[
https://issues.apache.org/jira/browse/HIVE-25081?focusedWorklogId=609141&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-609141
]
ASF GitHub Bot logged work on HIVE-25081:
-----------------------------------------
Author: ASF GitHub Bot
Created on: 09/Jun/21 13:28
Start Date: 09/Jun/21 13:28
Worklog Time Spent: 10m
Work Description: asinkovits commented on a change in pull request #2332:
URL: https://github.com/apache/hive/pull/2332#discussion_r648285067
##########
File path: ql/src/java/org/apache/hadoop/hive/ql/txn/compactor/Cleaner.java
##########
@@ -111,7 +111,7 @@ public void run() {
// so wrap it in a big catch Throwable statement.
try {
handle =
txnHandler.getMutexAPI().acquireLock(TxnStore.MUTEX_KEY.Cleaner.name());
- if (metricsEnabled) {
+ if (metricsEnabled && MetastoreConf.getBoolVar(conf,
MetastoreConf.ConfVars.METASTORE_ACIDMETRICS_EXT_ON)) {
Review comment:
yes. nice catch, fixed.
##########
File path:
standalone-metastore/metastore-common/src/main/java/org/apache/hadoop/hive/metastore/conf/MetastoreConf.java
##########
@@ -454,6 +454,8 @@ public static ConfVars getMetaConf(String name) {
"hive.metastore.acidmetrics.check.interval", 300,
TimeUnit.SECONDS,
"Time in seconds between acid related metric collection runs."),
+ METASTORE_ACIDMETRICS_EXT_ON("metastore.acidmetrics.ext.on",
"hive.metastore.acidmetrics.ext.on", true,
+ "Whether to collect additional acid related metrics outside of the
acid metrics service."),
Review comment:
And/Or HIVE_SERVER2_METRICS_ENABLED. Shall I add both?
##########
File path: ql/src/java/org/apache/hadoop/hive/ql/txn/compactor/Cleaner.java
##########
@@ -111,7 +111,7 @@ public void run() {
// so wrap it in a big catch Throwable statement.
try {
handle =
txnHandler.getMutexAPI().acquireLock(TxnStore.MUTEX_KEY.Cleaner.name());
- if (metricsEnabled) {
+ if (metricsEnabled && MetastoreConf.getBoolVar(conf,
MetastoreConf.ConfVars.METASTORE_ACIDMETRICS_EXT_ON)) {
Review comment:
fixed.
##########
File path:
ql/src/java/org/apache/hadoop/hive/ql/txn/compactor/metrics/DeltaFilesMetricReporter.java
##########
@@ -115,41 +117,45 @@ public static DeltaFilesMetricReporter getInstance() {
return InstanceHolder.instance;
}
- public static synchronized void init(HiveConf conf){
+ public static synchronized void init(HiveConf conf) {
getInstance().configure(conf);
}
public void submit(TezCounters counters) {
- updateMetrics(NUM_OBSOLETE_DELTAS,
- obsoleteDeltaCache, obsoleteDeltaTopN, obsoleteDeltasThreshold,
counters);
- updateMetrics(NUM_DELTAS,
- deltaCache, deltaTopN, deltasThreshold, counters);
- updateMetrics(NUM_SMALL_DELTAS,
- smallDeltaCache, smallDeltaTopN, deltasThreshold, counters);
+ if(acidMetricsExtEnabled) {
Review comment:
yeah.. So first issue is that this metrics is collected in HS2 but the
conf (METASTORE_ACIDMETRICS_EXT_ON) for the metrics collection is defined in
the MetastoreConf, so I wanted to minimize the exposure of that.
Second is in the end I put it here, because it seamed to make the class more
resilient. Here is my reasoning: This is a singleton class, so you can access
it basically anywhere, but you need to call the init method on it so that it
works properly. The acidMetricsExtEnabled flag is indeed set in the init
(configure), so if it was not called it will skip this code part which
otherwise would throw a NPE.
But I'm open to a better solution :)
##########
File path:
ql/src/java/org/apache/hadoop/hive/ql/txn/compactor/metrics/DeltaFilesMetricReporter.java
##########
@@ -115,41 +117,45 @@ public static DeltaFilesMetricReporter getInstance() {
return InstanceHolder.instance;
}
- public static synchronized void init(HiveConf conf){
+ public static synchronized void init(HiveConf conf) {
getInstance().configure(conf);
}
public void submit(TezCounters counters) {
- updateMetrics(NUM_OBSOLETE_DELTAS,
- obsoleteDeltaCache, obsoleteDeltaTopN, obsoleteDeltasThreshold,
counters);
- updateMetrics(NUM_DELTAS,
- deltaCache, deltaTopN, deltasThreshold, counters);
- updateMetrics(NUM_SMALL_DELTAS,
- smallDeltaCache, smallDeltaTopN, deltasThreshold, counters);
+ if(acidMetricsExtEnabled) {
+ updateMetrics(NUM_OBSOLETE_DELTAS,
+ obsoleteDeltaCache, obsoleteDeltaTopN, obsoleteDeltasThreshold,
counters);
+ updateMetrics(NUM_DELTAS,
+ deltaCache, deltaTopN, deltasThreshold, counters);
+ updateMetrics(NUM_SMALL_DELTAS,
+ smallDeltaCache, smallDeltaTopN, deltasThreshold, counters);
+ }
}
- public static void mergeDeltaFilesStats(AcidDirectory dir, long
checkThresholdInSec,
- float deltaPctThreshold, EnumMap<DeltaFilesMetricType, Map<String,
Integer>> deltaFilesStats) throws IOException {
- long baseSize = getBaseSize(dir);
- int numObsoleteDeltas = getNumObsoleteDeltas(dir, checkThresholdInSec);
+ public static void mergeDeltaFilesStats(AcidDirectory dir, long
checkThresholdInSec, float deltaPctThreshold,
+ EnumMap<DeltaFilesMetricType, Map<String, Integer>> deltaFilesStats,
Configuration conf) throws IOException {
+ if (MetastoreConf.getBoolVar(conf,
MetastoreConf.ConfVars.METASTORE_ACIDMETRICS_EXT_ON)) {
Review comment:
This is to minimize the exposure of the MetastoreConf. I've tried to put
as many checks into this class as I could.
##########
File path:
ql/src/java/org/apache/hadoop/hive/ql/txn/compactor/metrics/DeltaFilesMetricReporter.java
##########
@@ -206,7 +216,7 @@ public static void backPropagateAcidMetrics(JobConf
jobConf, Configuration conf)
}
public static void close() {
- if (getInstance() != null) {
+ if (getInstance() != null && getInstance().acidMetricsExtEnabled) {
Review comment:
the executorService is created in the configure method. So if it was not
called this will throw a NPE if I understand.
##########
File path: ql/src/java/org/apache/hadoop/hive/ql/txn/compactor/Initiator.java
##########
@@ -120,7 +120,7 @@ public void run() {
// don't doom the entire thread.
try {
handle =
txnHandler.getMutexAPI().acquireLock(TxnStore.MUTEX_KEY.Initiator.name());
- if (metricsEnabled) {
+ if (metricsEnabled && MetastoreConf.getBoolVar(conf,
MetastoreConf.ConfVars.METASTORE_ACIDMETRICS_EXT_ON)) {
Review comment:
fixed.
##########
File path:
ql/src/java/org/apache/hadoop/hive/ql/txn/compactor/metrics/DeltaFilesMetricReporter.java
##########
@@ -190,13 +197,16 @@ public static void
createCountersForAcidMetrics(TezCounters tezCounters, JobConf
}
public static void addAcidMetricsToConfObj(EnumMap<DeltaFilesMetricType,
Map<String, Integer>> deltaFilesStats, Configuration conf) {
- deltaFilesStats.forEach((type, value) ->
- conf.set(type.name(),
Joiner.on(",").withKeyValueSeparator("->").join(value))
- );
+ if (MetastoreConf.getBoolVar(conf,
MetastoreConf.ConfVars.METASTORE_ACIDMETRICS_EXT_ON)) {
Review comment:
nice, fixed.
##########
File path:
ql/src/java/org/apache/hadoop/hive/ql/txn/compactor/metrics/DeltaFilesMetricReporter.java
##########
@@ -230,23 +240,26 @@ private static long getDirSize(AcidUtils.ParsedDirectory
dir, FileSystem fs) thr
.sum();
}
- private void configure(HiveConf conf){
- deltasThreshold = HiveConf.getIntVar(conf,
HiveConf.ConfVars.HIVE_TXN_ACID_METRICS_DELTA_NUM_THRESHOLD);
- obsoleteDeltasThreshold = HiveConf.getIntVar(conf,
HiveConf.ConfVars.HIVE_TXN_ACID_METRICS_OBSOLETE_DELTA_NUM_THRESHOLD);
-
- initMetricsCache(conf);
- long reportingInterval = HiveConf.getTimeVar(conf,
- HiveConf.ConfVars.HIVE_TXN_ACID_METRICS_REPORTING_INTERVAL,
TimeUnit.SECONDS);
-
- ThreadFactory threadFactory =
- new ThreadFactoryBuilder()
- .setDaemon(true)
- .setNameFormat("DeltaFilesMetricReporter %d")
- .build();
- executorService =
Executors.newSingleThreadScheduledExecutor(threadFactory);
- executorService.scheduleAtFixedRate(
- new ReportingTask(), 0, reportingInterval, TimeUnit.SECONDS);
- LOG.info("Started DeltaFilesMetricReporter thread");
+ private void configure(HiveConf conf) {
+ acidMetricsExtEnabled = MetastoreConf.getBoolVar(conf,
MetastoreConf.ConfVars.METASTORE_ACIDMETRICS_EXT_ON);
+ if (acidMetricsExtEnabled) {
Review comment:
Again exposure of the MetastoreConf. It might not be a valid reason
though... :D
And because of my comment in the close method ("the executorService is
created in the configure method.")
we need to track the acidMetricsExtEnabled in this class.
--
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.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
Issue Time Tracking
-------------------
Worklog Id: (was: 609141)
Time Spent: 0.5h (was: 20m)
> Put metrics collection behind a feature flag
> --------------------------------------------
>
> Key: HIVE-25081
> URL: https://issues.apache.org/jira/browse/HIVE-25081
> Project: Hive
> Issue Type: Sub-task
> Reporter: Antal Sinkovits
> Assignee: Antal Sinkovits
> Priority: Major
> Labels: pull-request-available
> Time Spent: 0.5h
> Remaining Estimate: 0h
>
> Most metrics we're creating are collected in AcidMetricsService, which is
> behind a feature flag. However there are some metrics that are collected
> outside of the service. These should be behind a feature flag in addition to
> hive.metastore.metrics.enabled.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
