[GEODE-7481] can not start new member after config pdx with diskstore

• Type: Bug
• Status: Open
• Priority: Major
• Resolution: Unresolved
• Component/s: gfsh, management, rest (admin)
• Assignee: Unassigned
• Reporter: Gang Yan
• Created: 20/Nov/19 17:54
• Updated: 21/Nov/19 22:28

Description

the procedure:
1. start cluster, with 2 locator and 4 server
2. create disk-store
3. config pdx by REST API with diskstore
4. start new server

will met exception and can not start server, the exception is:

Exception in thread "main" java.lang.IllegalStateException: Cannot create region /PdxTypes DataPolicy withPersistence=true because another cache (10.118.20.125(g2-s2:1859)<v3>:41003) has the same region DataPolicy withPersistence=false. Persistent members must be started before non-persistent members

    at org.apache.geode.internal.cache.CreateRegionProcessor$CreateRegionMessage.handleCacheDistributionAdvisee(CreateRegionProcessor.java:472)

    at org.apache.geode.internal.cache.CreateRegionProcessor$CreateRegionMessage.process(CreateRegionProcessor.java:375)

    at org.apache.geode.distributed.internal.DistributionMessage.scheduleAction(DistributionMessage.java:372)

    at org.apache.geode.distributed.internal.DistributionMessage$1.run(DistributionMessage.java:436)

    at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)

    at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)

    at org.apache.geode.distributed.internal.ClusterOperationExecutors.runUntilShutdown(ClusterOperationExecutors.java:475)

    at org.apache.geode.distributed.internal.ClusterOperationExecutors.doHighPriorityThread(ClusterOperationExecutors.java:418)

    at org.apache.geode.logging.internal.executors.LoggingThreadFactory.lambda$newThread$0(LoggingThreadFactory.java:119)

    at java.lang.Thread.run(Thread.java:748)


And after destroy diskstore and all persistent region, can start new server again.