-
Notifications
You must be signed in to change notification settings - Fork 4.7k
/
IpcTraceTest.cs
417 lines (377 loc) · 17.5 KB
/
IpcTraceTest.cs
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
// Licensed to the .NET Foundation under one or more agreements.
// The .NET Foundation licenses this file to you under the MIT license.
using System;
using System.Collections.Generic;
using System.Collections.Concurrent;
using System.Diagnostics;
using System.Diagnostics.Tracing;
using System.IO;
using System.Linq;
using System.Runtime.InteropServices;
using System.Text.RegularExpressions;
using System.Threading;
using System.Threading.Tasks;
using Microsoft.Diagnostics.NETCore.Client;
using Microsoft.Diagnostics.Tracing;
namespace Tracing.Tests.Common
{
public class Logger
{
public static Logger logger = new Logger();
private TextWriter _log;
private Stopwatch _sw;
public Logger(TextWriter log = null)
{
_log = log ?? Console.Out;
_sw = new Stopwatch();
}
public void Log(string message)
{
if (!_sw.IsRunning)
_sw.Start();
_log.WriteLine($"{_sw.Elapsed.TotalSeconds,5:f1}s: {message}");
}
}
public class ExpectedEventCount
{
// The acceptable percent error on the expected value
// represented as a floating point value in [0,1].
public float Error { get; private set; }
// The expected count of events. A value of -1 indicates
// that count does not matter, and we are simply testing
// that the provider exists in the trace.
public int Count { get; private set; }
public ExpectedEventCount(int count, float error = 0.0f)
{
Count = count;
Error = error;
}
public bool Validate(int actualValue)
{
return Count == -1 || CheckErrorBounds(actualValue);
}
public bool CheckErrorBounds(int actualValue)
{
return Math.Abs(actualValue - Count) <= (Count * Error);
}
public static implicit operator ExpectedEventCount(int i)
{
return new ExpectedEventCount(i);
}
public override string ToString()
{
return $"{Count} +- {Count * Error}";
}
}
// This event source is used by the test infra to
// to insure that providers have finished being enabled
// for the session being observed. Since the client API
// returns the pipe for reading _before_ it finishes
// enabling the providers to write to that session,
// we need to guarantee that our providers are on before
// sending events. This is a _unique_ problem I imagine
// should _only_ affect scenarios like these tests
// where the reading and sending of events are required
// to synchronize.
public sealed class SentinelEventSource : EventSource
{
private SentinelEventSource() {}
public static SentinelEventSource Log = new SentinelEventSource();
public void SentinelEvent() { WriteEvent(1, "SentinelEvent"); }
}
public class IpcTraceTest
{
// This Action is executed while the trace is being collected.
private Action _eventGeneratingAction;
// A dictionary of event providers to number of events.
// A count of -1 indicates that you are only testing for the presence of the provider
// and don't care about the number of events sent
private Dictionary<string, ExpectedEventCount> _expectedEventCounts;
private Dictionary<string, int> _actualEventCounts = new Dictionary<string, int>();
// A function to be called with the EventPipeEventSource _before_
// the call to `source.Process()`. The function should return another
// function that will be called to check whether the optional test was validated.
// Example in situ: providervalidation.cs
private Func<EventPipeEventSource, Func<int>> _optionalTraceValidator;
/// <summary>
/// This is list of the EventPipe providers to turn on for the test execution
/// </summary>
private List<EventPipeProvider> _testProviders;
/// <summary>
/// This represents the current EventPipeSession
/// </summary>
private EventPipeSession _eventPipeSession;
/// <summary>
/// This is the list of EventPipe providers for the sentinel EventSource that indicates that the process is ready
/// </summary>
private List<EventPipeProvider> _sentinelProviders = new List<EventPipeProvider>()
{
new EventPipeProvider("SentinelEventSource", EventLevel.Verbose, -1)
};
IpcTraceTest(
Dictionary<string, ExpectedEventCount> expectedEventCounts,
Action eventGeneratingAction,
List<EventPipeProvider> providers,
int circularBufferMB,
Func<EventPipeEventSource, Func<int>> optionalTraceValidator = null)
{
_eventGeneratingAction = eventGeneratingAction;
_expectedEventCounts = expectedEventCounts;
_testProviders = providers;
_optionalTraceValidator = optionalTraceValidator;
}
private int Fail(string message = "")
{
Logger.logger.Log("Test FAILED!");
Logger.logger.Log(message);
Logger.logger.Log("Configuration:");
Logger.logger.Log("{");
Logger.logger.Log("\tproviders: [");
Logger.logger.Log("\t]");
Logger.logger.Log("}\n");
Logger.logger.Log("Expected:");
Logger.logger.Log("{");
foreach (var (k, v) in _expectedEventCounts)
{
Logger.logger.Log($"\t\"{k}\" = {v}");
}
Logger.logger.Log("}\n");
Logger.logger.Log("Actual:");
Logger.logger.Log("{");
foreach (var (k, v) in _actualEventCounts)
{
Logger.logger.Log($"\t\"{k}\" = {v}");
}
Logger.logger.Log("}");
return -1;
}
private int Validate(bool enableRundownProvider = true)
{
// FIXME: This is a bandaid fix for a deadlock in EventPipeEventSource caused by
// the lazy caching in the Regex library. The caching creates a ConcurrentDictionary
// and because it is the first one in the process, it creates an EventSource which
// results in a deadlock over a lock in EventPipe. These lines should be removed once the
// underlying issue is fixed by forcing these events to try to be written _before_ we shutdown.
//
// see: https://github.com/dotnet/runtime/pull/1794 for details on the issue
//
var emptyConcurrentDictionary = new ConcurrentDictionary<string, string>();
emptyConcurrentDictionary["foo"] = "bar";
var __count = emptyConcurrentDictionary.Count;
var isClean = IpcTraceTest.EnsureCleanEnvironment();
if (!isClean)
return -1;
// CollectTracing returns before EventPipe::Enable has returned, so the
// the sources we want to listen for may not have been enabled yet.
// We'll use this sentinel EventSource to check if Enable has finished
ManualResetEvent sentinelEventReceived = new ManualResetEvent(false);
var sentinelTask = new Task(() =>
{
Logger.logger.Log("Started sending sentinel events...");
while (!sentinelEventReceived.WaitOne(50))
{
SentinelEventSource.Log.SentinelEvent();
}
Logger.logger.Log("Stopped sending sentinel events");
});
sentinelTask.Start();
int processId = Process.GetCurrentProcess().Id;
object threadSync = new object(); // for locking eventpipeSession access
Func<int> optionalTraceValidationCallback = null;
DiagnosticsClient client = new DiagnosticsClient(processId);
#if DIAGNOSTICS_RUNTIME
if (OperatingSystem.IsAndroid() || OperatingSystem.IsIOS() || OperatingSystem.IsTvOS())
client = new DiagnosticsClient(new IpcEndpointConfig("127.0.0.1:9000", IpcEndpointConfig.TransportType.TcpSocket, IpcEndpointConfig.PortType.Listen));
#endif
var readerTask = new Task(() =>
{
Logger.logger.Log("Connecting to EventPipe...");
try
{
_eventPipeSession = client.StartEventPipeSession(_testProviders.Concat(_sentinelProviders), enableRundownProvider);
}
catch (DiagnosticsClientException ex)
{
Logger.logger.Log("Failed to connect to EventPipe!");
Logger.logger.Log(ex.ToString());
throw new ApplicationException("Failed to connect to EventPipe");
}
using (var eventPipeStream = new StreamProxy(_eventPipeSession.EventStream))
{
Logger.logger.Log("Creating EventPipeEventSource...");
using (EventPipeEventSource source = new EventPipeEventSource(eventPipeStream))
{
Logger.logger.Log("EventPipeEventSource created");
source.Dynamic.All += (eventData) =>
{
try
{
if (eventData.ProviderName == "SentinelEventSource")
{
if (!sentinelEventReceived.WaitOne(0))
Logger.logger.Log("Saw sentinel event");
sentinelEventReceived.Set();
}
else if (_actualEventCounts.TryGetValue(eventData.ProviderName, out _))
{
_actualEventCounts[eventData.ProviderName]++;
}
else
{
Logger.logger.Log($"Saw new provider '{eventData.ProviderName}'");
_actualEventCounts[eventData.ProviderName] = 1;
}
}
catch (Exception e)
{
Logger.logger.Log("Exception in Dynamic.All callback " + e.ToString());
}
};
Logger.logger.Log("Dynamic.All callback registered");
if (_optionalTraceValidator != null)
{
Logger.logger.Log("Running optional trace validator");
optionalTraceValidationCallback = _optionalTraceValidator(source);
Logger.logger.Log("Finished running optional trace validator");
}
Logger.logger.Log("Starting stream processing...");
try
{
source.Process();
}
catch (Exception)
{
Logger.logger.Log($"Exception thrown while reading; dumping culprit stream to disk...");
eventPipeStream.DumpStreamToDisk();
// rethrow it to fail the test
throw;
}
Logger.logger.Log("Stopping stream processing");
Logger.logger.Log($"Dropped {source.EventsLost} events");
}
}
});
var waitSentinelEventTask = new Task(() => {
sentinelEventReceived.WaitOne();
});
readerTask.Start();
waitSentinelEventTask.Start();
// Will throw if the reader task throws any exceptions before signaling sentinelEventReceived.
Task.WaitAny(readerTask, waitSentinelEventTask);
Logger.logger.Log("Starting event generating action...");
_eventGeneratingAction();
Logger.logger.Log("Stopping event generating action");
var stopTask = Task.Run(() =>
{
Logger.logger.Log("Sending StopTracing command...");
lock (threadSync) // eventpipeSession
{
_eventPipeSession.Stop();
}
Logger.logger.Log("Finished StopTracing command");
});
// Should throw if the reader task throws any exceptions
Task.WaitAll(readerTask, stopTask);
Logger.logger.Log("Reader task finished");
foreach (var (provider, expectedCount) in _expectedEventCounts)
{
if (_actualEventCounts.TryGetValue(provider, out var actualCount))
{
if (!expectedCount.Validate(actualCount))
{
return Fail($"Event count mismatch for provider \"{provider}\": expected {expectedCount}, but saw {actualCount}");
}
}
else
{
return Fail($"No events for provider \"{provider}\"");
}
}
if (optionalTraceValidationCallback != null)
{
Logger.logger.Log("Validating optional callback...");
// reader thread should be dead now, no need to lock
return optionalTraceValidationCallback();
}
else
{
return 100;
}
}
// Ensure that we have a clean environment for running the test.
// Specifically check that we don't have more than one match for
// Diagnostic IPC sockets in the TempPath. These can be left behind
// by bugs, catastrophic test failures, etc. from previous testing.
// The tmp directory is only cleared on reboot, so it is possible to
// run into these zombie pipes if there are failures over time.
// Note: Windows has some guarantees about named pipes not living longer
// the process that created them, so we don't need to check on that platform.
static public bool EnsureCleanEnvironment()
{
if (!OperatingSystem.IsWindows() && !OperatingSystem.IsBrowser() && !OperatingSystem.IsIOS() && !OperatingSystem.IsTvOS())
{
Func<(IEnumerable<IGrouping<int,FileInfo>>, List<int>)> getPidsAndSockets = () =>
{
IEnumerable<IGrouping<int,FileInfo>> currentIpcs = Directory.GetFiles(Path.GetTempPath(), "dotnet-diagnostic*")
.Select(filename => new { pid = int.Parse(Regex.Match(filename, @"dotnet-diagnostic-(?<pid>\d+)").Groups["pid"].Value), fileInfo = new FileInfo(filename) })
.GroupBy(fileInfos => fileInfos.pid, fileInfos => fileInfos.fileInfo);
List<int> currentPids = System.Diagnostics.Process.GetProcesses().Select(pid => pid.Id).ToList();
return (currentIpcs, currentPids);
};
var (currentIpcs, currentPids) = getPidsAndSockets();
foreach (var ipc in currentIpcs)
{
if (!currentPids.Contains(ipc.Key))
{
foreach (FileInfo fi in ipc)
{
Logger.logger.Log($"Attempting to delete the zombied pipe: {fi.FullName}");
fi.Delete();
Logger.logger.Log($"Deleted");
}
}
else
{
if (ipc.Count() > 1)
{
// delete zombied pipes except newest which is owned
var duplicates = ipc.OrderBy(fileInfo => fileInfo.CreationTime.Ticks).SkipLast(1);
foreach (FileInfo fi in duplicates)
{
Logger.logger.Log($"Attempting to delete the zombied pipe: {fi.FullName}");
fi.Delete();
}
}
}
}
}
return true;
}
public static int RunAndValidateEventCounts(
Dictionary<string, ExpectedEventCount> expectedEventCounts,
Action eventGeneratingAction,
List<EventPipeProvider> providers,
int circularBufferMB=1024,
Func<EventPipeEventSource, Func<int>> optionalTraceValidator = null,
bool enableRundownProvider = true)
{
Logger.logger.Log("==TEST STARTING==");
var test = new IpcTraceTest(expectedEventCounts, eventGeneratingAction, providers, circularBufferMB, optionalTraceValidator);
try
{
var ret = test.Validate(enableRundownProvider);
if (ret == 100)
Logger.logger.Log("==TEST FINISHED: PASSED!==");
else
Logger.logger.Log("==TEST FINISHED: FAILED!==");
return ret;
}
catch (Exception e)
{
Logger.logger.Log(e.ToString());
Logger.logger.Log("==TEST FINISHED: FAILED!==");
return -1;
}
}
}
}