core: Added changes to DelayedStream.setStream() should cancel the provided stream if not using it by Sangamesh1997 · Pull Request #11969 · grpc/grpc-java

Sangamesh1997 · 2025-03-19T06:49:48Z

core: Added changes to DelayedStream.setStream() should cancel the provided stream if not using it

…ovided stream if not using it

vinodhabib · 2025-03-19T08:19:32Z

core/src/test/java/io/grpc/internal/DelayedClientTransportTest.java

  }

+  @Test
+  public void newStreamThenShutDownNow() {


append test with UT method name -> testNewStreamThenShutDownNow()

vinodhabib · 2025-03-19T08:21:06Z

core/src/test/java/io/grpc/internal/DelayedClientTransportTest.java

+    ClientStream stream = delayedTransport.newStream(
+            method, new Metadata(), CallOptions.DEFAULT, tracers);
+    stream.start(streamListener);
+    assertEquals(1,delayedTransport.getPendingStreamsCount());


add an empty line above asserts statements in all applicable places.

vinodhabib

Added some minor comments, please check.

…ovided stream if not using it

vinodhabib · 2025-03-21T06:46:52Z

core/src/main/java/io/grpc/internal/DelayedStream.java

+      if (oldStream != null && !cancelOldStream) {
        return null;
      }
+


remove an empty line here

vinodhabib

Changes LGTM

AgraVator · 2025-04-02T10:44:21Z

core/src/main/java/io/grpc/internal/DelayedStream.java

+    boolean cancelOldStream = false;
+
    synchronized (this) {
-      // If realStream != null, then either setStream() or cancel() has been called.


Has this changed ?

No, added back the removed comment, it was accidentally missed while updating the code, but the logic still holds. if realStream() != null, it means setStream() was called, and if listener !=null, the old stream may have been cancelled.

AgraVator · 2025-04-02T10:48:09Z

core/src/test/java/io/grpc/internal/DelayedClientTransportTest.java

+    ClientStream stream = delayedTransport.newStream(
+            method, new Metadata(), CallOptions.DEFAULT, tracers);
+    stream.start(streamListener);
+    assertEquals(1,delayedTransport.getPendingStreamsCount());


Please add a space after the comma...

AgraVator · 2025-04-02T10:48:16Z

core/src/test/java/io/grpc/internal/DelayedClientTransportTest.java

+    verify(streamListener).closed(
+            statusCaptor.capture(), any(RpcProgress.class), any(Metadata.class));
+
+    assertEquals(0,delayedTransport.getPendingStreamsCount());


Please add a space after the comma...

This allows Filters to access the xds configuration for their own processing. From gRFC A83: > This data is available via the XdsConfig attribute introduced in A74. > If the xDS ConfigSelector is not already passing that attribute to the > filters, it will need to be changed to do so.

It is much harder to debug refcounting problems when we ignore impossible situations. So make such impossible cases complain loudly so the bug is obvious.

This is to support gRFC A83 xDS GCP Authentication Filter: > Otherwise, the filter will look in the CDS resource's metadata for a > key corresponding to the filter's instance name.

…ore (grpc#11962)

…rpc#11951)

panic() calls a good amount of code, so it could get another exception. The SynchronizationContext is running on an arbitrary thread and we don't want to propagate this secondary exception up its stack (to be handled by its UncaughtExceptionHandler); it we wanted that we'd propagate the original exception. This second exception will only be seen in the logs; the first exception was logged and will be used to fail RPCs. Also related to http://yaqs/8493785598685872128 and b692b9d

…already have their connections cancelled (grpc#11934) Some clients watching health status can cancel their watch and `HealthService` when trying to notify these watchers were getting CANCELLED exception because there was no cancellation handler set on the `StreamObserver`. This change sets the cancellation handler that removes the watcher from the set of watcher clients to be notified of the health status.

…verTest

…ction (grpc#11977) Fixes grpc#11975

Previously it would wait for the new LB to enter READY. However, that prevents there being an upper-bound on how long the old policy will continue to be used. The point of graceful switch is to avoid RPCs seeing increased latency when we swap config. We don't want it to prevent the system from becoming eventually consistent.

I think at some point there were more usages in the tests. But now it is pretty easy. PriorityLb.ChildLbState.picker is initialized to FixedResultPicker(NoResult). So now that GracefulSwitchLb is using the same picker, equals() is able to de-dup an update.

…CING_CONFIG (grpc#11982) ATTR_LOAD_BALANCING_CONFIG was deleted in bf7a42d.

"EXP" stood for experimental and all documentation that referenced it made it clear it was experimental. It's been some years since we started logging a message when it was used to say it will be deleted. There's no time like the present to delete it.

…c#11972)

…ovided stream if not using it

This reverts commit 67b2f3a.

…ed_1537 # Conflicts: # core/src/test/java/io/grpc/internal/DelayedClientTransportTest.java

Sangamesh1997 · 2025-04-08T09:11:38Z

@AgraVator Gentle reminder on review..

ejona86 · 2025-04-28T14:42:11Z

core/src/main/java/io/grpc/internal/DelayedStream.java

        return null;
      }
+      if (cancelOldStream) {
+        oldStream.cancel(Status.CANCELLED.withDescription("Replaced by a new Stream"));


No. We've already chosen a stream. We can't change to a different stream now. I don't understand why this is being done this way, especially since #1537 said that the stream passed into this function would be the one cancelled.

ejona86 · 2025-04-28T14:47:29Z

core/src/main/java/io/grpc/internal/DelayedClientTransport.java

    }

+    @Override
+    public void start(ClientStreamListener listener) {


I need to look into why this change was made. It looks like it delays registering the stream in pendingStreams to avoid createRealStream() from being called. We may need to double-check the transport state here, because there are state checks before createPendingStream() that are now out-of-date.

core: Added changes to DelayedStream.setStream() should cancel the pr…

270567d

…ovided stream if not using it

vinodhabib reviewed Mar 19, 2025

View reviewed changes

Sangamesh1997 added 2 commits March 19, 2025 16:10

Merge branch 'grpc:master' into Issue_fixed_1537

072d033

core: Added changes to DelayedStream.setStream() should cancel the pr…

8ac9678

…ovided stream if not using it

vinodhabib reviewed Mar 21, 2025

View reviewed changes

core: Fixed internal review points

cac52d4

Sangamesh1997 marked this pull request as ready for review March 21, 2025 07:02

vinodhabib approved these changes Mar 21, 2025

View reviewed changes

Sangamesh1997 requested a review from vinodhabib March 26, 2025 10:39

AgraVator reviewed Apr 2, 2025

View reviewed changes

ejona86 and others added 16 commits April 3, 2025 06:48

xds: Assert XdsNR's cluster ref counting is consistent

8cd4d5a

It is much harder to debug refcounting problems when we ignore impossible situations. So make such impossible cases complain loudly so the bug is obvious.

xds: Expose filter names to filter instances (grpc#11971)

2295bbe

This is to support gRFC A83 xDS GCP Authentication Filter: > Otherwise, the filter will look in the CDS resource's metadata for a > key corresponding to the filter's instance name.

Replace usages of deprecated ExpectedException in grpc-api and grpc-c…

9e6ece3

…ore (grpc#11962)

otel tracing: fix span names (grpc#11974)

77216bf

xds: add support for custom per-target credentials on the transport (g…

d28fcb2

…rpc#11951)

fix: cleans up FileWatcherCertificateProvider in XdsSecurityClientSer…

61bb878

…verTest

core: Use java.time.Time.getNano in InstantTimeProvider without refle…

3a35a76

…ction (grpc#11977) Fixes grpc#11975

Start 1.73.0 development cycle (grpc#11987)

f828a3c

okhttp: Per-rpc call option authority verification (grpc#11754)

8c7cf53

core: Delete stale SuppressWarnings("deprecated") for ATTR_LOAD_BALAN…

8879e94

…CING_CONFIG (grpc#11982) ATTR_LOAD_BALANCING_CONFIG was deleted in bf7a42d.

shivaspeaks and others added 10 commits April 3, 2025 06:48

xds: propagate audience from cluster resource in gcp auth filter (grp…

0ab9e11

…c#11972)

core: Fixed internal review points

b7df168

core: Added changes to DelayedStream.setStream() should cancel the pr…

58f39fa

…ovided stream if not using it

core: Added changes to DelayedStream.setStream() should cancel the pr…

013948d

…ovided stream if not using it

core: Fixed internal review points

a2220eb

core: Fixed internal review points

67b2f3a

Revert "core: Fixed internal review points"

696dd52

This reverts commit 67b2f3a.

Merge remote-tracking branch 'origin/Issue_fixed_1537' into Issue_fix…

6beb245

…ed_1537 # Conflicts: # core/src/test/java/io/grpc/internal/DelayedClientTransportTest.java

Merge branch 'master' into Issue_fixed_1537

17e6b41

core: Fixed internal review points

d1602ad

Sangamesh1997 requested a review from AgraVator April 3, 2025 10:53

Sangamesh1997 added 3 commits April 9, 2025 13:36

Merge branch 'grpc:master' into Issue_fixed_1537

163c958

Merge branch 'grpc:master' into Issue_fixed_1537

11adf6f

Merge branch 'grpc:master' into Issue_fixed_1537

529a75e

shivaspeaks requested review from ejona86 and removed request for vinodhabib April 28, 2025 10:05

ejona86 reviewed Apr 28, 2025

View reviewed changes

ejona86 closed this May 29, 2025

github-actions bot locked as resolved and limited conversation to collaborators Aug 28, 2025

Conversation

Sangamesh1997 commented Mar 19, 2025

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

vinodhabib Mar 19, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

vinodhabib 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!

vinodhabib left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Sangamesh1997 Apr 3, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

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!

Sangamesh1997 commented Apr 8, 2025

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!

10 participants

vinodhabib Mar 19, 2025 •

edited

Loading

Sangamesh1997 Apr 3, 2025 •

edited

Loading