zymap commented on a change in pull request #13833:
URL: https://github.com/apache/pulsar/pull/13833#discussion_r803220813
##########
File path:
pulsar-broker/src/main/java/org/apache/pulsar/broker/stats/prometheus/PrometheusMetricsGenerator.java
##########
@@ -160,6 +171,52 @@ private static void
generateBrokerBasicMetrics(PulsarService pulsar, SimpleTextO
clusterName, Collector.Type.GAUGE, stream);
}
+
+ private static void generateLedgerOffloaderMetrics(PulsarService pulsar,
SimpleTextOutputStream stream,
+ boolean
includeTopicMetrics) {
+ List<Metrics> metricList = new LinkedList<>();
+ if (includeTopicMetrics) {
+
pulsar.getBrokerService().getMultiLayerTopicMap().forEach((namespace,
bundlesMap) -> {
+ bundlesMap.forEach((bundle, topicsMap) -> {
+ topicsMap.forEach((topicName, topic) -> {
+ if (topic instanceof PersistentTopic) {
+ //check offlaoder
+ PersistentTopic persistentTopic =
(PersistentTopic) topic;
+ LedgerOffloader offloader =
+
persistentTopic.getManagedLedger().getConfig().getLedgerOffloader();
+ if (!(offloader instanceof NullLedgerOffloader)) {
+ try {
+ List<Metrics> metrics =
+ new LedgerOffloaderMetrics(pulsar,
true, namespace, topicName).generate();
+ metricList.addAll(metrics);
+ } catch (Exception ex) {
+ log.error("generate ledger offloader topic
level metrics error", ex);
Review comment:
Add topic name in the error log?
##########
File path:
tiered-storage/file-system/src/main/java/org/apache/bookkeeper/mledger/offload/filesystem/impl/FileStoreBackedReadHandleImpl.java
##########
@@ -113,7 +122,10 @@ public LedgerMetadata getLedgerMetadata() {
key.set(nextExpectedId - 1);
reader.seek(key);
while (entriesToRead > 0) {
+ long startReadTime = System.nanoTime();
reader.next(key, value);
+ mxBean.recordReadOffloadDataLatency(managedLedgerName,
System.nanoTime() - startReadTime,
Review comment:
Maybe we need to add try-catch for the `reader.next`? It may throw
IOException which would miss the latency calculation?
##########
File path:
tiered-storage/file-system/src/main/java/org/apache/bookkeeper/mledger/offload/filesystem/impl/FileSystemManagedLedgerOffloader.java
##########
@@ -208,7 +224,10 @@ public void run() {
long end = Math.min(needToOffloadFirstEntryNumber +
ENTRIES_PER_READ - 1,
readHandle.getLastAddConfirmed());
log.debug("read ledger entries. start: {}, end: {}",
needToOffloadFirstEntryNumber, end);
+ long startReadTime = System.nanoTime();
LedgerEntries ledgerEntriesOnce =
readHandle.readAsync(needToOffloadFirstEntryNumber, end).get();
+ long cost = System.nanoTime() - startReadTime;
Review comment:
`try-catch` here?
##########
File path:
tiered-storage/file-system/src/test/java/org/apache/bookkeeper/mledger/offload/filesystem/impl/FileSystemManagedLedgerOffloaderTest.java
##########
@@ -111,6 +112,31 @@ public void testOffloadAndRead() throws Exception {
}
}
+ @Test
+ public void testOffloadAndReadMetrics() throws Exception {
+ LedgerOffloader offloader = fileSystemManagedLedgerOffloader;
+ UUID uuid = UUID.randomUUID();
+ offloader.offload(toWrite, uuid, map).get();
+
+ LedgerOffloaderMXBeanImpl mbean =
(LedgerOffloaderMXBeanImpl)offloader.getStats();
+ assertTrue(mbean.getOffloadErrors(topic) == 0);
+ assertTrue(mbean.getOffloadBytes(topic) > 0 );
+ assertTrue(mbean.getReadLedgerLatencyBuckets(topic).getCount() > 0);
+ assertTrue(mbean.getWriteToStorageErrors(topic) == 0);
+
+ ReadHandle toTest = offloader.readOffloaded(toWrite.getId(), uuid,
map).get();
+ LedgerEntries toTestEntries = toTest.read(0, numberOfEntries - 1);
+ Iterator<LedgerEntry> toTestIter = toTestEntries.iterator();
+ while(toTestIter.hasNext()) {
+ LedgerEntry toTestEntry = toTestIter.next();
Review comment:
Can we compare the read entries size is equals the
`mbean.getReadOffloadBytes(topic)`?
##########
File path:
managed-ledger/src/main/java/org/apache/bookkeeper/mledger/LedgerOffloaderMXBean.java
##########
@@ -0,0 +1,95 @@
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing,
+ * software distributed under the License is distributed on an
+ * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+ * KIND, either express or implied. See the License for the
+ * specific language governing permissions and limitations
+ * under the License.
+ */
+package org.apache.bookkeeper.mledger;
+
+import org.apache.bookkeeper.common.annotation.InterfaceAudience;
+import org.apache.bookkeeper.common.annotation.InterfaceStability;
+import org.apache.bookkeeper.mledger.util.StatsBuckets;
+
+
+/**
+ * Management Bean for a {@link LedgerOffloader}.
+ */
[email protected]
[email protected]
+public interface LedgerOffloaderMXBean {
+
+ /**
+ * The ledger offloader name.
+ *
+ * @return ledger offloader name.
+ */
+ String getDriverName();
+
+ /**
+ * Record the offload error count.
+ *
+ * @return offload errors per topic.
+ */
+ long getOffloadErrors(String topic);
+
+ /**
+ * Record the offload rate to storage.
+ *
+ * @return offload rate per topic.
+ */
+ long getOffloadBytes(String topic);
+
+ /**
+ * Record the read ledger latency.
+ *
+ * @return read ledger latency per topic.
+ */
+ StatsBuckets getReadLedgerLatencyBuckets(String topic);
+
+ /**
+ * Record the write to storage error count.
+ *
+ * @return write to storage errors per topic.
+ */
+ long getWriteToStorageErrors(String topic);
+
+ /**
+ * Record read offload index latency.
+ *
+ * @return read offload index latency per topic.
+ */
+ StatsBuckets getReadOffloadIndexLatencyBuckets(String topic);
+
+ /**
+ * Record read offload data latency.
+ *
+ * @return read offload data latency per topic.
+ */
+ StatsBuckets getReadOffloadDataLatencyBuckets(String topic);
+
+ /**
+ * Record read offload method rate.
+ *
+ * @return read offload data rate.
+ */
+ long getReadOffloadBytes(String topic);
Review comment:
Same with above. I saw we only record the offload bytes, right? Not the
offload rate?
##########
File path:
managed-ledger/src/main/java/org/apache/bookkeeper/mledger/LedgerOffloaderMXBean.java
##########
@@ -0,0 +1,123 @@
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing,
+ * software distributed under the License is distributed on an
+ * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+ * KIND, either express or implied. See the License for the
+ * specific language governing permissions and limitations
+ * under the License.
+ */
+package org.apache.bookkeeper.mledger;
+
+import org.apache.bookkeeper.common.annotation.InterfaceAudience;
+import org.apache.bookkeeper.common.annotation.InterfaceStability;
+import org.apache.bookkeeper.mledger.util.StatsBuckets;
+
+
+/**
+ * Management Bean for a {@link LedgerOffloader}.
+ */
[email protected]
[email protected]
+public interface LedgerOffloaderMXBean {
+
+ /**
+ * The ledger offloader name.
+ *
+ * @return ledger offloader name.
+ */
+ String getDriverName();
+
+ /**
+ * Record the offload total time.
+ *
+ * @return offload time per topic.
+ */
+ long getOffloadTime(String topic);
+
+ /**
+ * Record the offload error count.
+ *
+ * @return offload errors per topic.
+ */
+ long getOffloadErrors(String topic);
+
+ /**
+ * Record the offload rate to storage.
+ *
+ * @return offload rate per topic.
+ */
+ long getOffloadBytes(String topic);
Review comment:
Do we record the rate or the data bytes? It is confusing.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
