[ 
https://issues.apache.org/jira/browse/GEODE-1128?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15511327#comment-15511327
 ] 

ASF subversion and git services commented on GEODE-1128:
--------------------------------------------------------

Commit 09cd4ad27298ebf20441e3285636f8e4c7e772c7 in incubator-geode's branch 
refs/heads/develop from [~khowe]
[ https://git-wip-us.apache.org/repos/asf?p=incubator-geode.git;h=09cd4ad ]

GEODE-1128: Add missing regions to the missing-disk-stores command

Refactored “show missing-disk-stores command” and added a section to the
output to show missing colocated regions. Corrected messages in
LocalizedStrings, change “gemfire list-missing-disk-stores” to
“gfsh show missing-disk-stores”. Added new command function class to gather
both missing disk-store and missing colocated region information on a
distributed system member.

Parameterized show missing-disk-store error message.

Added new DataSerializable class for sending missing colocated region
results to gfsh command.

Refactored ColocationLogger. Class originally deferred updating the list
of missing colocated children (i.e. removing regions from the missing
list that are now present) for a region until the next logging
interval. This has been changed to update the list on-demand, whenever
getMissingChildRegions() is called.

Added DUnit test for new functionality in show missing-disk-store
command.

Added JUnit tests for new classes

This closes #244


> Add log messages and enhance the output of list-missing-disk-stores to report 
> missing colocated regions
> -------------------------------------------------------------------------------------------------------
>
>                 Key: GEODE-1128
>                 URL: https://issues.apache.org/jira/browse/GEODE-1128
>             Project: Geode
>          Issue Type: Bug
>          Components: persistence
>            Reporter: Dan Smith
>            Assignee: Kenneth Howe
>
> Persistent recovery waits until all colocated regions are created. This is 
> necessary in order to maintain colocation. However, it can result in 
> surprising behavior if a user doesn't create a region or, in the case of 
> GEODE-1117, geode doesn't create an internal region.
> We need a warning level log message indicating that some colocated regions 
> are missing. We do have a log message after 15 seconds if other members have 
> not created persistent regions - see the code in 
> RedundancyLogger.logWaitingForMembers.
> We should enhance the logger to report any missing colocated regions as well.
> The list-missing-disk-stores command should also be enhanced to report 
> missing colocated regions



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Reply via email to