You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@directory.apache.org by "Stefan Seelmann (JIRA)" <ji...@apache.org> on 2016/10/02 16:48:20 UTC

[jira] [Commented] (DIRSERVER-2113) Integrate the 'partition-plumber' into ApacheDS

    [ https://issues.apache.org/jira/browse/DIRSERVER-2113?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15540636#comment-15540636 ] 

Stefan Seelmann commented on DIRSERVER-2113:
--------------------------------------------

Trunk looks a bit different (sysout instead of usage of logger).

When I remove the "start" I get an NPE:
{noformat}
java.lang.NullPointerException
	at org.apache.directory.server.ApacheDsService.getDirectoryService(ApacheDsService.java:587)
	at org.apache.directory.server.ApacheDsService.repair(ApacheDsService.java:251)
	at org.apache.directory.server.UberjarMain.repair(UberjarMain.java:192)
	at org.apache.directory.server.UberJarMainTest.repairTest(UberJarMainTest.java:364)
{noformat}

Anyway, in Studio I'll call the action"Start and Repair" for now, later we can change it again.

> Integrate the 'partition-plumber' into ApacheDS
> -----------------------------------------------
>
>                 Key: DIRSERVER-2113
>                 URL: https://issues.apache.org/jira/browse/DIRSERVER-2113
>             Project: Directory ApacheDS
>          Issue Type: New Feature
>    Affects Versions: 2.0.0-M20
>            Reporter: Emmanuel Lecharny
>            Priority: Critical
>             Fix For: 2.0.0-M22
>
>
> Kiran has developped a small tool that fix a corrupted JDBM database : the {{partition-plumber}} (see https://svn.apache.org/repos/asf/directory/sandbox/kayyagari/partition-plumber/).
> This is the only way we can get back on our feet when the database gets corrupted (if and only if the MasterTable is not itself corrupted, something that happens quite rarely).
> I suggest strongly we inject this smart tool in the server so that a user with a corrupted database can repair it by starting the server with a special option (something like apacheds -repair).
> That would save our users a lot of pain (like reloading the full database, assuming they have a recent backup...)



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