hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "stack (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-19341) Ensure CP can abort a Server
Date Wed, 06 Dec 2017 21:21:00 GMT

    [ https://issues.apache.org/jira/browse/HBASE-19341?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16280945#comment-16280945
] 

stack commented on HBASE-19341:
-------------------------------

Thanks for the reminder [~anoop.hbase]. Yes, it is as you describe it.  

<property>
      <name>hbase.coprocessor.abortonerror</name>
      <value>true</value>
      <description>Set to true to cause the hosting server (master or regionserver)
      to abort if a coprocessor fails to load, fails to initialize, or throws an
      unexpected Throwable object. Setting this to false will allow the server to
      continue execution but the system wide state of the coprocessor in question
      will become inconsistent as it will be properly executing in only a subset
      of servers, so this is most useful for debugging only.</description>
  </property>

...is the config.

HBASE-6873 set it to be true by default so we abort on CP exception.

It is hard to follow (two separate comments attest to the messy, distributed commit), but
it seems like the tests we want are already in place added by HBASE-4014 back in 0.92. They
look good.

Let me resolve this as not needed. Will add release note w/ the Anoop reminder and findings
here in case someone goes looking.

Hope this good by you [~ashish singhi]. Just have the CP throw and Error and default, we'll
abort the hosting serverr.



> Ensure CP can abort a Server
> ----------------------------
>
>                 Key: HBASE-19341
>                 URL: https://issues.apache.org/jira/browse/HBASE-19341
>             Project: HBase
>          Issue Type: Improvement
>          Components: Coprocessors
>            Reporter: stack
>            Assignee: stack
>            Priority: Critical
>             Fix For: 2.0.0-beta-1
>
>
> We used to allow a CP pull the Server#abort chain. We removed it in the CP refactor.
At the end of HBASE-18298, [~ankit@apache.org] describes a case where Phoenix needs to kill
Server in extreme case to maintain consistency. This issue is about ensuring that throwing
a CPException will indeed kill the running server.... Add a test.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

Mime
View raw message