You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@geode.apache.org by "ASF subversion and git services (JIRA)" <ji...@apache.org> on 2017/03/10 21:33:04 UTC

[jira] [Commented] (GEODE-2642) Sometime client doesn't ping server in ping-interval

    [ https://issues.apache.org/jira/browse/GEODE-2642?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15905729#comment-15905729 ] 

ASF subversion and git services commented on GEODE-2642:
--------------------------------------------------------

Commit 27995d6cc2f02ba7edbbc094117ffeaaf61d00e2 in geode's branch refs/heads/develop from [~hitesh.khamesra]
[ https://git-wip-us.apache.org/repos/asf?p=geode.git;h=27995d6 ]

GEODE-2642 Fixed ping thread issue on client.

Client ping thread wakes up every "ping-interval" and then it checks
whether someone already did some operation on that server or not, in the
last ping-interval. If there is some operation, then it doesn't ping
server and wakes up again after ping-interval.Otherwise ot pings the
server. But if operation happens just start of the previous ping-interval
then client thread will ping server after more than ping-interval.
This can cause an issue. Thus to avoid this race condition, now ping
thread will wake up in ping-interval/2 and will make sure it pings server.


> Sometime client doesn't ping server in ping-interval
> ----------------------------------------------------
>
>                 Key: GEODE-2642
>                 URL: https://issues.apache.org/jira/browse/GEODE-2642
>             Project: Geode
>          Issue Type: Bug
>          Components: client/server
>            Reporter: Hitesh Khamesra
>            Assignee: Hitesh Khamesra
>
> Client ping thread wakes up every "ping-interval" and then it checks whether someone already did some operation on that server or not, in the last ping-interval. If there is some operation, then it doesn't ping server and wakes up again after ping-interval. But if operation happens just start of the previous ping-interval then client thread will ping server after more then ping-interval. This can cause an issue.  Thus to avoid this race condition, now ping thread will wake up in ping-interval/2 and will make sure it pings server.



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)