[
https://issues.apache.org/jira/browse/HADOOP-16700?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16971989#comment-16971989
]
xuzq edited comment on HADOOP-16700 at 11/12/19 2:15 AM:
---------------------------------------------------------
Thanks [~xkrogen].
newCall(_timestamp1_) -> enQueue -> inQueue ->(_timestamp2_) handle ->
process -> response(_timestamp3_) -> updateMetric.
Like above, QueueTime is computed in _updateMetric_, and QueueTime =
_timestamp2 - call.timestamp(timestamp1) - enQueueTime._
But *_call.timestamp_ maybe changed in _sendResponse_*, so QueueTime =
timestamp2 - call.timestamp(timestamp3) - enQueueTime, it will be negative.
{code:java}
// processResponse
//
// If we were unable to write the entire response out, then
// insert in Selector queue.
//
call.connection.responseQueue.addFirst(call);
if (inHandler) {
// set the serve time when the response has to be sent later
call.timestampNanos = Time.monotonicNowNanos();
incPending();
try {
// Wakeup the thread blocked on select, only then can the call
// to channel.register() complete.
writeSelector.wakeup();
channel.register(writeSelector, SelectionKey.OP_WRITE, call);
} catch (ClosedChannelException e) {
//Its ok. channel might be closed else where.
done = true;
} finally {
decPending();
}
}
{code}
was (Author: xuzq_zander):
Thanks [~xkrogen].
newCall(_timestamp1_) -> enQueue -> inQueue ->(_timestamp2_) handle ->
process -> response(_timestamp3_) -> updateMetric.
Like above, QueueTime is computed in _updateMetric_, and QueueTime =
_timestamp2 - call.timestamp(timestamp1) - enQueueTime._
But *_call.timestamp_ maybe changed in _sendResponse_*, so QueueTime =
timestamp2 - call.timestamp(timestamp3) - enQueueTime, it will be negative.
{code:java}
// processResponse
//
// If we were unable to write the entire response out, then
// insert in Selector queue.
//
call.connection.responseQueue.addFirst(call);
if (inHandler) {
// set the serve time when the response has to be sent later
call.timestampNanos = Time.monotonicNowNanos();
incPending();
try {
// Wakeup the thread blocked on select, only then can the call
// to channel.register() complete.
writeSelector.wakeup();
channel.register(writeSelector, SelectionKey.OP_WRITE, call);
} catch (ClosedChannelException e) {
//Its ok. channel might be closed else where.
done = true;
} finally {
decPending();
}
}
{code}
> RpcQueueTime may be negative when the response has to be sent later
> -------------------------------------------------------------------
>
> Key: HADOOP-16700
> URL: https://issues.apache.org/jira/browse/HADOOP-16700
> Project: Hadoop Common
> Issue Type: Bug
> Reporter: xuzq
> Priority: Minor
> Attachments: HADOOP-16700-trunk-001.patch
>
>
> RpcQueueTime may be negative when the response has to be sent later.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
