[
https://issues.apache.org/jira/browse/HBASE-25634?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhengsicheng updated HBASE-25634:
---------------------------------
Description:
When the client scan operation, the server frequently returns
RpcThrottlingException, which will cause the meta table request to become high.
/hbase-client/src/main/java/org/apache/hadoop/hbase/client/RpcRetryingCallerImpl.java
{code:java}
// code placeholder
@Override
public T callWithRetries(RetryingCallable<T> callable, int callTimeout)
throws IOException, RuntimeException {
List<RetriesExhaustedException.ThrowableWithExtraContext> exceptions = new
ArrayList<>();
tracker.start();
context.clear();
for (int tries = 0;; tries++) {
long expectedSleep;
try {
// bad cache entries are cleared in the call to
RetryingCallable#throwable() in catch block
// callable.prepare() reload force reload of server location
callable.prepare(tries != 0);
interceptor.intercept(context.prepare(callable, tries));
return callable.call(getTimeout(callTimeout));
} catch (PreemptiveFastFailException e) {
throw e;
} catch (Throwable t) {
ExceptionUtil.rethrowIfInterrupt(t);
Throwable cause = t.getCause();
if (cause instanceof DoNotRetryIOException) {
// Fail fast
throw (DoNotRetryIOException) cause;
}
// translateException throws exception when should not retry: i.e. when
request is bad.
interceptor.handleFailure(context, t);
t = translateException(t);
{code}
!image-2021-03-05-12-00-33-522.png!
!image-2021-03-05-12-01-08-769.png!
was:
When the client scan operation, the server frequently returns
RpcThrottlingException, which will cause the meta table request to become high.
/hbase-client/src/main/java/org/apache/hadoop/hbase/client/RpcRetryingCallerImpl.java
{code:java}
// code placeholder
@Override
public T callWithRetries(RetryingCallable<T> callable, int callTimeout)
throws IOException, RuntimeException {
List<RetriesExhaustedException.ThrowableWithExtraContext> exceptions = new
ArrayList<>();
tracker.start();
context.clear();
for (int tries = 0;; tries++) {
long expectedSleep;
try {
// bad cache entries are cleared in the call to
RetryingCallable#throwable() in catch block
Throwable t = null;
if (exceptions != null && !exceptions.isEmpty()) {
t = exceptions.get(exceptions.size() - 1).throwable;
}
if (!(t instanceof RpcThrottlingException)) {
callable.prepare(tries != 0);
}
interceptor.intercept(context.prepare(callable, tries));
return callable.call(getTimeout(callTimeout));
} catch (PreemptiveFastFailException e) {
throw e;
} catch (Throwable t) {
ExceptionUtil.rethrowIfInterrupt(t);
Throwable cause = t.getCause();
if (cause instanceof DoNotRetryIOException) {
// Fail fast
throw (DoNotRetryIOException) cause;
}
// translateException throws exception when should not retry: i.e. when
request is bad.
interceptor.handleFailure(context, t);
t = translateException(t);
{code}
!image-2021-03-05-12-00-33-522.png!
!image-2021-03-05-12-01-08-769.png!
> The client frequently exceeds the quota, which causes the meta table scan to
> be too high
> ----------------------------------------------------------------------------------------
>
> Key: HBASE-25634
> URL: https://issues.apache.org/jira/browse/HBASE-25634
> Project: HBase
> Issue Type: Bug
> Affects Versions: 2.3.4
> Reporter: zhengsicheng
> Assignee: zhengsicheng
> Priority: Minor
> Attachments: image-2021-03-05-12-00-33-522.png,
> image-2021-03-05-12-01-08-769.png
>
>
> When the client scan operation, the server frequently returns
> RpcThrottlingException, which will cause the meta table request to become
> high.
>
> /hbase-client/src/main/java/org/apache/hadoop/hbase/client/RpcRetryingCallerImpl.java
> {code:java}
> // code placeholder
> @Override
> public T callWithRetries(RetryingCallable<T> callable, int callTimeout)
> throws IOException, RuntimeException {
> List<RetriesExhaustedException.ThrowableWithExtraContext> exceptions = new
> ArrayList<>();
> tracker.start();
> context.clear();
> for (int tries = 0;; tries++) {
> long expectedSleep;
> try {
> // bad cache entries are cleared in the call to
> RetryingCallable#throwable() in catch block
> // callable.prepare() reload force reload of server location
> callable.prepare(tries != 0);
> interceptor.intercept(context.prepare(callable, tries));
> return callable.call(getTimeout(callTimeout));
> } catch (PreemptiveFastFailException e) {
> throw e;
> } catch (Throwable t) {
> ExceptionUtil.rethrowIfInterrupt(t);
> Throwable cause = t.getCause();
> if (cause instanceof DoNotRetryIOException) {
> // Fail fast
> throw (DoNotRetryIOException) cause;
> }
> // translateException throws exception when should not retry: i.e. when
> request is bad.
> interceptor.handleFailure(context, t);
> t = translateException(t);
> {code}
>
>
>
> !image-2021-03-05-12-00-33-522.png!
> !image-2021-03-05-12-01-08-769.png!
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
