Unable to collect data with Oracle Database Management Pack: Failed to test adapter: null
search cancel

Unable to collect data with Oracle Database Management Pack: Failed to test adapter: null

book

Article ID: 429419

calendar_today

Updated On:

Products

VCF Operations/Automation (formerly VMware Aria Suite)

Issue/Introduction

When testing the adapter connection, the following error message is consistently returned:

“Error | Failed to test adapter: null”



Error in OracleDatabaseAdapter_###.log:

DEBUG [Collector worker thread 11] (#########) com.bluemedora.vrops.compiler.execution.AdapterLiveCollector.getCurrentMetrics - Encountered a failure when collecting data: null
java.lang.UnsupportedOperationException: null
        at com.bluemedora.oracle.database.DataCollector.describeTestConnectionResult(DataCollector.kt:##) ~[oracle-database-dp-5.0.11-all.jar:?]
        at com.bluemedora.exuno.DataProvider.describeTestConnectionResultInternal(DataProvider.kt:###) ~[ex-uno-6.2.0.jar:?]
        at com.bluemedora.exuno.DataProvider.addTestConnectionResultDefinitions(DataProvider.kt:###) ~[ex-uno-6.2.0.jar:?]
        at com.bluemedora.exuno.DataProvider.describeCollectionResultInternal(DataProvider.kt:###) ~[ex-uno-6.2.0.jar:?]
        at com.bluemedora.exuno.DataProvider.setupThreadLocal(DataProvider.kt:###) ~[ex-uno-6.2.0.jar:?]
        at com.bluemedora.exuno.DataProvider.withThreadLocal(DataProvider.kt:###) ~[ex-uno-6.2.0.jar:?]
        at com.bluemedora.exuno.DataProvider.collect(DataProvider.kt:###) ~[ex-uno-6.2.0.jar:?]
        at com.bluemedora.vrops.compiler.execution.configuration.DataProviderConfiguration.runCollection(DataProviderConfiguration.kt:###) ~[library-10.7.0.jar:?]
        at com.bluemedora.vrops.compiler.execution.AdapterLiveCollector.getCurrentMetrics(AdapterLiveCollector.kt:##) ~[library-10.7.0.jar:?]
        at com.vmware.tvs.vrealize.adapter.core.collection.live.LiveCollectionExecutor.getCurrentMetricsSynchronous(LiveCollectionExecutor.java:##) ~[aria-ops-core-8.2.0.jar:?]
        at com.vmware.tvs.vrealize.adapter.core.UnlicensedAdapter.onCollect(UnlicensedAdapter.java:###) ~[aria-ops-core-8.2.0.jar:?]
        at com.vmware.tvs.vrealize.adapter.core.UnlicensedAdapter.onCollect(UnlicensedAdapter.java:###) ~[aria-ops-core-8.2.0.jar:?]
        at com.bluemedora.vrops.compiler.execution.UnlicensedManagementPackExecutor.onCollect(UnlicensedManagementPackExecutor.kt:66) ~[library-10.7.0.jar:?]

Environment

Aria Operations 8.18.x
Oracle Database Management Pack 9.0.0.0.24730771

Cause

  • The latest release of the Oracle DB Management Pack (version 9.0.0.0.24730771) includes two versions of the Ex Uno library:
  • 6-6.6.0.jar – the version intended for use with this Management Pack release
  • ex-uno-6.2.0.jar – an earlier version of the library
  • The Management Pack does not enforce a specific load order for JAR files. As a result, the first reference to an Ex Uno class may be resolved from either library.
  • If the earlier ex-uno-6.2.0.jar is loaded first, the newer 6-6.6.0.jar may not be loaded, which can cause the Management Pack to fail.
  • Because the JAR load order is non-deterministic, the issue may occur intermittently. The Management Pack may appear to function normally or encounter failures depending on which Ex Uno library is loaded first after restarting the collector service.

Resolution

This issue has been identified in Oracle DB Management Pack version 9.0.0.0.24730771. A permanent fix will be included in a future release of the Oracle DB Management Pack.
 
Workaround:
 
Apply the following workaround only to the nodes / Cloud Proxies (CPs) where an Oracle DB Management Pack adapter instance is currently running.

To fully resolve the issue across the environment, perform these steps on all nodes / Cloud Proxies (CPs).
 
1. SSH into the collecting Aria Operations node as root.
2. Remove the obsolete JAR file from the Oracle DB Management Pack library directory:
rm /usr/lib/vmware-vcops/user/plugins/inbound/oracledatabase_adapter_3/lib/ex-uno-6.2.0.jar
3. Restart the collector service:
systemctl restart collector
 
Note: If the Management Pack is reinstalled, this workaround must be applied again.

Additional Information

References: