[Tracer] Rework exporter settings by pierotibou · Pull Request #2686 · DataDog/dd-trace-dotnet

pierotibou · 2022-04-14T07:41:53Z

Summary of changes

First, this is a subset of #2455. I did another one as I thought it would be easier to follow up the changes. Also note that I'm trying to make the review easier if you do it commit by commit.
Then, here are the changes so far:

Add more tests
Change the order of precedence of the settings to put the URL back on top
Added a few validation checks (eg the URI is well formed), allowed fallbacking and added validation warnings
Changing a property recomputes the strategy. This one is a bit tricky as we keep the precedence. So if a URL had been set by config, and the user changes the socket path, then it won't change anything.

Reason for change

The main reason is that we introduced a bug by breaking the order of precedence of the parameters. BTW, this order is now documented here. But this implementation has many flaws:

Setting a unix:// URI after a http uri would just break as we're not changing the transport strategy
Setting TracesUnixDomainSocketPath, MetricsUnixDomainSocketPath, MetricsPipeName, TracesPipeName properties is useless if you hadn't given an equivalent configuration through environment. Indeed, we don't recompute the strategy when we set those.

Implementation details

In this PR, I'm not changing the way it works. Thus, all the computation stilll happen in ExporterSettings. Doing so to avoid breaking changes in behavior.

Minor detail, I've pushed commit 5332cf0 in order to prevent an issue in partial trust environment. As we don't support it, I think I can revert it, but we'd need to drop the Sandbox tests in that case. FYI, here the issue was Cannot change initonly field outside its .ctor. I've verified with PEVerify that the issue has been solved (though the code is even uglier, so I'd be happy to revert)

Test coverage

I've done TDD. So I've added tests that match the target behavior. Added them in the first commit marking the ones that fail. And commit by commit I remove the comments.

Other details

Fixes #2426

Next steps

Shall we set TracesUnixDomainSocketPath if a path has been provided by URL?
Switch to a builder template in 3.0

Add the URL as the first parameter to consider Check Uri before creation and allow fallbacks Adds validation warnings Changing the Uri through the property now recomputes the strategy Change the tests as well to order them by order of precedence But didn't add as much tests as in my previous PR (yet)

andrewlock · 2022-04-18T22:50:06Z

Code Coverage Report 📊

✔️ Merging #2686 into master will not change line coverage
✔️ Merging #2686 into master will not change branch coverage
⛔ Merging #2686 into master will will increase complexity by 56

	master	#2686	Change
Lines	`13609` / `18627`	`13571` / `18690`
Lines %	`73%`	`73%`	`0%` ✔️
Branches	`7873` / `11125`	`7860` / `11167`
Branches %	`71%`	`70%`	`0%` ✔️
Complexity	`12431`	`12487`	`56` ⛔

View the full report for further details:

Datadog.Trace Breakdown ✔️

	master	#2686	Change
Lines %	`73%`	`73%`	`0%` ✔️
Branches %	`71%`	`70%`	`0%` ✔️
Complexity	`12431`	`12487`	`56` ⛔

The following classes have significant coverage changes.

File	Line coverage change	Branch coverage change	Complexity change
Datadog.Trace.Telemetry.JsonWebRequestTelemetryTransport	`-27%` ⛔	`-57%` ⛔	`0` ✔️
Datadog.Trace.Ci.GitInfo	`-17%` ⛔	`-11%` ⛔	`0` ✔️
Datadog.Trace.Configuration.ExporterSettings	`-4%` ⚠️	`-6%` ⛔	`43` ⛔
Datadog.Trace.ClrProfiler.ServerlessInstrumentation.AWS.LambdaCommon	`-1%` ⚠️	`15%` ✔️	`11` ⛔
Datadog.Trace.Ci.CIVisibility	`7%` ✔️	`7%` ✔️	`0` ✔️

View the full reports for further details:

andrewlock · 2022-04-18T23:03:26Z

Benchmarks Report 🐌

Benchmarks for #2686 compared to master:

9 benchmarks are faster, with geometric mean 1.180
All benchmarks have the same 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.AgentWriterBenchmark - Faster 🎉 Same allocations ✔️

Faster 🎉 in #2686

Benchmark	base/diff	Base Median (ns)	Diff Median (ns)	Modality
Benchmarks.Trace.AgentWriterBenchmark.WriteAndFlushEnrichedTraces‑net472	1.286	1,180,268.56	917,805.89

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Allocated
master	`WriteAndFlushEnrichedTraces`	net472	1.18ms	6.29μs	35.6μs	3.16 KB
master	`WriteAndFlushEnrichedTraces`	netcoreapp3.1	818μs	4.72μs	38μs	2.57 KB
#2686	`WriteAndFlushEnrichedTraces`	net472	922μs	4.86μs	23.3μs	3.16 KB
#2686	`WriteAndFlushEnrichedTraces`	netcoreapp3.1	741μs	4.02μs	25.7μs	2.57 KB

Benchmarks.Trace.AppSecBodyBenchmark - Faster 🎉 Same allocations ✔️

Faster 🎉 in #2686

Benchmark	base/diff	Base Median (ns)	Diff Median (ns)	Modality
Benchmarks.Trace.AppSecBodyBenchmark.BodyExtractorSimpleBody‑net472	1.135	482.25	424.79
Benchmarks.Trace.AppSecBodyBenchmark.AllCycleMoreComplexBody‑net472	1.119	325.03	290.52

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Gen 1	Allocated
master	`AllCycleSimpleBody`	net472	332ns	1.52ns	5.9ns	0.0519	0	329 B
master	`AllCycleSimpleBody`	netcoreapp3.1	400ns	2.08ns	9.75ns	0.00454	0	328 B
master	`AllCycleMoreComplexBody`	net472	327ns	1.31ns	5.06ns	0.0482	0	305 B
master	`AllCycleMoreComplexBody`	netcoreapp3.1	394ns	1.93ns	8.64ns	0.00432	0	304 B
master	`BodyExtractorSimpleBody`	net472	482ns	1.97ns	7.36ns	0.0568	0	361 B
master	`BodyExtractorSimpleBody`	netcoreapp3.1	464ns	2.17ns	8.13ns	0.00372	0	272 B
master	`BodyExtractorMoreComplexBody`	net472	27μs	146ns	840ns	1.18	0.0131	7.62 KB
master	`BodyExtractorMoreComplexBody`	netcoreapp3.1	23.1μs	130ns	842ns	0.0895	0	6.75 KB
#2686	`AllCycleSimpleBody`	net472	305ns	1.68ns	10.2ns	0.052	0	329 B
#2686	`AllCycleSimpleBody`	netcoreapp3.1	383ns	2.01ns	10.1ns	0.00457	0	328 B
#2686	`AllCycleMoreComplexBody`	net472	291ns	1.38ns	5.34ns	0.0482	0	305 B
#2686	`AllCycleMoreComplexBody`	netcoreapp3.1	374ns	1.81ns	7ns	0.00424	0	304 B
#2686	`BodyExtractorSimpleBody`	net472	430ns	2.28ns	14.8ns	0.057	0	361 B
#2686	`BodyExtractorSimpleBody`	netcoreapp3.1	446ns	2.32ns	11.1ns	0.00378	0	272 B
#2686	`BodyExtractorMoreComplexBody`	net472	23.9μs	95.1ns	392ns	1.18	0.0119	7.62 KB
#2686	`BodyExtractorMoreComplexBody`	netcoreapp3.1	20.9μs	107ns	477ns	0.0928	0	6.75 KB

Benchmarks.Trace.AspNetCoreBenchmark - Faster 🎉 Same allocations ✔️

Faster 🎉 in #2686

Benchmark	base/diff	Base Median (ns)	Diff Median (ns)	Modality
Benchmarks.Trace.AspNetCoreBenchmark.SendRequest‑netcoreapp3.1	1.134	342,424.35	302,044.84

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`SendRequest`	net472	0ns	0ns	0ns	0	0 b
master	`SendRequest`	netcoreapp3.1	343μs	1.62μs	7.08μs	0.179	19.74 KB
#2686	`SendRequest`	net472	0ns	0ns	0ns	0	0 b
#2686	`SendRequest`	netcoreapp3.1	300μs	1.62μs	9.32μs	0.147	19.74 KB

Benchmarks.Trace.DbCommandBenchmark - Faster 🎉 Same allocations ✔️

Faster 🎉 in #2686

Benchmark	base/diff	Base Median (ns)	Diff Median (ns)	Modality
Benchmarks.Trace.DbCommandBenchmark.ExecuteNonQuery‑net472	1.169	2,026.12	1,733.17

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`ExecuteNonQuery`	net472	2.02μs	11.4ns	70.9ns	0.0943	594 B
master	`ExecuteNonQuery`	netcoreapp3.1	1.69μs	8.97ns	44.8ns	0.00883	632 B
#2686	`ExecuteNonQuery`	net472	1.74μs	10.1ns	78.7ns	0.0938	594 B
#2686	`ExecuteNonQuery`	netcoreapp3.1	1.51μs	8.12ns	41.4ns	0.00889	632 B

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

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Gen 1	Allocated
master	`CallElasticsearch`	net472	2.82μs	13.5ns	52.2ns	0.126	0	803 B
master	`CallElasticsearch`	netcoreapp3.1	1.87μs	8.79ns	35.2ns	0.0106	0	792 B
master	`CallElasticsearchAsync`	net472	3.5μs	20.2ns	169ns	0.146	0	939 B
master	`CallElasticsearchAsync`	netcoreapp3.1	1.96μs	7.14ns	25.8ns	0.0126	0	912 B
#2686	`CallElasticsearch`	net472	2.72μs	13.9ns	68ns	0.127	0	803 B
#2686	`CallElasticsearch`	netcoreapp3.1	1.68μs	7.88ns	30.5ns	0.0112	0	792 B
#2686	`CallElasticsearchAsync`	net472	3.1μs	13.8ns	66.4ns	0.149	0.00155	939 B
#2686	`CallElasticsearchAsync`	netcoreapp3.1	1.92μs	8.71ns	34.8ns	0.0124	0	912 B

Benchmarks.Trace.GraphQLBenchmark - Faster 🎉 Same allocations ✔️

Faster 🎉 in #2686

Benchmark	base/diff	Base Median (ns)	Diff Median (ns)	Modality
Benchmarks.Trace.GraphQLBenchmark.ExecuteAsync‑net472	1.131	3,618.94	3,200.85

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Gen 1	Allocated
master	`ExecuteAsync`	net472	3.66μs	17.1ns	130ns	0.165	0	1.06 KB
master	`ExecuteAsync`	netcoreapp3.1	2.21μs	12.8ns	102ns	0.0147	0	1.03 KB
#2686	`ExecuteAsync`	net472	3.21μs	15.7ns	64.9ns	0.168	0.00158	1.06 KB
#2686	`ExecuteAsync`	netcoreapp3.1	2.01μs	7.42ns	27.8ns	0.0147	0	1.03 KB

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

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`SendAsync`	net472	7.57μs	42.8ns	280ns	0.356	2.28 KB
master	`SendAsync`	netcoreapp3.1	4.97μs	17.3ns	64.6ns	0.0314	2.21 KB
#2686	`SendAsync`	net472	6.96μs	37.2ns	197ns	0.36	2.28 KB
#2686	`SendAsync`	netcoreapp3.1	4.8μs	19.5ns	75.4ns	0.0309	2.21 KB

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

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`EnrichedLog`	net472	3.74μs	20.2ns	121ns	0.228	1.45 KB
master	`EnrichedLog`	netcoreapp3.1	3.29μs	17.3ns	91.5ns	0.0223	1.53 KB
#2686	`EnrichedLog`	net472	3.35μs	15.8ns	61.4ns	0.228	1.45 KB
#2686	`EnrichedLog`	netcoreapp3.1	3.19μs	17.5ns	132ns	0.0207	1.53 KB

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

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Gen 1	Allocated
master	`EnrichedLog`	net472	301μs	1.62μs	8.57μs	0.437	0.146	4.33 KB
master	`EnrichedLog`	netcoreapp3.1	238μs	1.16μs	4.76μs	0	0	4.21 KB
#2686	`EnrichedLog`	net472	284μs	940ns	3.64μs	0.423	0.141	4.33 KB
#2686	`EnrichedLog`	netcoreapp3.1	227μs	463ns	1.6μs	0	0	4.21 KB

Benchmarks.Trace.NLogBenchmark - Faster 🎉 Same allocations ✔️

Faster 🎉 in #2686

Benchmark	base/diff	Base Median (ns)	Diff Median (ns)	Modality
Benchmarks.Trace.NLogBenchmark.EnrichedLog‑net472	1.221	9,041.15	7,405.58	several?

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`EnrichedLog`	net472	9.04μs	47.5ns	251ns	0.506	3.23 KB
master	`EnrichedLog`	netcoreapp3.1	7.06μs	37.2ns	193ns	0.0517	3.6 KB
#2686	`EnrichedLog`	net472	7.41μs	36.6ns	155ns	0.504	3.23 KB
#2686	`EnrichedLog`	netcoreapp3.1	7.15μs	29.6ns	111ns	0.0498	3.6 KB

Benchmarks.Trace.RedisBenchmark - Faster 🎉 Same allocations ✔️

Faster 🎉 in #2686

Benchmark	base/diff	Base Median (ns)	Diff Median (ns)	Modality
Benchmarks.Trace.RedisBenchmark.SendReceive‑net472	1.202	2,881.07	2,396.76	several?

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`SendReceive`	net472	2.88μs	16.9ns	165ns	0.158	1.01 KB
master	`SendReceive`	netcoreapp3.1	2.05μs	9.83ns	38.1ns	0.0142	1.01 KB
#2686	`SendReceive`	net472	2.41μs	12.7ns	63.5ns	0.16	1.01 KB
#2686	`SendReceive`	netcoreapp3.1	2.02μs	6.02ns	21.7ns	0.0142	1.01 KB

Benchmarks.Trace.SerilogBenchmark - Faster 🎉 Same allocations ✔️

Faster 🎉 in #2686

Benchmark	base/diff	Base Median (ns)	Diff Median (ns)	Modality
Benchmarks.Trace.SerilogBenchmark.EnrichedLog‑net472	1.237	7,275.28	5,882.86

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`EnrichedLog`	net472	7.29μs	36.1ns	153ns	0.292	1.87 KB
master	`EnrichedLog`	netcoreapp3.1	5.84μs	28.6ns	125ns	0.0187	1.49 KB
#2686	`EnrichedLog`	net472	5.96μs	32.4ns	174ns	0.291	1.87 KB
#2686	`EnrichedLog`	netcoreapp3.1	5.18μs	14.1ns	51ns	0.0209	1.49 KB

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

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`StartFinishSpan`	net472	1.01μs	3.63ns	14ns	0.0716	457 B
master	`StartFinishSpan`	netcoreapp3.1	947ns	5.08ns	25.9ns	0.00644	456 B
master	`StartFinishScope`	net472	1.15μs	5.87ns	27.5ns	0.0842	538 B
master	`StartFinishScope`	netcoreapp3.1	1.16μs	5.11ns	19.8ns	0.00782	576 B
#2686	`StartFinishSpan`	net472	961ns	4.97ns	23.3ns	0.0719	457 B
#2686	`StartFinishSpan`	netcoreapp3.1	876ns	4.18ns	16.7ns	0.00622	456 B
#2686	`StartFinishScope`	net472	1.08μs	3.41ns	12.8ns	0.0843	538 B
#2686	`StartFinishScope`	netcoreapp3.1	1.05μs	6.06ns	50.7ns	0.00797	576 B

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

Raw results

Branch	Method	Toolchain	Mean	StdError	StdDev	Gen 0	Allocated
master	`RunOnMethodBegin`	net472	1.37μs	5.12ns	17.7ns	0.0839	538 B
master	`RunOnMethodBegin`	netcoreapp3.1	1.22μs	6.28ns	39.7ns	0.00791	576 B
#2686	`RunOnMethodBegin`	net472	1.36μs	6.64ns	27.4ns	0.0841	538 B
#2686	`RunOnMethodBegin`	netcoreapp3.1	1.22μs	6.21ns	30.4ns	0.00835	576 B

andrewlock

I regret suggesting this as the way to go

andrewlock · 2022-04-21T08:13:37Z

+        [Fact] // Fails for now because of the 20 000
+        public void TracesPipeName()
+        {
+            var param = "/var/path";


nit: you may want to use a Windows file path here for testing, as that's what's used in practice AFAIK e.g. @"C:\temp\someval"

andrewlock · 2022-04-21T08:15:08Z

+        public void MetricsUnixDomainSocketPath()
+        {
+            var param = "/var/path";
+            var settings = new ExporterSettings() { MetricsUnixDomainSocketPath = param };


AFAICT we don't have a test here for the "non-metrics" UDS? we have http +

named pipes

named pipes metrics

UDS metrics

EDIT: oh wait, lower down?

andrewlock · 2022-04-21T08:26:54Z

+            Assert.Equal(expected: TracesTransportType.Default, actual: settings.TracesTransport);
+            Assert.Equal(expected: expectedUri, actual: settings.AgentUri);
+            Assert.False(string.Equals(settings.AgentUri.Host, "localhost", StringComparison.OrdinalIgnoreCase));


nit: we generally prefer the Should() style this days. No need to change them, just for the future!

andrewlock · 2022-04-21T08:28:21Z

+            Assert.Equal(expected: TracesTransportType.UnixDomainSocket, actual: settings.TracesTransport);
+            Assert.Equal(expected: socketPath, actual: settings.TracesUnixDomainSocketPath);
+            Assert.NotNull(settings.AgentUri);
+            Assert.False(string.Equals(settings.AgentUri.Host, "localhost", StringComparison.OrdinalIgnoreCase));


I think we should be explicitly checking the URI here is either socketPath or $"unix//{socketPath}". Same for the named pipe

andrewlock · 2022-04-21T08:32:04Z

+            // Check the parameters in order of precedence
+            // For some cases, we allow falling back on another configuration (eg invalid url as the application will need to be restarted to fix it anyway).
+            // For other cases (eg a configured unix domain socket path not found), we don't fallback as the problem could be fixed outside the application.
+            if (agentUri is not null)


nit: maybe string.IsNullOrWhiteSpace(agentUri)?

andrewlock · 2022-04-21T08:46:08Z

-                DogStatsdPort = dogStatsdPort ?? DefaultDogstatsdPort;
-                metricsTransport = MetricsTransportType.UDP;
+                SetUdsAsTraceTransport(DefaultTracesUnixDomainSocket);
+                SetAgentUri(DefaultAgentHost, DefaultAgentPort);


No, this isn't right is it?🤔 We should set the Agent URI to DefaultTracesUnixDomainSocket?

andrewlock · 2022-04-21T08:50:14Z

-            TracesPipeName = source?.GetString(ConfigurationKeys.TracesPipeName);
+        private bool TrySetAgentUriAndTransport(string url)
+        {
+            if (!Uri.TryCreate(url, UriKind.RelativeOrAbsolute, out var uri))


If url starts with unix://, will this fail? 🤔?

Nope, I've even added a UTest to make sure it works

andrewlock · 2022-04-21T08:57:38Z

+            // Still build the Uri no matter what the transport as we send it in the http message
+            // TBH, I don't know if we should handle the case where we use agentHost or agentPort.
+            // Can user have configured both agenthost, agentport, and a UDS path (or an url or a pipe)?


If we have set UDS path, I think AgentUri should also be the UDS path. In #2688 we don't necessarily use the AgentUri directly in the request, so that's not a concern.

andrewlock · 2022-04-21T08:58:45Z

+        public UnixDomainSocketConfig(string traces)
+        {
+            Traces = traces;
+        }


Suggested change

public UnixDomainSocketConfig(string traces)

{

Traces = traces;

}

public UnixDomainSocketConfig(string traces)

: base(traces, metrics: null)

andrewlock · 2022-04-21T09:00:40Z

+    /// This class has many flaws that we shall change the pattern in our next major version. That said, as of now, here's how it works:
+    /// You can set configuration at construction or through properties.
+    /// There's an order of precedence between the configuration knobs (Uri is considered first for instance)
+    /// If you try to override through code a configuration, the order of precedence still prevails. Then, if you had set an Uri at construction, setting anything else through code will not have any effect.


I don't think we should include this extra. It's brutally honest, but not what I expect in docs 😂

This comment has been minimized.

Sign in to view

pierotibou force-pushed the pierre/rework-exporter-settings branch 2 times, most recently from a9fc4c6 to a55547d Compare April 15, 2022 11:04

This comment has been minimized.

Sign in to view

Add more tests

94c1696

This comment has been minimized.

Sign in to view

pierotibou force-pushed the pierre/rework-exporter-settings branch from 7e2cf43 to a0a1063 Compare April 15, 2022 21:31

pierotibou added 2 commits April 16, 2022 00:51

Simplify UnixDomainSocketConfig in Tests

7f61d29

Remove references of DD_APM_RECEIVER_SOCKET

a9fbc95

pierotibou force-pushed the pierre/rework-exporter-settings branch from a0a1063 to b7f3ae5 Compare April 15, 2022 22:52

pierotibou marked this pull request as ready for review April 15, 2022 22:58

pierotibou requested a review from a team as a code owner April 15, 2022 22:58

pierotibou added type:bug area:tracer The core tracer library (Datadog.Trace, does not include OpenTracing, native code, or integrations) type:refactor labels Apr 15, 2022

This comment has been minimized.

Sign in to view

pierotibou added 2 commits April 18, 2022 23:19

Fix the properties by recomputing parameters

cb0e1fd

Prevent parial trust issue

5332cf0

pierotibou force-pushed the pierre/rework-exporter-settings branch from b7f3ae5 to 5332cf0 Compare April 18, 2022 21:32

andrewlock mentioned this pull request Apr 20, 2022

If DirectLogSubmission is enabled, enable Logs Injection by default #2697

Merged

andrewlock reviewed Apr 21, 2022

View reviewed changes

pierotibou mentioned this pull request Apr 27, 2022

[Tracer] Fix precedence in ExporterSettings #2737

Merged

pierotibou closed this May 2, 2022

Uh oh!

Conversation

pierotibou commented Apr 14, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary of changes

Reason for change

Implementation details

Test coverage

Other details

Next steps

Uh oh!

This comment has been minimized.

This comment has been minimized.

This comment has been minimized.

This comment has been minimized.

andrewlock commented Apr 18, 2022

Code Coverage Report 📊

Datadog.Trace Breakdown ✔️

Uh oh!

andrewlock commented Apr 18, 2022

Benchmarks Report 🐌

Benchmark details

Faster 🎉 in #2686

Raw results

Faster 🎉 in #2686

Raw results

Faster 🎉 in #2686

Raw results

Faster 🎉 in #2686

Raw results

Raw results

Faster 🎉 in #2686

Raw results

Raw results

Raw results

Raw results

Faster 🎉 in #2686

Raw results

Faster 🎉 in #2686

Raw results

Faster 🎉 in #2686

Raw results

Raw results

Raw results

Uh oh!

andrewlock left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

pierotibou commented Apr 14, 2022 •

edited

Loading