HDB Won't Start - Nameserver Crashing

[Former Member]

Hi,

Has anybody come across this error before and can help please?

I am trying to start HDB after initially trying to perform an import from another system using Sapinst (Database Refresh option).  Sapinst performed the Create and Load step of the database successfully, but then it timed out on the hdbnsutil -initTopology step.  I aborted the sapinst import and have since tried performing a restore in Hana Studio and also a manual inittopology command, but the HDB will not start.

Error from the nameserver_alert_<hostname>.trc file is:

NOTE: full crash dump will be written to /usr/sap/BWQ/HDB35/augdnspi51/trace/nameserver_augdnspi51.33501.crashdump.20141118-201115.036400.trc

Call stack of crashing context:

1: 0x00007f159d6c6901 in TrexThreads::Thread::getState() const+0x0 at Thread.cpp:145 (libhdbbasement.so)

2: 0x00007f159ed39d3a in NameServer::BaseThread::stop(int*)+0x16 at ThreadBase.h:27 (libhdbns.so)

3: 0x00007f159ed0627f in NameServer::TREXNameServer::prepareToShutdown()+0xe0b at TREXNameServer.cpp:2833 (libhdbns.so)

[36413]{-1}[-1/-1] 2014-11-18 20:11:15.958504 e Basis            TREXNameServer.cpp(02906) : Process exited due to an error via explicit exit call with exit code 1 , no crash dump will be written

[40995]{-1}[-1/-1] 2014-11-18 20:34:41.892397 e tns_ddl          TNSClient.cpp(00917) : setServiceType no topology databaseId use service default:  2

[41233]{-1}[-1/-1] 2014-11-18 20:34:53.769741 e sr_nameserver    TNSClient.cpp(06756) : unknown data center 2 when sending request dr_retrieveinifilecontents

[41008]{-1}[-1/-1] 2014-11-18 20:38:33.783375 e assign           TREXNameServer.cpp(01810) : assign failed with ltt exception. stopping service...

exception  1: no.3000334  (DataAccess/impl/DisasterRecoveryPrimaryImpl.cpp:1031)

    Operation aborted, service in shutdown

exception throw location:

1: 0x00007fb4f86c1c91 in DataAccess::ShutdownAwareBarrier::wait()+0x50 at DisasterRecoveryPrimaryImpl.cpp:1031 (libhdbdataaccess.so)

2: 0x00007fb4f86c416d in DataAccess::SessionRegistry::SessionRegistryItem::waitUntilActive(DataAccess::ShutdownAwareBarrier*)+0x89 at DisasterRecoveryPrimaryImpl.cpp:1441 (libhdbdataaccess.so)

3: 0x00007fb4f86c439b in DataAccess::DisasterRecoveryPrimaryHandlerImpl::waitUntilSecondaryIsActive()+0xa7 at DisasterRecoveryPrimaryImpl.cpp:1053 (libhdbdataaccess.so)

4: 0x00007fb4fe0261c1 in NameServer::TREXNameServer::assign(NameServer::ServiceStartInfo&)+0x20d0 at TREXNameServer.cpp:1760 (libhdbns.so)

5: 0x00007fb4fe0566c7 in NameServer::SelfAssignThread::run(void*)+0x13 at TREXNameServer.cpp:302 (libhdbns.so)

6: 0x00007fb4fc9ea1f2 in TrexThreads::PoolThread::run()+0x850 at PoolThread.cpp:265 (libhdbbasement.so)

7: 0x00007fb4fc9ebd58 in TrexThreads::PoolThread::run(void*&)+0x14 at PoolThread.cpp:124 (libhdbbasement.so)

8: 0x00007fb4fb088ccf in Execution::Thread::staticMainImp(void**)+0x98b at Thread.cpp:475 (libhdbbasis.so)

9: 0x00007fb4fb08921d in Execution::Thread::staticMain(void*)+0x39 at Thread.cpp:545 (libhdbbasis.so)

[42520]{-1}[-1/-1] 2014-11-18 20:38:56.352439 e tns_ddl          TNSClient.cpp(00917) : setServiceType no topology databaseId use service default:  2

[42759]{-1}[-1/-1] 2014-11-18 20:39:08.217642 e sr_nameserver    TNSClient.cpp(06756) : unknown data center 2 when sending request dr_retrieveinifilecontents

Thanks,

Chaz.

[Former Member]

Fixed.

The issue was found to be related to replication settings performed for testing recently.

Ran replication commands to cleanup the replication config on both primary and secondary, then ran initttopology command on both as well.  HDB then started ok.

[andreas_zigann]

Hello Chaz,

where can I find the guide with the replication and inittopology commands?

I got Problems with the replication database, too. I can not start it anymore, so I can not configure it with HANA Studio.

Best Regards

Andreas

[andreas_zigann]

Hello Chaz,

I got my instance running.

I had to

stop the replication database,

execute hdbnsutil -sr_register --remoteHost=... --remoteInstance=00 --mode=sync -name=HDBREP

start the replication database

That was enough at my error. I did not Need to init the topology.

Bye

Andreas