Skip to content

HDDS-13346. Intermittent failure in TestCloseContainer#testContainerChecksumForClosedContainer #8771

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
errose28 merged 1 commit into from
Jul 16, 2025
Merged

HDDS-13346. Intermittent failure in TestCloseContainer#testContainerChecksumForClosedContainer #8771

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
20 changes: 10 additions & 10 deletions ...p-ozone/integration-test/src/test/java/org/apache/hadoop/hdds/scm/TestCloseContainer.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -229,8 +229,8 @@ public void testContainerChecksumForClosedContainer() throws Exception {
// Checksum file exists after container close and matches the expected container
// merkle tree for all the datanodes
for (HddsDatanodeService hddsDatanode : hddsDatanodes) {
GenericTestUtils.waitFor(() -> checkContainerCloseInDatanode(hddsDatanode, containerInfo1), 100, 5000);
assertTrue(containerChecksumFileExists(hddsDatanode, containerInfo1.getContainerID()));
GenericTestUtils.waitFor(() -> checkContainerCloseInDatanode(hddsDatanode, containerInfo1) &&
containerChecksumFileExists(hddsDatanode, containerInfo1.getContainerID()), 100, 5000);
OzoneContainer ozoneContainer = hddsDatanode.getDatanodeStateMachine().getContainer();
Container<?> container1 = ozoneContainer.getController().getContainer(containerInfo1.getContainerID());
ContainerProtos.ContainerChecksumInfo containerChecksumInfo = ContainerMerkleTreeTestUtils.readChecksumFile(
Expand All @@ -256,8 +256,8 @@ public void testContainerChecksumForClosedContainer() throws Exception {
// Checksum file exists after container close and matches the expected container
// merkle tree for all the datanodes
for (HddsDatanodeService hddsDatanode : hddsDatanodes) {
GenericTestUtils.waitFor(() -> checkContainerCloseInDatanode(hddsDatanode, containerInfo2), 100, 5000);
assertTrue(containerChecksumFileExists(hddsDatanode, containerInfo2.getContainerID()));
GenericTestUtils.waitFor(() -> checkContainerCloseInDatanode(hddsDatanode, containerInfo2) &&
containerChecksumFileExists(hddsDatanode, containerInfo2.getContainerID()), 100, 5000);
OzoneContainer ozoneContainer = hddsDatanode.getDatanodeStateMachine().getContainer();
Container<?> container2 = ozoneContainer.getController().getContainer(containerInfo2.getContainerID());
ContainerProtos.ContainerChecksumInfo containerChecksumInfo = ContainerMerkleTreeTestUtils.readChecksumFile(
Expand All @@ -274,12 +274,12 @@ public void testContainerChecksumForClosedContainer() throws Exception {
assertNotEquals(prevExpectedChecksumInfo1.getContainerID(), prevExpectedChecksumInfo2.getContainerID());
assertNotEquals(prevExpectedChecksumInfo1.getContainerMerkleTree().getDataChecksum(),
prevExpectedChecksumInfo2.getContainerMerkleTree().getDataChecksum());
for (ContainerReplica replica : getContainerReplicas(containerInfo1)) {
assertNotEquals(0, replica.getDataChecksum());
}
for (ContainerReplica replica : getContainerReplicas(containerInfo2)) {
assertNotEquals(0, replica.getDataChecksum());
}

// Wait for SCM to receive container reports with non-zero checksums for all replicas
GenericTestUtils.waitFor(() -> getContainerReplicas(containerInfo1).stream()
.allMatch(r -> r.getDataChecksum() != 0), 200, 5000);
GenericTestUtils.waitFor(() -> getContainerReplicas(containerInfo2).stream()
.allMatch(r -> r.getDataChecksum() != 0), 200, 5000);
}

private boolean checkContainerCloseInDatanode(HddsDatanodeService hddsDatanode,
Expand Down