[OPENDJ-2223] Administration Connector hangs while deleting backend Created: 23/Jul/15  Updated: 08/Nov/19  Resolved: 17/May/16

Status: Done
Project: OpenDJ
Component/s: backends
Affects Version/s: 2.6.2
Fix Version/s: Not applicable

Type: Bug Priority: Major
Reporter: patrick diligent Assignee: Chris Ridd
Resolution: Cannot Reproduce Votes: 0
Labels: release-notes

Dev Assignee: Chris Ridd

 Description   

User creating by mistake backend on a suffix for which the root entry is already contained in the userRoot backend, and attempted to configure replication, with initialisation failing. After disabling the replication and attempting to delete the backend, the thread blocks within the backend delete operation, preventing further administration to the server; to recover from this situation the server process had to be killed.

Relevant stacks below, showing two threads stuck :

LDAP Request Handler 0 for connection handler Administration Connector 0.0.0.0 port 4444" prio=10 tid=0x00007f1cf85c6800 nid=0x1c19 waiting on condition [0x00007f1ca316f000]
   java.lang.Thread.State: TIMED_WAITING (sleeping)
    at java.lang.Thread.sleep(Native Method)
    at org.opends.server.backends.jeb.BackendImpl.waitUntilQuiescent(BackendImpl.java:212)
    at org.opends.server.backends.jeb.BackendImpl.finalizeBackend(BackendImpl.java:413)
    at org.opends.server.core.BackendConfigManager.applyConfigurationDelete(BackendConfigManager.java:1235)
    at org.opends.server.core.BackendConfigManager.applyConfigurationDelete(BackendConfigManager.java:69)
    at org.opends.server.admin.server.ServerManagedObjectDeleteListenerAdaptor.applyConfigurationDelete(ServerManagedObjectDeleteListenerAdaptor.java:74)
    at org.opends.server.admin.server.ConfigDeleteListenerAdaptor.applyConfigurationDelete(ConfigDeleteListenerAdaptor.java:187)
    at org.opends.server.extensions.ConfigFileHandler.deleteEntry(ConfigFileHandler.java:1593)
    - locked <0x000000071722c820> (a java.lang.Object)
    at org.opends.server.workflowelement.localbackend.LocalBackendDeleteOperation.processLocalDelete(LocalBackendDeleteOperation.java:341)
    at org.opends.server.workflowelement.localbackend.LocalBackendWorkflowElement.execute(LocalBackendWorkflowElement.java:556)
    at org.opends.server.core.WorkflowImpl.execute(WorkflowImpl.java:197)
    at org.opends.server.core.WorkflowTopologyNode.execute(WorkflowTopologyNode.java:100)
    at org.opends.server.core.DeleteOperationBasis.run(DeleteOperationBasis.java:417)
    at org.opends.server.core.SynchronousStrategy.enqueueRequest(SynchronousStrategy.java:49)
    at org.opends.server.protocols.ldap.LDAPClientConnection.addOperationInProgress(LDAPClientConnection.java:1269)
    at org.opends.server.protocols.ldap.LDAPClientConnection.processDeleteRequest(LDAPClientConnection.java:2130)
    at org.opends.server.protocols.ldap.LDAPClientConnection.processLDAPMessage(LDAPClientConnection.java:1721)
    at org.opends.server.protocols.ldap.LDAPRequestHandler.run(LDAPRequestHandler.java:194)

"Replication server RS(27161) writing to Replication server RS(31538) for domain "o=business,dc=example,dc=com" at w3prd1core4.example.com/10.242.69.46:40553" prio=10 tid=0x00007f1cb40ae800 nid=0x4b72 in Object.wait() [0x00007f1c95090000]
   java.lang.Thread.State: TIMED_WAITING (on object monitor)
	at java.lang.Object.wait(Native Method)
	at org.opends.server.replication.server.MessageHandler.getNextMessage(MessageHandler.java:428)
	- locked <0x000000079d13e6f8> (a org.opends.server.replication.server.MsgQueue)
	at org.opends.server.replication.server.ServerHandler.take(ServerHandler.java:1097)
	at org.opends.server.replication.server.ReplicationServerDomain.take(ReplicationServerDomain.java:1398)
	at org.opends.server.replication.server.ServerWriter.run(ServerWriter.java:108)

The backend "ou=business,dc=example,dc=com" should have had replication disabled, and is the one for which deletion is blocked.



 Comments   
Comment by Matthew Swift [ 12/Mar/16 ]

Needs re-evaluating: may be resolved in 3.0.0.

Comment by Matthew Swift [ 17/May/16 ]

We've not been able to reproduce this issue.

Comment by Matthew Swift [ 07/Nov/19 ]

Moved to closed state because the fixVersion has already been released.

Generated at Tue Jul 07 22:27:40 UTC 2020 using Jira 7.13.12#713012-sha1:6e07c38070d5191bbf7353952ed38f111754533a.