[
https://issues.apache.org/jira/browse/YARN-6640?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16138559#comment-16138559
]
Jason Lowe commented on YARN-6640:
----------------------------------
Yes, sorry I didn't call it out explicitly. I agree that we should only expect
a request to have the same ID we sent in the last response or the previous ID.
Anything else should be an error since the AM is out of sync with the RM. A
sane AM could send a request ID that is far larger than the RM's current ID
after the RM restarts, but I think that case should already be covered by the
!hasApplicationMasterRegistered check before we compare the request ID to the
last response ID.
> AM heartbeat stuck when responseId overflows MAX_INT
> -----------------------------------------------------
>
> Key: YARN-6640
> URL: https://issues.apache.org/jira/browse/YARN-6640
> Project: Hadoop YARN
> Issue Type: Bug
> Reporter: Botong Huang
> Assignee: Botong Huang
> Priority: Blocker
> Attachments: YARN-6640.v1.patch
>
>
> The current code in {{ApplicationMasterService}}:
> if ((request.getResponseId() + 1) == lastResponse.getResponseId()) {/* old
> heartbeat */ return lastResponse;}
> else if (request.getResponseId() + 1 < lastResponse.getResponseId()) { throw
> ... }
> process the heartbeat...
> When a heartbeat comes in, in usual case we are expecting
> request.getResponseId() == lastResponse.getResponseId(). The “if“ is for the
> duplicate heartbeat that’s one step old, the “else if” is to throw and
> complain for heartbeats more than two steps old, otherwise we accept the new
> heartbeat and process it.
> So the bug is: when lastResponse.getResponseId() == MAX_INT, the newest
> heartbeat comes in with responseId == MAX_INT. However reponseId + 1 will be
> MIN_INT, and we will fall into the “else if” case and RM will throw. Then we
> are stuck here…
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
