hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Chinmay Kulkarni (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-17988) get-active-master.rb and draining_servers.rb no longer work
Date Mon, 05 Jun 2017 23:32:12 GMT

    [ https://issues.apache.org/jira/browse/HBASE-17988?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16037837#comment-16037837

Chinmay Kulkarni commented on HBASE-17988:

[~busbey] The {{bin/get-active-master.rb}} and {{bin/draining_servers.rb}} scripts seem to
be working fine for branches branch-1 and branch-1.3. {{bin/get-active-master.rb}} works fine
for branch 0.98 (0.98 does not have {{bin/draining_servers.rb}}).

This patch is only for the master branch, where _masterAddressZNode_  and _drainingZNode_
have been moved to be inside the _znodePaths_ member variable and is thus inaccessible directly
from the ZooKeeperWatcher object. I verified that this change works for master and that the
expected values are got by running these scripts.

> get-active-master.rb and draining_servers.rb no longer work
> -----------------------------------------------------------
>                 Key: HBASE-17988
>                 URL: https://issues.apache.org/jira/browse/HBASE-17988
>             Project: HBase
>          Issue Type: Bug
>          Components: scripts
>    Affects Versions: 2.0.0
>            Reporter: Mike Drob
>            Assignee: Chinmay Kulkarni
>            Priority: Critical
>             Fix For: 2.0.0, 1.4.0, 1.3.2, 1.1.11, 1.2.7
>         Attachments: HBASE-17988.patch
> The scripts {{bin/get-active-master.rb}} and {{bin/draining_servers.rb}} no longer work
on current master branch. Here is an example error message:
> {noformat}
> $ bin/hbase-jruby bin/get-active-master.rb 
> NoMethodError: undefined method `masterAddressZNode' for #<Java::OrgApacheHadoopHbaseZookeeper::ZooKeeperWatcher:0x1b0e9707>
>   <main> at bin/get-active-master.rb:35
> {noformat}
> My initial probing suggests that this is likely due to movement that happened in HBASE-16690.
Perhaps instead of reworking the ruby, there is similar Java functionality already existing
> Putting priority at critical since it's impossible to know whether users rely on the

This message was sent by Atlassian JIRA

View raw message