grpc
/
grpc-java
mirror of https://github.com/grpc/grpc-java.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Fix flakiness of testConfigureDeadlineNanoTime with warmup 

Without the warm up I saw large deltas, like 2,262,968ns and
1,712,558ns, on my machine. With the single-line warm up the deltas
decreased dramitically, like 385ns and 536ns. Since our times are so
much better now, decreasing the required delta to 10ms seems reasonable.

This would seem to support the theory that the flakiness was caused by
the class loader, which may even be doing I/O.

Fixes #1646
This commit is contained in:
Eric Anderson 2016-05-06 08:54:47 -07:00
parent 91b087e526
commit 83bf5e6f42
1 changed files with 3 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
interop-testing/src/test/java/io/grpc/stub/StubConfigTest.java
View File

@ -107,11 +107,13 @@ public class StubConfigTest {
// Create a default stub
TestServiceGrpc.TestServiceBlockingStub stub = TestServiceGrpc.newBlockingStub(channel);
assertNull(stub.getCallOptions().getDeadlineNanoTime());
// Warm up JVM
stub.withDeadlineNanoTime(deadline);
// Reconfigure it
TestServiceGrpc.TestServiceBlockingStub reconfiguredStub = stub.withDeadlineNanoTime(deadline);
// New altered config
assertNotNull(reconfiguredStub.getCallOptions().getDeadlineNanoTime());
long maxDelta = MILLISECONDS.toNanos(30);
long maxDelta = MILLISECONDS.toNanos(10);
long actualDelta = Math.abs(reconfiguredStub.getCallOptions().getDeadlineNanoTime() - deadline);
assertTrue(maxDelta + " < " + actualDelta, maxDelta >= actualDelta);
// Default config unchanged