Enable some Aeron tests in nightly-builds, #31010 (#31218)

* and enable gc logging via -Dakka.ci-server=true in .jvmopts-ci
* and some cleanup
* exclude ClusterReceptionistSpec and JoinConfigCompatCheckerRollingUpdateSpec
  with new tag

.github/workflows/build-test-prValidation.yml

@@ -56,7 +56,6 @@ jobs:
         run: |-
           sbt -jvm-opts .jvmopts-ci \
           -Dakka.mima.enabled=false \
-          -Dakka.ci-server=true \
           -Dakka.test.multi-in-test=false \
           -Dakka.test.timefactor=2 \
           -Dakka.actor.testkit.typed.timefactor=2 \

.github/workflows/nightly-builds.yml

@@ -94,6 +94,7 @@ jobs:
         uses: coursier/cache-action@v6.2
 
       - name: sbt ${{ matrix.command }}
+        # note that this is not running any multi-jvm tests because multi-in-test=false
         run: |-
           sbt -jvm-opts .jvmopts-ci \
           -Djava.security.egd=file:/dev/./urandom \
@@ -101,13 +102,8 @@ jobs:
           -Dakka.test.timefactor=2 \
           -Dakka.actor.testkit.typed.timefactor=2 \
           -Dakka.test.tags.exclude=gh-exclude,timing \
+          -Dakka.test.multi-in-test=false \
           -Dakka.cluster.assert=on \
-          -Dakka.test.names.exclude=akka.cluster.Stress \
-          -Dmultinode.XX:MetaspaceSize=128M \
-          -Dmultinode.Xms256M \
-          -Dmultinode.Xmx256M \
-          -Dmultinode.Xlog:gc \
-          -Dmultinode.XX:+AlwaysActAsServerClassMachine \
           clean ${{ matrix.command }}
 
       - name: Email on failure
@@ -139,11 +135,11 @@ jobs:
         jdkVersion: ["adopt@1.8.0", "adopt@1.11", "openjdk@1.17.0"]
         include:
           - jdkVersion: adopt@1.8.0
-            extraOpts: "-Dmultinode.XX:+PrintGCDetails -Dmultinode.XX:+PrintGCTimeStamps"
+            extraOpts: ""
           - jdkVersion: adopt@1.11
-            extraOpts: "-Dmultinode.Xlog:gc"
+            extraOpts: ""
           - jdkVersion: openjdk@1.17.0
-            extraopts: "-Dmultinode.Xlog:gc"
+            extraopts: ""
     steps:
       - name: Checkout
         uses: actions/checkout@v2
@@ -159,7 +155,7 @@ jobs:
         uses: coursier/cache-action@v6.2
 
       - name: Compile and Test
-        # note that this is not running any multi-jvm tests (yet) because multi-in-test=false
+        # note that this is not running any multi-jvm tests because multi-in-test=false
         run: |-
           sbt -jvm-opts .jvmopts-ci \
             -Dakka.cluster.assert=on \
@@ -168,10 +164,6 @@ jobs:
             -Dakka.actor.testkit.typed.timefactor=2 \
             -Dakka.test.tags.exclude=gh-exclude,timing \
             -Dakka.test.multi-in-test=false \
-            -Dmultinode.XX:MetaspaceSize=128M \
-            -Dmultinode.Xms256M \
-            -Dmultinode.Xmx256M \
-            -Dmultinode.XX:+AlwaysActAsServerClassMachine \
             ${{ matrix.extraOpts }} \
             clean "+~ ${{ matrix.scalaVersion }} test" checkTestsHaveRun
 
@@ -225,3 +217,58 @@ jobs:
           body: |
             Job ${{ github.job }} in workflow ${{ github.workflow }} of ${{github.repository}} failed!
             https://github.com/${{github.repository}}/actions/runs/${{github.run_id}}
+
+  akka-artery-aeron-tests:
+    name: Akka Artery Aeron Tests
+    runs-on: ubuntu-20.04
+    if: github.repository == 'akka/akka'
+    strategy:
+      fail-fast: false
+      matrix:
+        command:
+          - akka-remote/test akka-remote-tests/test
+          - akka-cluster/test akka-cluster-typed/test
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v2
+        with:
+          # we don't know what commit the last tag was it's safer to get entire repo so previousStableVersion resolves
+          fetch-depth: 0
+
+      - name: Set up JDK 11
+        uses: olafurpg/setup-scala@v10
+        with:
+          java-version: adopt@1.11
+
+      - name: Cache Coursier cache
+        uses: coursier/cache-action@v6.2
+
+      - name: sbt ${{ matrix.command }}
+        # note that this is not running any multi-jvm tests because multi-in-test=false
+        run: |-
+          sbt -jvm-opts .jvmopts-ci \
+          -Djava.security.egd=file:/dev/./urandom \
+          -Dakka.remote.artery.transport=aeron-udp \
+          -Dakka.test.timefactor=2 \
+          -Dakka.actor.testkit.typed.timefactor=2 \
+          -Dakka.test.tags.exclude=gh-exclude,gh-exclude-aeron,timing \
+          -Dakka.test.multi-in-test=false \
+          -Dakka.cluster.assert=on \
+          -Daeron.dir=/opt/volumes/media-driver \
+          -Daeron.term.buffer.length=33554432 \
+          clean ${{ matrix.command }}
+
+      - name: Email on failure
+        if: ${{ failure() }}
+        uses: dawidd6/action-send-mail@v3
+        with:
+          server_address: smtp.gmail.com
+          server_port: 465
+          username: ${{secrets.MAIL_USERNAME}}
+          password: ${{secrets.MAIL_PASSWORD}}
+          subject: "Failed: ${{ github.workflow }} / ${{ github.job }}"
+          to: akka.official@gmail.com
+          from: Akka CI (GHActions)
+          body: |
+            Job ${{ github.job }} in workflow ${{ github.workflow }} of ${{github.repository}} failed!
+            https://github.com/${{github.repository}}/actions/runs/${{github.run_id}}

.github/workflows/scala3-build.yml

@@ -57,7 +57,6 @@ jobs:
           -Dmultinode.XX:MetaspaceSize=128M \
           -Dmultinode.Xms256M \
           -Dmultinode.Xmx256M \
-          -Dmultinode.Xlog:gc \
           -Dmultinode.XX:+AlwaysActAsServerClassMachine \
           "+~ 3 ${{ matrix.command }}"
 

.github/workflows/timing-tests.yml

@@ -39,7 +39,6 @@ jobs:
             -Dmultinode.XX:MetaspaceSize=128M \
             -Dmultinode.Xms256M \
             -Dmultinode.Xmx256M \
-            -Dmultinode.Xlog:gc \
             -Dmultinode.XX:+AlwaysActAsServerClassMachine \
             clean test
 

.jvmopts-ci

@@ -12,3 +12,4 @@
 -XX:MetaspaceSize=512M
 -XX:-ClassUnloadingWithConcurrentMark
 -Djava.security.egd=file:/dev/./urandom
+-Dakka.ci-server=true

akka-cluster-typed/src/test/scala/akka/cluster/typed/internal/receptionist/ClusterReceptionistSpec.scala

@@ -8,9 +8,11 @@ import java.util.concurrent.ThreadLocalRandom
 
 import scala.concurrent.Await
 import scala.concurrent.duration._
+
 import com.typesafe.config.ConfigFactory
 import org.scalatest.matchers.should.Matchers
 import org.scalatest.wordspec.AnyWordSpec
+
 import akka.actor.RootActorPath
 import akka.actor.testkit.typed.FishingOutcome
 import akka.actor.testkit.typed.scaladsl.ActorTestKit
@@ -31,6 +33,8 @@ import akka.serialization.jackson.CborSerializable
 import akka.testkit.LongRunningTest
 import org.scalatest.concurrent.ScalaFutures
 
+import akka.testkit.GHExcludeAeronTest
+
 object ClusterReceptionistSpec {
   val config = ConfigFactory.parseString(s"""
       akka.loglevel = DEBUG # issue #24960
@@ -80,7 +84,7 @@ class ClusterReceptionistSpec extends AnyWordSpec with Matchers with LogCapturin
 
   "The cluster receptionist" must {
 
-    "eventually replicate registrations to the other side" taggedAs (LongRunningTest) in {
+    "eventually replicate registrations to the other side".taggedAs(LongRunningTest, GHExcludeAeronTest) in {
       val testKit1 = ActorTestKit("ClusterReceptionistSpec-test-1", ClusterReceptionistSpec.config)
       val system1 = testKit1.system
       val testKit2 = ActorTestKit(system1.name, system1.settings.config)
@@ -118,7 +122,7 @@ class ClusterReceptionistSpec extends AnyWordSpec with Matchers with LogCapturin
       }
     }
 
-    "handle registrations before joining" taggedAs (LongRunningTest) in {
+    "handle registrations before joining".taggedAs(LongRunningTest, GHExcludeAeronTest) in {
       val testKit1 = ActorTestKit("ClusterReceptionistSpec-test-2", ClusterReceptionistSpec.config)
       val system1 = testKit1.system
       val testKit2 = ActorTestKit(system1.name, system1.settings.config)
@@ -148,11 +152,11 @@ class ClusterReceptionistSpec extends AnyWordSpec with Matchers with LogCapturin
       }
     }
 
-    "remove registrations when node dies" taggedAs (LongRunningTest) in {
+    "remove registrations when node dies".taggedAs(LongRunningTest, GHExcludeAeronTest) in {
       testNodeRemoval(down = true)
     }
 
-    "remove registrations when node leaves" taggedAs (LongRunningTest) in {
+    "remove registrations when node leaves".taggedAs(LongRunningTest, GHExcludeAeronTest) in {
       testNodeRemoval(down = false)
     }
 
@@ -215,7 +219,7 @@ class ClusterReceptionistSpec extends AnyWordSpec with Matchers with LogCapturin
       }
     }
 
-    "not remove registrations when self is shutdown" taggedAs (LongRunningTest) in {
+    "not remove registrations when self is shutdown".taggedAs(LongRunningTest, GHExcludeAeronTest) in {
       val testKit1 = ActorTestKit("ClusterReceptionistSpec-test-4", ClusterReceptionistSpec.config)
       val system1 = testKit1.system
       val testKit2 = ActorTestKit(system1.name, system1.settings.config)
@@ -264,7 +268,7 @@ class ClusterReceptionistSpec extends AnyWordSpec with Matchers with LogCapturin
 
     }
 
-    "work with services registered before node joins cluster" taggedAs (LongRunningTest) in {
+    "work with services registered before node joins cluster".taggedAs(LongRunningTest, GHExcludeAeronTest) in {
       val testKit1 = ActorTestKit("ClusterReceptionistSpec-test-5", ClusterReceptionistSpec.config)
       val system1 = testKit1.system
       val testKit2 = ActorTestKit(system1.name, system1.settings.config)
@@ -322,7 +326,7 @@ class ClusterReceptionistSpec extends AnyWordSpec with Matchers with LogCapturin
       }
     }
 
-    "handle a new incarnation of the same node well" taggedAs (LongRunningTest) in {
+    "handle a new incarnation of the same node well".taggedAs(LongRunningTest, GHExcludeAeronTest) in {
       val testKit1 = ActorTestKit("ClusterReceptionistSpec-test-6", ClusterReceptionistSpec.config)
       val system1 = testKit1.system
       val testKit2 = ActorTestKit(system1.name, system1.settings.config)
@@ -423,7 +427,9 @@ class ClusterReceptionistSpec extends AnyWordSpec with Matchers with LogCapturin
     }
 
     // reproducer of issue #26284
-    "handle a new incarnation of the same node that is no longer part of same cluster" taggedAs (LongRunningTest) in {
+    "handle a new incarnation of the same node that is no longer part of same cluster".taggedAs(
+      LongRunningTest,
+      GHExcludeAeronTest) in {
       val testKit1 = ActorTestKit(
         "ClusterReceptionistSpec-test-7",
         ConfigFactory.parseString("""
@@ -527,7 +533,7 @@ class ClusterReceptionistSpec extends AnyWordSpec with Matchers with LogCapturin
       }
     }
 
-    "not lose removals on concurrent updates to same key" taggedAs (LongRunningTest) in {
+    "not lose removals on concurrent updates to same key".taggedAs(LongRunningTest, GHExcludeAeronTest) in {
       val config = ConfigFactory.parseString("""
           # disable delta propagation so we can have repeatable concurrent writes
           # without delta reaching between nodes already
@@ -598,7 +604,7 @@ class ClusterReceptionistSpec extends AnyWordSpec with Matchers with LogCapturin
       }
     }
 
-    "not conflict with the ClusterClient receptionist default name" taggedAs (LongRunningTest) in {
+    "not conflict with the ClusterClient receptionist default name".taggedAs(LongRunningTest, GHExcludeAeronTest) in {
       val testKit = ActorTestKit(s"ClusterReceptionistSpec-test-9", ClusterReceptionistSpec.config)
       try {
         testKit.system.systemActorOf(Behaviors.ignore, "receptionist")
@@ -607,7 +613,7 @@ class ClusterReceptionistSpec extends AnyWordSpec with Matchers with LogCapturin
       }
     }
 
-    "handle unregistration and re-registration of services" taggedAs (LongRunningTest) in {
+    "handle unregistration and re-registration of services".taggedAs(LongRunningTest, GHExcludeAeronTest) in {
       val testKit1 = ActorTestKit("ClusterReceptionistSpec-test-10", ClusterReceptionistSpec.config)
       val system1 = testKit1.system
       val testKit2 = ActorTestKit(system1.name, system1.settings.config)
@@ -665,7 +671,7 @@ class ClusterReceptionistSpec extends AnyWordSpec with Matchers with LogCapturin
 
     }
 
-    "handle unregistration per key not per actor" taggedAs (LongRunningTest) in {
+    "handle unregistration per key not per actor".taggedAs(LongRunningTest, GHExcludeAeronTest) in {
       val testKit1 = ActorTestKit("ClusterReceptionistSpec-test-11", ClusterReceptionistSpec.config)
       val system1 = testKit1.system
       val testKit2 = ActorTestKit(system1.name, system1.settings.config)
@@ -717,7 +723,7 @@ class ClusterReceptionistSpec extends AnyWordSpec with Matchers with LogCapturin
 
     }
 
-    "handle concurrent unregistration and registration on different nodes" taggedAs (LongRunningTest) in {
+    "handle concurrent unregistration and registration on different nodes".taggedAs(LongRunningTest, GHExcludeAeronTest) in {
       // this covers the fact that with ddata a removal can be lost
       val testKit1 = ActorTestKit("ClusterReceptionistSpec-test-12", ClusterReceptionistSpec.config)
       val system1 = testKit1.system
@@ -776,7 +782,7 @@ class ClusterReceptionistSpec extends AnyWordSpec with Matchers with LogCapturin
     }
     // Fixme concurrent registration and unregistration
 
-    "notify subscribers when registering and joining simultaneously" taggedAs (LongRunningTest) in {
+    "notify subscribers when registering and joining simultaneously".taggedAs(LongRunningTest, GHExcludeAeronTest) in {
       // failing test reproducer for issue #28792
       // It's possible that the registry entry from the ddata update arrives before MemberJoined.
       val config = ConfigFactory.parseString("""
@@ -839,7 +845,7 @@ class ClusterReceptionistSpec extends AnyWordSpec with Matchers with LogCapturin
       }
     }
 
-    "never use durable store" taggedAs (LongRunningTest) in {
+    "never use durable store".taggedAs(LongRunningTest, GHExcludeAeronTest) in {
       val testKit = ActorTestKit("ClusterReceptionistSpec-test-14", ClusterReceptionistSpec.config)
       val system = testKit.system
       try {

akka-cluster/src/test/scala/akka/cluster/JoinConfigCompatCheckerRollingUpdateSpec.scala

@@ -6,7 +6,10 @@ package akka.cluster
 
 import scala.collection.{ immutable => im }
 import scala.concurrent.duration._
+
 import com.typesafe.config.{ Config, ConfigFactory }
+
+import akka.testkit.GHExcludeAeronTest
 import akka.testkit.LongRunningTest
 
 object JoinConfigCompatCheckerRollingUpdateSpec {
@@ -47,14 +50,18 @@ class JoinConfigCompatCheckerRollingUpdateSpec
 
   "A Node" must {
     val timeout = 20.seconds
-    "NOT be allowed to re-join a cluster if it has a new, additional configuration the others do not have and not the old" taggedAs LongRunningTest in {
+    "NOT be allowed to re-join a cluster if it has a new, additional configuration the others do not have and not the old"
+      .taggedAs(LongRunningTest, GHExcludeAeronTest) in {
       // confirms the 2 attempted re-joins fail with both nodes being terminated
       upgradeCluster(3, v1Config, v2ConfigIncompatible, timeout, timeout, enforced = true, shouldRejoin = false)
     }
-    "be allowed to re-join a cluster if it has a new, additional property and checker the others do not have" taggedAs LongRunningTest in {
+    "be allowed to re-join a cluster if it has a new, additional property and checker the others do not have".taggedAs(
+      LongRunningTest,
+      GHExcludeAeronTest) in {
       upgradeCluster(3, v1Config, v2Config, timeout, timeout * 3, enforced = true, shouldRejoin = true)
     }
-    "be allowed to re-join a cluster if it has a new, additional configuration the others do not have and configured to NOT enforce it" taggedAs LongRunningTest in {
+    "be allowed to re-join a cluster if it has a new, additional configuration the others do not have and configured to NOT enforce it"
+      .taggedAs(LongRunningTest, GHExcludeAeronTest) in {
       upgradeCluster(3, v1Config, v2Config, timeout, timeout * 3, enforced = false, shouldRejoin = true)
     }
   }

akka-testkit/src/test/scala/akka/testkit/TestTags.scala

@@ -11,3 +11,4 @@ object LongRunningTest extends Tag("long-running")
 object PerformanceTest extends Tag("performance")
 
 object GHExcludeTest extends Tag("gh-exclude")
+object GHExcludeAeronTest extends Tag("gh-exclude-aeron")