Fix bugs with DogStatsD when using named pipes or UDS #4933

andrewlock · 2023-11-28T12:58:08Z

Summary of changes

Explicitly pass in the calculated pipename to StatsD when using named pipes
Correctly fallback when using UDS for traces and UDP for metrics
Add tests for StatsDService creation

Reason for change

If there's no PipeName set, no ServerName set, and no environment variables set, StatsD throws an ArgumentNullException

Similarly, if DD_TRACE_AGENT_URL is set to a unix:// path, and the metrics socket is not in the default location, stats falls back to UDP, but we then currently fail to create the StatsDService.

Implementation details

Explicitly pass the named pipe through the config.
Add workaround for UDS/UDP combo by checking for an empty host
Added an [EnvironmentRestorer] attribute for use in testing to make it easier to clear and restore env vars

Test coverage

Happy path is already covered by named pipe tests - the specific error is hard to reproduce with current infrastructure.

For the UDS/UDP combo, reproduced in unit tests, which are now passing after the fix

Other details

These errors may well be coming from lib-injection which appears to use /opt/datadog/apm/inject/run/dsd.socket for metrics rather than the expected /var/run/datadog/dsd.socket

andrewlock · 2023-11-28T12:59:38Z

tracer/src/Datadog.Trace/TracerManagerFactory.cs

-                        // Environment variables for windows named pipes are not explicitly passed to statsd.
-                        // They are retrieved within the vendored code, so there is nothing to pass.
-                        // Passing anything through StatsdConfig may cause bugs when windows named pipes should be used.


This is a pretty old comment, which I think is incorrect now. We don't want statsd to grab the values directly from the environment, because

It may not have permission too, so it's unsafe

We may have a "better" pipename to pass to it via config etc

I suspect this comment only applied to a previous version than the currently vendored one

datadog-ddstaging · 2023-11-28T13:18:35Z

Datadog Report

Branch report: andrew/fix-statsdconfig
Commit report: 5a5924f

❄️ dd-trace-dotnet: 0 Failed, 1 New Flaky, 302842 Passed, 994 Skipped, 1 Regression, 40m 54.8s Wall Time

New Flaky Tests (1)

ManualOnly - Datadog.Trace.ClrProfiler.IntegrationTests.ManualInstrumentationTests - Last Failure

Expand for error

 Results do not match.
 Differences:
 Received: ManualInstrumentationTests.ManualOnly.received.txt
 Verified: ManualInstrumentationTests.ManualOnly.verified.txt
 Received Content:
 [
   {
     TraceId: Id_1,
     SpanId: Id_2,
     Name: Manual-1.Initial,
 ...

⌛ Performance Regressions vs Default Branch (1)

Profiler_liveheap_cpu_walltime - scenarios 4.65s (+222.83ms, +5%) - Details

andrewlock · 2023-11-28T13:29:20Z

Execution-Time Benchmarks Report ⏱️

Execution-time results for samples comparing the following branches/commits:

Execution-time benchmarks measure the whole time it takes to execute a program. And are intended to measure the one-off costs. Cases where the execution time results for the PR are worse than latest master results are shown in red. The following thresholds were used for comparing the execution times:

Welch test with statistical test for significance of 5%
Only results indicating a difference greater than 5% and 5 ms are considered.

Note that these results are based on a single point-in-time result for each branch. For full results, see the dashboard.

Graphs show the p99 interval based on the mean and StdDev of the test run, as well as the mean value of the run (shown as a diamond below the graph).

gantt
    title Execution time (ms) FakeDbCommand (.NET Framework 4.6.2) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (4933) - mean (71ms)  : 64, 79
     .   : milestone, 71,
    master - mean (69ms)  : 63, 75
     .   : milestone, 69,

    section CallTarget+Inlining+NGEN
    This PR (4933) - mean (1,006ms)  : 981, 1031
     .   : milestone, 1006,
    master - mean (1,002ms)  : 979, 1024
     .   : milestone, 1002,

gantt
    title Execution time (ms) FakeDbCommand (.NET Core 3.1) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (4933) - mean (106ms)  : 104, 108
     .   : milestone, 106,
    master - mean (106ms)  : 102, 109
     .   : milestone, 106,

    section CallTarget+Inlining+NGEN
    This PR (4933) - mean (693ms)  : 675, 710
     .   : milestone, 693,
    master - mean (687ms)  : 675, 698
     .   : milestone, 687,

gantt
    title Execution time (ms) FakeDbCommand (.NET 6) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (4933) - mean (90ms)  : 86, 94
     .   : milestone, 90,
    master - mean (90ms)  : 88, 92
     .   : milestone, 90,

    section CallTarget+Inlining+NGEN
    This PR (4933) - mean (658ms)  : 636, 681
     .   : milestone, 658,
    master - mean (659ms)  : 638, 681
     .   : milestone, 659,

gantt
    title Execution time (ms) HttpMessageHandler (.NET Framework 4.6.2) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (4933) - mean (189ms)  : 186, 192
     .   : milestone, 189,
    master - mean (189ms)  : 185, 193
     .   : milestone, 189,

    section CallTarget+Inlining+NGEN
    This PR (4933) - mean (1,134ms)  : 1115, 1152
     .   : milestone, 1134,
    master - mean (1,132ms)  : 1111, 1153
     .   : milestone, 1132,

gantt
    title Execution time (ms) HttpMessageHandler (.NET Core 3.1) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (4933) - mean (272ms)  : 266, 278
     .   : milestone, 272,
    master - mean (272ms)  : 267, 277
     .   : milestone, 272,

    section CallTarget+Inlining+NGEN
    This PR (4933) - mean (1,086ms)  : 1062, 1109
     .   : milestone, 1086,
    master - mean (1,087ms)  : 1057, 1117
     .   : milestone, 1087,

gantt
    title Execution time (ms) HttpMessageHandler (.NET 6) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (4933) - mean (262ms)  : 259, 265
     .   : milestone, 262,
    master - mean (262ms)  : 257, 266
     .   : milestone, 262,

    section CallTarget+Inlining+NGEN
    This PR (4933) - mean (1,055ms)  : 1033, 1076
     .   : milestone, 1055,
    master - mean (1,059ms)  : 1035, 1083
     .   : milestone, 1059,

andrewlock · 2023-11-28T14:44:44Z

Benchmarks Report 🐌

Benchmarks for #4933 compared to master:

3 benchmarks are faster, with geometric mean 1.202
1 benchmarks are slower, with geometric mean 1.214
1 benchmarks have fewer allocations

The following thresholds were used for comparing the benchmark speeds:

Mann–Whitney U test with statistical test for significance of 5%
Only results indicating a difference greater than 10% and 0.3 ns are considered.

Allocation changes below 0.5% are ignored.

Benchmark details

Benchmarks.Trace.ActivityBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Gen 1	Gen 2	Allocated
master	`StartStopWithChild`	net6.0	8.61μs	48.8ns	331ns	0.0303	0.013	0	7.48 KB
master	`StartStopWithChild`	netcoreapp3.1	10.6μs	59.3ns	365ns	0.0319	0.0106	0	7.57 KB
master	`StartStopWithChild`	net472	17.1μs	47.2ns	183ns	1.36	0.382	0.136	7.96 KB
#4933	`StartStopWithChild`	net6.0	8.81μs	48.9ns	320ns	0.0173	0.00867	0	7.48 KB
#4933	`StartStopWithChild`	netcoreapp3.1	10.5μs	57ns	312ns	0.0208	0.0104	0	7.58 KB
#4933	`StartStopWithChild`	net472	17.3μs	41.8ns	156ns	1.34	0.363	0.112	7.96 KB

Benchmarks.Trace.AgentWriterBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`WriteAndFlushEnrichedTraces`	net6.0	467μs	339ns	1.31μs	0	2.7 KB
master	`WriteAndFlushEnrichedTraces`	netcoreapp3.1	639μs	274ns	1.06μs	0	2.7 KB
master	`WriteAndFlushEnrichedTraces`	net472	788μs	451ns	1.75μs	0.393	3.3 KB
#4933	`WriteAndFlushEnrichedTraces`	net6.0	459μs	311ns	1.2μs	0	2.7 KB
#4933	`WriteAndFlushEnrichedTraces`	netcoreapp3.1	643μs	287ns	1.08μs	0	2.7 KB
#4933	`WriteAndFlushEnrichedTraces`	net472	782μs	480ns	1.73μs	0.388	3.3 KB

Benchmarks.Trace.Asm.AppSecBodyBenchmark - Slower ⚠️ Same allocations ✔️

Slower ⚠️ in #4933

Benchmark	diff/base	Base Median (ns)	Diff Median (ns)	Modality
Benchmarks.Trace.Asm.AppSecBodyBenchmark.ObjectExtractorMoreComplexBody‑net472	1.214	3,390.77	4,116.18

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Gen 1	Allocated
master	`AllCycleSimpleBody`	net6.0	43.5μs	24.2ns	90.5ns	0.0218	0	1.77 KB
master	`AllCycleSimpleBody`	netcoreapp3.1	47.7μs	37.7ns	141ns	0	0	1.74 KB
master	`AllCycleSimpleBody`	net472	49.4μs	24.8ns	96.1ns	0.27	0	1.81 KB
master	`AllCycleMoreComplexBody`	net6.0	212μs	84.4ns	327ns	0.106	0	9.25 KB
master	`AllCycleMoreComplexBody`	netcoreapp3.1	228μs	198ns	768ns	0.112	0	9.14 KB
master	`AllCycleMoreComplexBody`	net472	237μs	117ns	452ns	1.42	0	9.32 KB
master	`ObjectExtractorSimpleBody`	net6.0	130ns	0.129ns	0.465ns	0.00392	0	280 B
master	`ObjectExtractorSimpleBody`	netcoreapp3.1	188ns	0.182ns	0.705ns	0.00377	0	272 B
master	`ObjectExtractorSimpleBody`	net472	153ns	0.0573ns	0.214ns	0.0446	0	281 B
master	`ObjectExtractorMoreComplexBody`	net6.0	2.74μs	1.38ns	5.33ns	0.0534	0	3.78 KB
master	`ObjectExtractorMoreComplexBody`	netcoreapp3.1	3.83μs	1.15ns	4.47ns	0.0498	0	3.69 KB
master	`ObjectExtractorMoreComplexBody`	net472	3.39μs	1.03ns	3.73ns	0.602	0.0051	3.8 KB
#4933	`AllCycleSimpleBody`	net6.0	44.6μs	139ns	539ns	0.022	0	1.77 KB
#4933	`AllCycleSimpleBody`	netcoreapp3.1	46.8μs	66.8ns	259ns	0.0233	0	1.74 KB
#4933	`AllCycleSimpleBody`	net472	50.1μs	27.1ns	105ns	0.274	0	1.81 KB
#4933	`AllCycleMoreComplexBody`	net6.0	212μs	80.3ns	290ns	0.106	0	9.25 KB
#4933	`AllCycleMoreComplexBody`	netcoreapp3.1	225μs	251ns	939ns	0.112	0	9.14 KB
#4933	`AllCycleMoreComplexBody`	net472	239μs	60.3ns	217ns	1.42	0	9.32 KB
#4933	`ObjectExtractorSimpleBody`	net6.0	133ns	0.0428ns	0.16ns	0.00396	0	280 B
#4933	`ObjectExtractorSimpleBody`	netcoreapp3.1	185ns	0.113ns	0.425ns	0.00372	0	272 B
#4933	`ObjectExtractorSimpleBody`	net472	152ns	0.13ns	0.503ns	0.0446	0	281 B
#4933	`ObjectExtractorMoreComplexBody`	net6.0	2.95μs	2.07ns	7.75ns	0.0531	0	3.78 KB
#4933	`ObjectExtractorMoreComplexBody`	netcoreapp3.1	3.56μs	1.04ns	3.89ns	0.0498	0	3.69 KB
#4933	`ObjectExtractorMoreComplexBody`	net472	4.12μs	3.41ns	13.2ns	0.601	0.00618	3.8 KB

Benchmarks.Trace.Asm.AppSecWafBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Gen 1	Allocated
master	`RunWaf(args=NestedMap (10))`	net6.0	51.8μs	266ns	1.27μs	0.217	0	16.06 KB
master	`RunWaf(args=NestedMap (10))`	netcoreapp3.1	65.4μs	231ns	802ns	0.193	0	16.06 KB
master	`RunWaf(args=NestedMap (10))`	net472	97.7μs	490ns	2.08μs	2.52	0.0934	16.14 KB
master	`RunWafWithAttack(args=Neste(...)tack) [22])`	net6.0	122μs	631ns	3.22μs	0.289	0	22.41 KB
master	`RunWafWithAttack(args=Neste(...)tack) [22])`	netcoreapp3.1	141μs	283ns	1.09μs	0.269	0	22.36 KB
master	`RunWafWithAttack(args=Neste(...)tack) [22])`	net472	171μs	851ns	3.71μs	3.56	0.166	22.7 KB
master	`RunWaf(args=NestedMap (100))`	net6.0	102μs	595ns	5.26μs	0.458	0	32.76 KB
master	`RunWaf(args=NestedMap (100))`	netcoreapp3.1	132μs	336ns	1.3μs	0.395	0	33.33 KB
master	`RunWaf(args=NestedMap (100))`	net472	190μs	65.9ns	255ns	5.3	0.365	33.67 KB
master	`RunWafWithAttack(args=Neste(...)tack) [23])`	net6.0	170μs	925ns	5.31μs	0.484	0	39.1 KB
master	`RunWafWithAttack(args=Neste(...)tack) [23])`	netcoreapp3.1	205μs	957ns	3.71μs	0.508	0	39.63 KB
master	`RunWafWithAttack(args=Neste(...)tack) [23])`	net472	270μs	224ns	869ns	6.38	0.521	40.23 KB
master	`RunWaf(args=NestedMap (20))`	net6.0	106μs	40.6ns	157ns	0.422	0	32.18 KB
master	`RunWaf(args=NestedMap (20))`	netcoreapp3.1	133μs	277ns	1.07μs	0.381	0	32.3 KB
master	`RunWaf(args=NestedMap (20))`	net472	182μs	105ns	406ns	5.1	0.364	32.63 KB
master	`RunWafWithAttack(args=Neste(...)tack) [22])`	net6.0	168μs	59.2ns	222ns	0.497	0	38.53 KB
master	`RunWafWithAttack(args=Neste(...)tack) [22])`	netcoreapp3.1	210μs	186ns	719ns	0.528	0	38.6 KB
master	`RunWafWithAttack(args=Neste(...)tack) [22])`	net472	262μs	122ns	457ns	6.21	0.518	39.2 KB
#4933	`RunWaf(args=NestedMap (10))`	net6.0	52.6μs	279ns	1.48μs	0.207	0	16.06 KB
#4933	`RunWaf(args=NestedMap (10))`	netcoreapp3.1	64.7μs	232ns	869ns	0.199	0	16.06 KB
#4933	`RunWaf(args=NestedMap (10))`	net472	94.2μs	48.7ns	182ns	2.53	0.0936	16.14 KB
#4933	`RunWafWithAttack(args=Neste(...)tack) [22])`	net6.0	112μs	49.8ns	193ns	0.282	0	22.41 KB
#4933	`RunWafWithAttack(args=Neste(...)tack) [22])`	netcoreapp3.1	134μs	697ns	3.42μs	0.274	0	22.36 KB
#4933	`RunWafWithAttack(args=Neste(...)tack) [22])`	net472	165μs	121ns	452ns	3.54	0.164	22.7 KB
#4933	`RunWaf(args=NestedMap (100))`	net6.0	105μs	593ns	4.44μs	0.427	0	32.76 KB
#4933	`RunWaf(args=NestedMap (100))`	netcoreapp3.1	132μs	715ns	3.92μs	0.436	0	33.33 KB
#4933	`RunWaf(args=NestedMap (100))`	net472	184μs	131ns	471ns	5.32	0.367	33.67 KB
#4933	`RunWafWithAttack(args=Neste(...)tack) [23])`	net6.0	176μs	90.8ns	352ns	0.523	0	39.1 KB
#4933	`RunWafWithAttack(args=Neste(...)tack) [23])`	netcoreapp3.1	202μs	758ns	2.84μs	0.501	0	39.63 KB
#4933	`RunWafWithAttack(args=Neste(...)tack) [23])`	net472	258μs	279ns	968ns	6.28	0.512	40.23 KB
#4933	`RunWaf(args=NestedMap (20))`	net6.0	98.8μs	32.7ns	118ns	0.4	0	32.18 KB
#4933	`RunWaf(args=NestedMap (20))`	netcoreapp3.1	131μs	123ns	443ns	0.434	0	32.3 KB
#4933	`RunWaf(args=NestedMap (20))`	net472	181μs	89.6ns	347ns	5.16	0.362	32.63 KB
#4933	`RunWafWithAttack(args=Neste(...)tack) [22])`	net6.0	162μs	51.4ns	199ns	0.487	0	38.53 KB
#4933	`RunWafWithAttack(args=Neste(...)tack) [22])`	netcoreapp3.1	204μs	705ns	2.73μs	0.496	0	38.6 KB
#4933	`RunWafWithAttack(args=Neste(...)tack) [22])`	net472	258μs	141ns	526ns	6.19	0.515	39.2 KB

Benchmarks.Trace.AspNetCoreBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`SendRequest`	net6.0	170μs	158ns	546ns	0.17	18.25 KB
master	`SendRequest`	netcoreapp3.1	193μs	247ns	956ns	0.192	20.41 KB
master	`SendRequest`	net472	0ns	0ns	0ns	0	0 b
#4933	`SendRequest`	net6.0	175μs	196ns	759ns	0.175	18.25 KB
#4933	`SendRequest`	netcoreapp3.1	195μs	362ns	1.4μs	0.195	20.41 KB
#4933	`SendRequest`	net472	0.000334ns	0.000185ns	0.000716ns	0	0 b

Benchmarks.Trace.CIVisibilityProtocolWriterBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Gen 1	Gen 2	Allocated
master	`WriteAndFlushEnrichedTraces`	net6.0	548μs	403ns	1.56μs	0.561	0	0	41.72 KB
master	`WriteAndFlushEnrichedTraces`	netcoreapp3.1	653μs	725ns	2.71μs	0.331	0	0	41.79 KB
master	`WriteAndFlushEnrichedTraces`	net472	803μs	3μs	11.2μs	8.41	2.4	0.401	53.21 KB
#4933	`WriteAndFlushEnrichedTraces`	net6.0	545μs	1.26μs	4.89μs	0.539	0	0	41.81 KB
#4933	`WriteAndFlushEnrichedTraces`	netcoreapp3.1	677μs	963ns	3.73μs	0.331	0	0	41.63 KB
#4933	`WriteAndFlushEnrichedTraces`	net472	826μs	4.09μs	18.3μs	8.22	2.47	0.411	53.22 KB

Benchmarks.Trace.DbCommandBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Gen 1	Allocated
master	`ExecuteNonQuery`	net6.0	1.09μs	0.906ns	3.51ns	0.011	0	768 B
master	`ExecuteNonQuery`	netcoreapp3.1	1.44μs	0.347ns	1.34ns	0.0102	0	768 B
master	`ExecuteNonQuery`	net472	1.72μs	0.676ns	2.44ns	0.116	0.000853	730 B
#4933	`ExecuteNonQuery`	net6.0	1.18μs	0.731ns	2.83ns	0.0106	0	768 B
#4933	`ExecuteNonQuery`	netcoreapp3.1	1.48μs	0.974ns	3.37ns	0.00968	0	768 B
#4933	`ExecuteNonQuery`	net472	1.78μs	2.08ns	7.77ns	0.116	0	730 B

Benchmarks.Trace.ElasticsearchBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Gen 1	Allocated
master	`CallElasticsearch`	net6.0	1.13μs	1.12ns	4.19ns	0.0131	0	936 B
master	`CallElasticsearch`	netcoreapp3.1	1.5μs	0.587ns	2.2ns	0.0122	0	936 B
master	`CallElasticsearch`	net472	2.49μs	0.929ns	3.6ns	0.151	0.00124	955 B
master	`CallElasticsearchAsync`	net6.0	1.26μs	0.571ns	2.21ns	0.0126	0	912 B
master	`CallElasticsearchAsync`	netcoreapp3.1	1.66μs	1.58ns	6.13ns	0.0134	0	984 B
master	`CallElasticsearchAsync`	net472	2.57μs	0.653ns	2.35ns	0.159	0	1.01 KB
#4933	`CallElasticsearch`	net6.0	1.17μs	0.651ns	2.44ns	0.013	0	936 B
#4933	`CallElasticsearch`	netcoreapp3.1	1.56μs	0.812ns	3.14ns	0.0125	0	936 B
#4933	`CallElasticsearch`	net472	2.54μs	0.516ns	1.86ns	0.151	0	955 B
#4933	`CallElasticsearchAsync`	net6.0	1.34μs	0.364ns	1.36ns	0.0128	0	912 B
#4933	`CallElasticsearchAsync`	netcoreapp3.1	1.61μs	3.19ns	11.9ns	0.013	0	984 B
#4933	`CallElasticsearchAsync`	net472	2.6μs	1.27ns	4.92ns	0.16	0	1.01 KB

Benchmarks.Trace.GraphQLBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Gen 1	Allocated
master	`ExecuteAsync`	net6.0	1.33μs	0.914ns	3.54ns	0.0125	0	912 B
master	`ExecuteAsync`	netcoreapp3.1	1.61μs	0.661ns	2.47ns	0.0121	0	912 B
master	`ExecuteAsync`	net472	1.78μs	1.06ns	3.98ns	0.138	0.000887	875 B
#4933	`ExecuteAsync`	net6.0	1.23μs	1.12ns	4.34ns	0.0131	0	912 B
#4933	`ExecuteAsync`	netcoreapp3.1	1.55μs	0.667ns	2.49ns	0.0124	0	912 B
#4933	`ExecuteAsync`	net472	1.8μs	1.02ns	3.93ns	0.139	0.0009	875 B

Benchmarks.Trace.HttpClientBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`SendAsync`	net6.0	4.08μs	2.18ns	8.16ns	0.0286	2.1 KB
master	`SendAsync`	netcoreapp3.1	4.95μs	2.59ns	10ns	0.0345	2.63 KB
master	`SendAsync`	net472	7.8μs	2.53ns	9.47ns	0.523	3.31 KB
#4933	`SendAsync`	net6.0	4.03μs	1.61ns	6.22ns	0.0284	2.1 KB
#4933	`SendAsync`	netcoreapp3.1	5.05μs	1.88ns	7.3ns	0.0353	2.63 KB
#4933	`SendAsync`	net472	7.79μs	3.02ns	11.7ns	0.525	3.31 KB

Benchmarks.Trace.Iast.StringAspectsBenchmark - Faster 🎉 Fewer allocations 🎉

Faster 🎉 in #4933

Benchmark	base/diff	Base Median (ns)	Diff Median (ns)	Modality
Benchmarks.Trace.Iast.StringAspectsBenchmark.StringConcatBenchmark(parameters: System.Collections.Generic.List`1[System.String])‑netcoreapp3.1	1.173	61,800.00	52,700.00	multimodal

Fewer allocations 🎉 in #4933

Benchmark	Base Allocated	Diff Allocated	Change	Change %
Benchmarks.Trace.Iast.StringAspectsBenchmark.StringConcatAspectBenchmark(parameters: System.Collections.Generic.List`1[System.String])‑net472	65.54 KB	57.34 KB	-8.19 KB	-12.50%

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Allocated
master	`StringConcatBenchmark(parameters=Syste(...)ring] [48])`	net6.0	54.4μs	269ns	1.11μs	43.44 KB
master	`StringConcatBenchmark(parameters=Syste(...)ring] [48])`	netcoreapp3.1	62.3μs	748ns	7.44μs	42.64 KB
master	`StringConcatBenchmark(parameters=Syste(...)ring] [48])`	net472	37.7μs	116ns	419ns	59.77 KB
master	`StringConcatAspectBenchmark(parameters=Syste(...)ring] [48])`	net6.0	65.2μs	288ns	1.08μs	43.29 KB
master	`StringConcatAspectBenchmark(parameters=Syste(...)ring] [48])`	netcoreapp3.1	107μs	3.6μs	35.6μs	42.64 KB
master	`StringConcatAspectBenchmark(parameters=Syste(...)ring] [48])`	net472	56.6μs	781ns	7.77μs	65.54 KB
#4933	`StringConcatBenchmark(parameters=Syste(...)ring] [48])`	net6.0	52.2μs	270ns	1.27μs	43.44 KB
#4933	`StringConcatBenchmark(parameters=Syste(...)ring] [48])`	netcoreapp3.1	53μs	275ns	1.26μs	42.64 KB
#4933	`StringConcatBenchmark(parameters=Syste(...)ring] [48])`	net472	37.9μs	177ns	771ns	59.81 KB
#4933	`StringConcatAspectBenchmark(parameters=Syste(...)ring] [48])`	net6.0	65.9μs	317ns	1.34μs	43.29 KB
#4933	`StringConcatAspectBenchmark(parameters=Syste(...)ring] [48])`	netcoreapp3.1	97.6μs	3.2μs	31.7μs	42.64 KB
#4933	`StringConcatAspectBenchmark(parameters=Syste(...)ring] [48])`	net472	65.8μs	258ns	997ns	57.34 KB

Benchmarks.Trace.ILoggerBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`EnrichedLog`	net6.0	1.43μs	1.17ns	4.39ns	0.0223	1.57 KB
master	`EnrichedLog`	netcoreapp3.1	2.21μs	0.758ns	2.94ns	0.0213	1.57 KB
master	`EnrichedLog`	net472	2.47μs	2.08ns	8.04ns	0.237	1.5 KB
#4933	`EnrichedLog`	net6.0	1.45μs	0.397ns	1.49ns	0.022	1.57 KB
#4933	`EnrichedLog`	netcoreapp3.1	2.2μs	5.68ns	22ns	0.0215	1.57 KB
#4933	`EnrichedLog`	net472	2.61μs	2.02ns	7.81ns	0.238	1.5 KB

Benchmarks.Trace.Log4netBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Gen 1	Allocated
master	`EnrichedLog`	net6.0	114μs	208ns	804ns	0.0573	0	4.21 KB
master	`EnrichedLog`	netcoreapp3.1	119μs	373ns	1.45μs	0	0	4.21 KB
master	`EnrichedLog`	net472	147μs	129ns	500ns	0.663	0.221	4.39 KB
#4933	`EnrichedLog`	net6.0	112μs	111ns	429ns	0	0	4.21 KB
#4933	`EnrichedLog`	netcoreapp3.1	118μs	171ns	639ns	0	0	4.21 KB
#4933	`EnrichedLog`	net472	150μs	120ns	463ns	0.676	0.225	4.39 KB

Benchmarks.Trace.NLogBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`EnrichedLog`	net6.0	2.98μs	0.976ns	3.78ns	0.0296	2.13 KB
master	`EnrichedLog`	netcoreapp3.1	4.24μs	1.46ns	5.46ns	0.0276	2.13 KB
master	`EnrichedLog`	net472	4.89μs	3.46ns	13.4ns	0.308	1.95 KB
#4933	`EnrichedLog`	net6.0	2.97μs	1.23ns	4.44ns	0.0297	2.13 KB
#4933	`EnrichedLog`	netcoreapp3.1	4.14μs	1.32ns	5.13ns	0.0289	2.13 KB
#4933	`EnrichedLog`	net472	4.88μs	4.19ns	16.2ns	0.307	1.95 KB

Benchmarks.Trace.RedisBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`SendReceive`	net6.0	1.28μs	0.58ns	2.17ns	0.0155	1.1 KB
master	`SendReceive`	netcoreapp3.1	1.79μs	0.978ns	3.66ns	0.0152	1.1 KB
master	`SendReceive`	net472	2.2μs	1.92ns	7.45ns	0.176	1.12 KB
#4933	`SendReceive`	net6.0	1.3μs	0.559ns	2.02ns	0.0157	1.1 KB
#4933	`SendReceive`	netcoreapp3.1	1.7μs	0.882ns	3.42ns	0.0144	1.1 KB
#4933	`SendReceive`	net472	2.16μs	1.97ns	7.62ns	0.177	1.12 KB

Benchmarks.Trace.SerilogBenchmark - Same speed ✔️ Same allocations ✔️

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`EnrichedLog`	net6.0	2.63μs	1.07ns	4.01ns	0.0213	1.53 KB
master	`EnrichedLog`	netcoreapp3.1	3.81μs	1.48ns	5.54ns	0.0211	1.58 KB
master	`EnrichedLog`	net472	4.31μs	1.28ns	4.95ns	0.311	1.97 KB
#4933	`EnrichedLog`	net6.0	2.69μs	1.44ns	5.4ns	0.0204	1.53 KB
#4933	`EnrichedLog`	netcoreapp3.1	3.75μs	2.17ns	8.41ns	0.0206	1.58 KB
#4933	`EnrichedLog`	net472	4.27μs	6.11ns	23.6ns	0.311	1.97 KB

Benchmarks.Trace.SpanBenchmark - Faster 🎉 Same allocations ✔️

Faster 🎉 in #4933

Benchmark	base/diff	Base Median (ns)	Diff Median (ns)	Modality
Benchmarks.Trace.SpanBenchmark.StartFinishScope‑net6.0	1.226	636.80	519.24

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`StartFinishSpan`	net6.0	448ns	0.114ns	0.442ns	0.00745	536 B
master	`StartFinishSpan`	netcoreapp3.1	689ns	0.353ns	1.37ns	0.00722	536 B
master	`StartFinishSpan`	net472	718ns	0.817ns	3.06ns	0.0853	538 B
master	`StartFinishScope`	net6.0	644ns	3.26ns	14.2ns	0.00903	656 B
master	`StartFinishScope`	netcoreapp3.1	945ns	0.469ns	1.82ns	0.00878	656 B
master	`StartFinishScope`	net472	1.01μs	1.08ns	4.2ns	0.0978	618 B
#4933	`StartFinishSpan`	net6.0	471ns	2.24ns	8.97ns	0.00763	536 B
#4933	`StartFinishSpan`	netcoreapp3.1	742ns	0.263ns	1.02ns	0.00732	536 B
#4933	`StartFinishSpan`	net472	713ns	0.196ns	0.759ns	0.0854	538 B
#4933	`StartFinishScope`	net6.0	519ns	0.378ns	1.46ns	0.00915	656 B
#4933	`StartFinishScope`	netcoreapp3.1	917ns	0.217ns	0.839ns	0.00875	656 B
#4933	`StartFinishScope`	net472	956ns	1.06ns	4.12ns	0.0981	618 B

Benchmarks.Trace.TraceAnnotationsBenchmark - Faster 🎉 Same allocations ✔️

Faster 🎉 in #4933

Benchmark	base/diff	Base Median (ns)	Diff Median (ns)	Modality
Benchmarks.Trace.TraceAnnotationsBenchmark.RunOnMethodBegin‑net6.0	1.209	751.51	621.67

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`RunOnMethodBegin`	net6.0	752ns	0.349ns	1.35ns	0.00905	656 B
master	`RunOnMethodBegin`	netcoreapp3.1	968ns	0.884ns	3.42ns	0.00882	656 B
master	`RunOnMethodBegin`	net472	1.03μs	0.258ns	0.929ns	0.0978	618 B
#4933	`RunOnMethodBegin`	net6.0	622ns	0.154ns	0.595ns	0.00907	656 B
#4933	`RunOnMethodBegin`	netcoreapp3.1	967ns	0.419ns	1.62ns	0.00911	656 B
#4933	`RunOnMethodBegin`	net472	1.12μs	0.263ns	1.02ns	0.098	618 B

bouwkast

Looks good to me, just want to double check if we need to see if MetricsPipeNameInternal has a value first

tracer/src/Datadog.Trace/TracerManagerFactory.cs

andrewlock · 2023-11-29T16:27:29Z

Throughput/Crank Report:zap:

Throughput results for AspNetCoreSimpleController comparing the following branches/commits:

Cases where throughput results for the PR are worse than latest master (5% drop or greater), results are shown in red.

Note that these results are based on a single point-in-time result for each branch. For full results, see one of the many, many dashboards!

gantt
    title Throughput Linux x64 (Total requests) 
    dateFormat  X
    axisFormat %s
    section Baseline
    This PR (4933) (10.935M)   : 0, 10935134
    master (10.843M)   : 0, 10842955
    benchmarks/2.9.0 (10.902M)   : 0, 10901856

    section Automatic
    This PR (4933) (7.670M)   : 0, 7669617
    master (7.717M)   : 0, 7717296
    benchmarks/2.9.0 (7.812M)   : 0, 7812244

    section Trace stats
    This PR (4933) (8.092M)   : 0, 8091913
    master (7.987M)   : 0, 7986898

    section Manual
    This PR (4933) (9.912M)   : 0, 9911554
    master (9.641M)   : 0, 9641442

    section Manual + Automatic
    This PR (4933) (7.213M)   : 0, 7212599
    master (7.226M)   : 0, 7226484

    section Version Conflict
    This PR (4933) (6.355M)   : 0, 6355156
    master (6.403M)   : 0, 6403123

gantt
    title Throughput Linux arm64 (Total requests) 
    dateFormat  X
    axisFormat %s
    section Baseline
    This PR (4933) (9.510M)   : 0, 9509711
    master (9.713M)   : 0, 9712994
    benchmarks/2.9.0 (9.771M)   : 0, 9771450

    section Automatic
    This PR (4933) (6.663M)   : 0, 6662872
    master (6.607M)   : 0, 6606537

    section Trace stats
    This PR (4933) (6.994M)   : 0, 6994071
    master (6.888M)   : 0, 6887932

    section Manual
    This PR (4933) (8.025M)   : 0, 8025253
    master (8.221M)   : 0, 8220805

    section Manual + Automatic
    This PR (4933) (6.242M)   : 0, 6241852
    master (6.003M)   : 0, 6003182

    section Version Conflict
    This PR (4933) (5.705M)   : 0, 5704990
    master (5.548M)   : 0, 5548209

gantt
    title Throughput Windows x64 (Total requests) 
    dateFormat  X
    axisFormat %s
    section Baseline
    This PR (4933) (10.474M)   : 0, 10474084
    master (10.099M)   : 0, 10099485
    benchmarks/2.9.0 (10.877M)   : 0, 10876503

    section Automatic
    This PR (4933) (7.307M)   : 0, 7306629
    master (7.366M)   : 0, 7366129
    benchmarks/2.9.0 (8.110M)   : 0, 8109662

    section Trace stats
    This PR (4933) (7.722M)   : 0, 7722195
    master (7.755M)   : 0, 7755352

    section Manual
    This PR (4933) (9.194M)   : 0, 9193513
    master (9.088M)   : 0, 9088139

    section Manual + Automatic
    This PR (4933) (7.171M)   : 0, 7171075
    master (6.936M)   : 0, 6936402

    section Version Conflict
    This PR (4933) (6.325M)   : 0, 6325003
    master (6.126M)   : 0, 6125533

gantt
    title Throughput Linux x64 (ASM) (Total requests) 
    dateFormat  X
    axisFormat %s
    section Baseline
    master (7.591M)   : 0, 7590892
    benchmarks/2.9.0 (7.803M)   : 0, 7803140

    section No attack
    master (2.014M)   : 0, 2013599
    benchmarks/2.9.0 (3.216M)   : 0, 3215869

    section Attack
    master (1.625M)   : 0, 1624581
    benchmarks/2.9.0 (2.464M)   : 0, 2463754

    section Blocking
    master (3.393M)   : 0, 3393406

    section IAST default
    master (6.693M)   : 0, 6692725

    section IAST full
    master (5.723M)   : 0, 5722791

    section Base vuln
    master (0.922M)   : 0, 922289

    section IAST vuln
    master (0.891M)   : 0, 891407

…ment variables

…trics

andrewlock added type:bug area:runtime-metrics labels Nov 28, 2023

andrewlock requested a review from a team as a code owner November 28, 2023 12:58

github-actions bot added the area:tracer The core tracer library (Datadog.Trace, does not include OpenTracing, native code, or integrations) label Nov 28, 2023

andrewlock commented Nov 28, 2023

View reviewed changes

bouwkast approved these changes Nov 29, 2023

View reviewed changes

tracer/src/Datadog.Trace/TracerManagerFactory.cs Show resolved Hide resolved

andrewlock changed the title ~~Explicitly set the pipename when using named pipes with dogstatsd~~ Fix bugs with DogStatsD when using named pipes or UDS Nov 30, 2023

lucaspimentel approved these changes Nov 30, 2023

View reviewed changes

andrewlock force-pushed the andrew/fix-statsdconfig branch from 1343e82 to 49d3537 Compare November 30, 2023 15:42

andrewlock added 5 commits December 1, 2023 10:59

Explicitly set the pipename when using named pipes with dogstatsd

53c50a0

Add EnvironmentRestorerAttribute to make it easier to test on environ…

63dcbb2

…ment variables

Add tests for creation of DogStatsD instance

7dab4fb

Add workaround for dogstatsd when using UDS for traces but UDP for me…

5cf08f7

…trics

I forgot, macos doesn't like UDS

9986c29

andrewlock force-pushed the andrew/fix-statsdconfig branch from 49d3537 to 34d7f99 Compare December 1, 2023 11:01

Forgot macos doesn't even like UDP

5a5924f

andrewlock force-pushed the andrew/fix-statsdconfig branch from 34d7f99 to 5a5924f Compare December 1, 2023 13:04

andrewlock merged commit 7373886 into master Dec 1, 2023
52 of 54 checks passed

andrewlock deleted the andrew/fix-statsdconfig branch December 1, 2023 17:13

github-actions bot added this to the vNext milestone Dec 1, 2023

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Fix bugs with DogStatsD when using named pipes or UDS #4933

Fix bugs with DogStatsD when using named pipes or UDS #4933

andrewlock commented Nov 28, 2023 •

edited

Loading

andrewlock Nov 28, 2023

datadog-ddstaging bot commented Nov 28, 2023 •

edited

Loading

andrewlock commented Nov 28, 2023 •

edited

Loading

andrewlock commented Nov 28, 2023 •

edited

Loading

Raw results

Raw results

Slower ⚠️ in #4933

Raw results

Raw results

Raw results

Raw results

Raw results

Raw results

Raw results

Raw results

Faster 🎉 in #4933

Fewer allocations 🎉 in #4933

Raw results

Raw results

Raw results

Raw results

Raw results

Raw results

Faster 🎉 in #4933

Raw results

Faster 🎉 in #4933

Raw results

bouwkast left a comment

andrewlock commented Nov 29, 2023 •

edited

Loading

Fix bugs with DogStatsD when using named pipes or UDS #4933

Fix bugs with DogStatsD when using named pipes or UDS #4933

Conversation

andrewlock commented Nov 28, 2023 • edited Loading

Summary of changes

Reason for change

Implementation details

Test coverage

Other details

andrewlock Nov 28, 2023

Choose a reason for hiding this comment

datadog-ddstaging bot commented Nov 28, 2023 • edited Loading

Datadog Report

New Flaky Tests (1)

⌛ Performance Regressions vs Default Branch (1)

andrewlock commented Nov 28, 2023 • edited Loading

Execution-Time Benchmarks Report ⏱️

andrewlock commented Nov 28, 2023 • edited Loading

Benchmarks Report 🐌

Benchmark details

Raw results

Raw results

Slower ⚠️ in #4933

Raw results

Raw results

Raw results

Raw results

Raw results

Raw results

Raw results

Raw results

Faster 🎉 in #4933

Fewer allocations 🎉 in #4933

Raw results

Raw results

Raw results

Raw results

Raw results

Raw results

Faster 🎉 in #4933

Raw results

Faster 🎉 in #4933

Raw results

bouwkast left a comment

Choose a reason for hiding this comment

andrewlock commented Nov 29, 2023 • edited Loading

Throughput/Crank Report:zap:

andrewlock commented Nov 28, 2023 •

edited

Loading

datadog-ddstaging bot commented Nov 28, 2023 •

edited

Loading

andrewlock commented Nov 28, 2023 •

edited

Loading

andrewlock commented Nov 28, 2023 •

edited

Loading

andrewlock commented Nov 29, 2023 •

edited

Loading