helix-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "kishore gopalakrishna (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (HELIX-31) Detect flapping and disable the participant/controller/spectator
Date Tue, 29 Jan 2013 23:03:13 GMT

     [ https://issues.apache.org/jira/browse/HELIX-31?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel

kishore gopalakrishna updated HELIX-31:

    Fix Version/s: 0.6.1-incubating
> Detect flapping and disable the participant/controller/spectator 
> -----------------------------------------------------------------
>                 Key: HELIX-31
>                 URL: https://issues.apache.org/jira/browse/HELIX-31
>             Project: Apache Helix
>          Issue Type: Improvement
>            Reporter: kishore gopalakrishna
>             Fix For: 0.6.1-incubating
> There are many cases where the zk client gets into infinite loop of connect/disconnect.
Few use cases where this may occur
> * GC
> * Bugs( for example when a node tries to write something more than 1 MB in a znode)
> * Network glitches 
> Right thing to do in this case, it to keep track of connect/disconnect that happen over
a period of time and either let the controller know about it and disable yourself.

This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

View raw message