This is an automated email from the ASF dual-hosted git repository.
jojochuang pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/ozone.git
The following commit(s) were added to refs/heads/master by this push:
new 29758e4ac1d HDDS-16441. Orphaned snapshot diff task marks a
resubmitted job DONE with another run's metadata (#11321)
29758e4ac1d is described below
commit 29758e4ac1d433f6aa7016bfddc67cb13f9bb081
Author: Siyao Meng <[email protected]>
AuthorDate: Thu Oct 1 12:35:46 2026 -0700
HDDS-16441. Orphaned snapshot diff task marks a resubmitted job DONE with
another run's metadata (#11321)
Generated-by: Claude Code (Opus 5)
---
.../hadoop/ozone/om/snapshot/SnapshotDiffManager.java | 15 +++++++++++----
.../ozone/om/snapshot/TestSnapshotDiffManager.java | 18 ++++++++++++++++++
2 files changed, 29 insertions(+), 4 deletions(-)
diff --git
a/hadoop-ozone/ozone-manager/src/main/java/org/apache/hadoop/ozone/om/snapshot/SnapshotDiffManager.java
b/hadoop-ozone/ozone-manager/src/main/java/org/apache/hadoop/ozone/om/snapshot/SnapshotDiffManager.java
index 1452999bb5c..d37fcab50c3 100644
---
a/hadoop-ozone/ozone-manager/src/main/java/org/apache/hadoop/ozone/om/snapshot/SnapshotDiffManager.java
+++
b/hadoop-ozone/ozone-manager/src/main/java/org/apache/hadoop/ozone/om/snapshot/SnapshotDiffManager.java
@@ -1185,7 +1185,7 @@ void generateSnapshotDiffReport(final String jobKey,
LOG.debug("Generated snapshot diff report, entry count: {},
elapsed: {}ms, jobId: {}",
reportEntries.getKey(), Time.monotonicNow() -
reportGenStart, jobId);
}
- updateJobStatusToDone(jobKey, reportEntries.getKey(),
reportEntries.getValue());
+ updateJobStatusToDone(jobKey, jobId, reportEntries.getKey(),
reportEntries.getValue());
}
return null;
}
@@ -1744,10 +1744,17 @@ private synchronized void
updateJobStatusToFailed(String jobKey,
snapDiffJobTable.put(jobKey, snapshotDiffJob);
}
- private synchronized void updateJobStatusToDone(String jobKey,
- long totalDiffEntries,
- String largestJobKey) {
+ synchronized void updateJobStatusToDone(String jobKey,
+ String jobId,
+ long totalDiffEntries,
+ String largestJobKey) {
SnapshotDiffJob snapshotDiffJob = snapDiffJobTable.get(jobKey);
+ // A cancelled job's task keeps running, and a resubmit reuses the jobKey
with a new jobId.
+ if (!snapshotDiffJob.getJobId().equals(jobId)) {
+ LOG.warn("Not marking snapshot diff job {} as DONE: task jobId {} was
superseded by jobId {}.",
+ jobKey, jobId, snapshotDiffJob.getJobId());
+ return;
+ }
if (snapshotDiffJob.getStatus() != IN_PROGRESS) {
throw new IllegalStateException("Invalid job status for jobID: " +
snapshotDiffJob.getJobId() + ". Job's current status is '" +
diff --git
a/hadoop-ozone/ozone-manager/src/test/java/org/apache/hadoop/ozone/om/snapshot/TestSnapshotDiffManager.java
b/hadoop-ozone/ozone-manager/src/test/java/org/apache/hadoop/ozone/om/snapshot/TestSnapshotDiffManager.java
index f391125fe06..2bc474d5be1 100644
---
a/hadoop-ozone/ozone-manager/src/test/java/org/apache/hadoop/ozone/om/snapshot/TestSnapshotDiffManager.java
+++
b/hadoop-ozone/ozone-manager/src/test/java/org/apache/hadoop/ozone/om/snapshot/TestSnapshotDiffManager.java
@@ -1564,6 +1564,24 @@ public void
testSubmitSnapshotDiffResubmitsWhenPreviouslyCancelled()
assertEquals(IN_PROGRESS, updated.getStatus());
}
+ @Test
+ public void testUpdateJobStatusToDoneSkipsSupersededJob() throws IOException
{
+ SnapDiffTestContext ctx = setupRandomSnapDiffTestContext();
+ String liveJobId = UUID.randomUUID().toString();
+ snapshotDiffManager.getSnapDiffJobTable().put(ctx.diffJobKey,
+ new SnapshotDiffJob(0L, liveJobId, IN_PROGRESS, ctx.volumeName,
ctx.bucketName,
+ ctx.fromSnapshotName, ctx.toSnapshotName, false, false, 0L, null,
0.0, null));
+
+ // A superseded task's publish is ignored, the live job's own publish goes
through.
+ snapshotDiffManager.updateJobStatusToDone(ctx.diffJobKey,
UUID.randomUUID().toString(), 5L, "staleKey");
+ assertEquals(IN_PROGRESS,
snapshotDiffManager.getSnapDiffJobTable().get(ctx.diffJobKey).getStatus());
+
+ snapshotDiffManager.updateJobStatusToDone(ctx.diffJobKey, liveJobId, 5L,
"liveKey");
+ SnapshotDiffJob done =
snapshotDiffManager.getSnapDiffJobTable().get(ctx.diffJobKey);
+ assertEquals(DONE, done.getStatus());
+ assertEquals("liveKey", done.getLargestEntryKey());
+ }
+
@Test
public void testSubmitSnapshotDiffResubmitsWhenPreviouslyRejected()
throws IOException {
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
