5 method='write' c=100 len=131072 type='buf' benchmarker='wrk' 0.90 % ±2.47% ±3.29% ±4.28% 23:20:26 http/end-vs-write-end.js duration=5 method='write' c=100 len=131072 type='utf' benchmarker='wrk' -1.88 % ±3.44% ±4.57% ±5.96% 23:20:26 http/end-vs-write-end.js duration=5 method='write' c=100 len=262144 type='asc' benchmarker='wrk' 2.08 % ±4.64% ±6.17% ±8.04% 23:20:26 http/end-vs-write-end.js duration=5 method='write' c=100 len=262144 type='buf' benchmarker='wrk' -3.69 % ±4.93% ±6.56% ±8.54% 23:20:26 http/end-vs-write-end.js duration=5 method='write' c=100 len=262144 type='utf' benchmarker='wrk' 2.09 % ±4.51% ±6.01% ±7.82% 23:20:26 http/end-vs-write-end.js duration=5 method='write' c=100 len=65536 type='asc' benchmarker='wrk' -0.63 % ±1.89% ±2.51% ±3.27% 23:20:26 http/end-vs-write-end.js duration=5 method='write' c=100 len=65536 type='buf' benchmarker='wrk' * 2.42 % ±1.91% ±2.54% ±3.30% 23:20:26 http/end-vs-write-end.js duration=5 method='write' c=100 len=65536 type='utf' benchmarker='wrk' -0.89 % ±2.05% ±2.72% ±3.55% 23:20:26 http/headers.js duration=5 len=100 n=1000 benchmarker='wrk' 2.03 % ±2.67% ±3.55% ±4.63% 23:20:26 http/headers.js duration=5 len=100 n=10 benchmarker='wrk' 4.50 % ±5.24% ±6.98% ±9.09% 23:20:26 http/headers.js duration=5 len=1 n=1000 benchmarker='wrk' 1.56 % ±2.33% ±3.11% ±4.04% 23:20:26 http/headers.js duration=5 len=1 n=10 benchmarker='wrk' 1.22 % ±5.07% ±6.75% ±8.81% 23:20:26 ```

addaleax
addaleax approved these changes Mar 29, 2020
View reviewed changes
Comment threadsrc/node_http_parser.cc Outdated
Comment threadsrc/node_http_parser.cc
auto parsing_time = (now - header_parsing_start_time_) / 1e6;

if (parsing_time > headers_timeout_) {
Local<Value> cb =

@addaleaxaddaleaxMar 29, 2020

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@OrKoN I don’t think so. But if it only fires in the timeout case, that should be okay.

Comment threadsrc/node_http_parser.cc
if (!cb->IsFunction())
return;

MakeCallback(cb.As<Function>(), 0, nullptr);

@addaleaxaddaleaxMar 29, 2020

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@OrKoN Since you’d be return;ing here either way, whether the call threw an exception or not, I don’t think so. You can wrap the call in USE() to make it explicit that you’re discarding the return value because of that.

@addaleaxaddaleax added author ready PRs that have at least one approval, no pending requests for changes, and a CI started. review wanted PRs that need reviews. labels Mar 29, 2020
mcollina
mcollina approved these changes Mar 29, 2020
View reviewed changes

@mcollinamcollina left a comment

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

lgtm

@nodejs-github-bot

nodejs-github-bot commented Mar 29, 2020

Copy link
Copy Markdown
Collaborator

CI: https://ci.nodejs.org/job/node-test-pull-request/30220/

@OrKoN@addaleax
Update src/node_http_parser.cc
9536c1b
Co-Authored-By: Anna Henningsen <github@addaleax.net>
@nodejs-github-bot

nodejs-github-bot commented Mar 30, 2020

Copy link
Copy Markdown
Collaborator

CI: https://ci.nodejs.org/job/node-test-pull-request/30257/

mcollina pushed a commit that referenced this pull request Apr 2, 2020
@OrKoN@mcollina
http: fix incorrect headersTimeout measurement
b149eef
For keep-alive connections, the headersTimeout may fire during
subsequent request because the measurement was reset after
a request and not before a request.
PR-URL: #32329Fixes: #27363
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
@mcollina

mcollina commented Apr 2, 2020

Copy link
Copy Markdown
Member

Landed in b149eef

@mcollinamcollina closed this Apr 2, 2020
BethGriggs pushed a commit that referenced this pull request Apr 7, 2020
@OrKoN@BethGriggs
http: fix incorrect headersTimeout measurement
fcb302f
For keep-alive connections, the headersTimeout may fire during
subsequent request because the measurement was reset after
a request and not before a request.
PR-URL: #32329Fixes: #27363
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
@mcollina

mcollina commented Apr 10, 2020

Copy link
Copy Markdown
Member

@BethGriggs this should ideally be backported down to 10 in the next patch release. When is that scheduled? I assume this is going to need backport PRs, should we open them already?

@BethGriggs

BethGriggs commented Apr 10, 2020

Copy link
Copy Markdown
Member

The next v12.x is due out 28th April and the next v10.x May 5th. Assuming that this commit goes out with v13.x current (scheduled next Tuesday), then this change should be able to make it into the upcoming releases. Opening backport PRs anytime after it has gone out in current (next Tuesday) makes sense to me.

@Flarna

Flarna commented Jul 2, 2020•
edited
Loading

Copy link
Copy Markdown
Member

@richardlau Should I create one more backport PR for 10.x? My assumption was that it should be in 12.x for a while first.

@richardlau

richardlau commented Jul 2, 2020

Copy link
Copy Markdown
Member

@Flarna I've been going around the lts-watch-v10.x labelled PRs to see what's landable for the upcoming 10.x release I'm preparing. Your assumption is correct that usually we'd want things to flow back in an ordered manner (so 12.x before 10.x). I haven't checked to see if the 12.x PR would land cleanly on 10.x. If it does we can add a lts-watch-v10.x label to the 12.x backport PR, otherwise yes please open a backport PR for 10.x so it can be ready for a future 10.x release (there's no regular schedule for non-security 10.x releases as it is now in maintenance -- the release team will review periodically to see if there is enough content to warrant a release).

@Flarna

Flarna commented Jul 2, 2020

Copy link
Copy Markdown
Member

I did a fast check and it seems that the commit for 12.x doesn't land on 10.x so seems a dedicated backport PR will be needed. I may create one once the backport to 12.x is finished.

MylesBorins pushed a commit that referenced this pull request Aug 18, 2020
@OrKoN@MylesBorins
http: fix incorrect headersTimeout measurement
11a87ed
For keep-alive connections, the headersTimeout may fire during
subsequent request because the measurement was reset after
a request and not before a request.
Backport-PR-URL: #34131
PR-URL: #32329Fixes: #27363
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
@codebyterecodebytere mentioned this pull request Sep 28, 2020
Merged
@andrewnesterandrewnester mentioned this pull request Oct 20, 2020
Closed
@andrewnester

andrewnester commented Oct 20, 2020

Copy link
Copy Markdown

@OrKoN Could this change cause the following issue with uploading?
#35661

@LxxyxLxxyx mentioned this pull request Dec 25, 2020
Merged
2 tasks
Trott pushed a commit to Lxxyx/node that referenced this pull request Dec 27, 2020
@Lxxyx@Trott
http: remove dead code from internal/http.js
acaa58e
PR-URL: nodejs#36630
Refs: nodejs#32329
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Rich Trott <rtrott@gmail.com>
danielleadams pushed a commit that referenced this pull request Jan 12, 2021
@Lxxyx@danielleadams
http: remove dead code from internal/http.js
9156f43
PR-URL: #36630
Refs: #32329
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Rich Trott <rtrott@gmail.com>
@aduh95aduh95 mentioned this pull request Jan 12, 2021
Closed
aduh95 pushed a commit to aduh95/node that referenced this pull request Jan 13, 2021
@Lxxyx@aduh95
http: remove dead code from internal/http.js
bae71f0
PR-URL: nodejs#36630
Refs: nodejs#32329
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Rich Trott <rtrott@gmail.com>
danielleadams pushed a commit that referenced this pull request Jan 13, 2021
@Lxxyx@danielleadams
Revert "http: remove dead code from internal/http.js"
7e3ad1b
This reverts commit 9156f43.
PR-URL: #36890
Refs: #32329
Reviewed-By: Danielle Adams <adamzdanielle@gmail.com>
danielleadams pushed a commit that referenced this pull request Jan 13, 2021
@Lxxyx@danielleadams
http: remove dead code from internal/http.js
d858c95
PR-URL: #36630
Backport-PR-URL: #36890
Refs: #32329
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Rich Trott <rtrott@gmail.com>
targos pushed a commit that referenced this pull request May 1, 2021
@Lxxyx@targos
http: remove dead code from internal/http.js
8163f31
PR-URL: #36630
Refs: #32329
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Rich Trott <rtrott@gmail.com>
@ArnoldZokas

ArnoldZokas commented Jun 10, 2021

Copy link
Copy Markdown
Contributor

has this been backported to v14?

@AdamKatzDevAdamKatzDev mentioned this pull request Feb 17, 2023
Merged
ijjk added a commit to vercel/next.js that referenced this pull request Mar 17, 2023
@AdamKatzDev@ijjk
feat(standalone): allow configuring KEEP_ALIVE_TIMEOUT via env var (#…
9dbd8d7
…46052)
Resolves #39689, partially resolves #28642 (see notes below)
Inspired by #44627
In #28642 it was also asked to expose `server.headersTimeout`, but it is
probably not needed for most use cases and not implemented even in `next
start`. It was needed to change this option before
nodejs/node#27363.
There also exists a rare bug that is described here
nodejs/node#32329 (comment). To fix
this exposing `server.headersTimeout` might be required both in
`server.js` and in `next start`.
Co-authored-by: JJ Kasper <jj@jjsweb.site>
@codyebbersoncodyebberson mentioned this pull request Jun 19, 2023
Merged
abhishekumar-tyagi pushed a commit to abhishekumar-tyagi/node that referenced this pull request May 5, 2024
@OrKoN@Flarna
http: fix incorrect headersTimeout measurement
7cb2c9b
For keep-alive connections, the headersTimeout may fire during
subsequent request because the measurement was reset after
a request and not before a request.
PR-URL: nodejs/node#32329Fixes: nodejs/node#27363
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
@pimterrypimterry mentioned this pull request Aug 4, 2025
Closed
jimmy54 pushed a commit to jimmy54/nodejs-mobile that referenced this pull request Dec 31, 2025
@OrKoN@BethGriggs
http: fix incorrect headersTimeout measurement
ec1df7b
For keep-alive connections, the headersTimeout may fire during
subsequent request because the measurement was reset after
a request and not before a request.
Backport-PR-URL: nodejs/node#34131
PR-URL: nodejs/node#32329Fixes: nodejs/node#27363
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Sign up for freeto join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@mcollinamcollinamcollina approved these changes
@addaleaxaddaleaxaddaleax approved these changes

Assignees

No one assigned

Labels

author readyPRs that have at least one approval, no pending requests for changes, and a CI started.c++Issues and PRs that require attention from people who are familiar with C++.lib / srcIssues and PRs related to general changes in the lib or src directory.review wantedPRs that need reviews.

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

Regression issue with keep alive connections

12 participants

@OrKoN@mcollina@nodejs-github-bot@BethGriggs@nahuel@targos@Flarna@richardlau@andrewnester@ArnoldZokas@addaleax@MylesBorins