[jira] [Commented] (DIRSHARED-108) Recursively drilling into the directory structure causes java.lang.OutOfMemoryError

["Emmanuel Lecharny (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/DIRSHARED-108?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13038157#comment-13038157 ] 

Emmanuel Lecharny commented on DIRSHARED-108:
---------------------------------------------

I have tested the server and the API, injecting 30 000 entries within 9 levels, and everything works fine, assuming the server is started with enough memory.

I'll profile the code now

> Recursively drilling into the directory structure causes java.lang.OutOfMemoryError
> -----------------------------------------------------------------------------------
>
>                 Key: DIRSHARED-108
>                 URL: https://issues.apache.org/jira/browse/DIRSHARED-108
>             Project: Directory Shared
>          Issue Type: Bug
>    Affects Versions: 1.0.0-M1
>         Environment: Windows XP SP3
>            Reporter: Sebu Koleth
>             Fix For: 1.0.0-M4
>
>         Attachments: ApacheLdapAPITest.java, UnboundidLdapAPITest.java
>
>
>         private static void recursivelyDescend(LdapConnection connection, String dn) {
> 		System.out.println("Searching for children of dn : " + dn);
> 		try 
> 		{
> 			Cursor<SearchResponse> cursor = connection.search(dn, "(objectclass=*)", SearchScope.ONELEVEL, "*");
> 			while (cursor.next())
> 			{
> 				SearchResponse response = cursor.get();
> 				if(response instanceof SearchResultEntry) {
> 					recursivelyDescend(connection, ((SearchResultEntry)response).getObjectName().getName());
> 				} else {
> 					System.out.println("Unusable response type " + response);
> 				}
> 			}
> 		} catch (LdapException le) {
> 			le.printStackTrace();
> 		} catch (Exception e) {
> 			e.printStackTrace();
> 		}
> 	}
> The above piece of code is exercised after obtaining an SSL-based LDAP connection. The target server has hundreds of thousands of records at different levels. Logging at WARN level shows a *lot* of messages :
> WARN NioProcessor-1 org.apache.directory.shared.asn1.ber.Asn1Decoder - The PDU has been fully decoded but there are still bytes in the buffer.
> The code chokes at processing the 52nd entry that is two levels deep from the base DN. At this level there are around 1000 sub-levels.

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira