hadoop-common-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Erik Krogen (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (HADOOP-16192) CallQueue backoff bug fixes: doesn't perform backoff when add() is used, and doesn't update backoff when refreshed
Date Thu, 14 Mar 2019 17:12:00 GMT

     [ https://issues.apache.org/jira/browse/HADOOP-16192?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Erik Krogen updated HADOOP-16192:
---------------------------------
    Description: 
The {{CallQueueManager}} has a mechanism to enforce backoff when various criteria are met,
as defined by the implementation of {{RpcScheduler}} used. However, it currently only checks
for these backoff criteria when {{put()}} is used; {{add()}} calls are passed directly to
the underlying call queue. It should check the backoff criteria for either call.

Also, when {{refreshCallQueue()}} is called, the backoff configuration is not refreshed. This
should be updated as well.

  was:The {{CallQueueManager}} has a mechanism to enforce backoff when various criteria are
met, as defined by the implementation of {{RpcScheduler}} used. However, it currently only
checks for these backoff criteria when {{put()}} is used; {{add()}} calls are passed directly
to the underlying call queue. It should check the backoff criteria for either call.


> CallQueue backoff bug fixes: doesn't perform backoff when add() is used, and doesn't
update backoff when refreshed
> ------------------------------------------------------------------------------------------------------------------
>
>                 Key: HADOOP-16192
>                 URL: https://issues.apache.org/jira/browse/HADOOP-16192
>             Project: Hadoop Common
>          Issue Type: Bug
>          Components: ipc
>            Reporter: Erik Krogen
>            Assignee: Erik Krogen
>            Priority: Major
>
> The {{CallQueueManager}} has a mechanism to enforce backoff when various criteria are
met, as defined by the implementation of {{RpcScheduler}} used. However, it currently only
checks for these backoff criteria when {{put()}} is used; {{add()}} calls are passed directly
to the underlying call queue. It should check the backoff criteria for either call.
> Also, when {{refreshCallQueue()}} is called, the backoff configuration is not refreshed.
This should be updated as well.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: common-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: common-issues-help@hadoop.apache.org


Mime
View raw message