AnsweredAssumed Answered

Total memory required to run all configured services is exceeding available memory on the box....

Question asked by subroto on Jul 14, 2013
Latest reply on Jul 15, 2013 by subroto
Hi,

I have tried using ec2 instances of type m1.large and m1.xlarge but, I keep getting the error like:
<pre>
2013-07-15 08:57:40,151 INFO  com.mapr.warden.service.baseservice.Service [main]: Connected to ZK: ip-10-140-5-194.ec2.internal:5181With State: State:CONNECTED Timeout:30000 sessionid:0x13fe2685e280009 local:/10.140.5.194:45281 remoteserver:ip-10-140-5-194.ec2.internal/10.140.5.194:5181 lastZxid:0 xid:1 sent:1 recv:1 queuedpkts:0 pendingresp:0 queuedevents:0
2013-07-15 08:57:40,152 INFO  com.mapr.warden.WardenManager [main]: Total node memory: 7864460
2013-07-15 08:57:40,152 FATAL com.mapr.warden.WardenManager manage [main]:
--------------- Total memory required to run all configured services is exceeding available memory on the box.
--------------- Total memory needed: 8294MB where Total node memory is: 7680MB
--------------- EXITING
2013-07-15 08:57:40,177 INFO  com.mapr.warden.WardenMain [Thread-2]: ShutdownHook started
2013-07-15 08:57:40,177 INFO  com.mapr.warden.centralconfig.PullCentralConfigTaskScheduler [Thread-2]: Cancelling outstanding PullCentralConfigTask
2013-07-15 08:57:40,177 WARN  com.mapr.warden.service.baseservice.zksessionmgmnt.ZookeeperClientSessionManagement [Thread-6]: Reinit Service thread's for Warden was interrupted. Most likely because of the shutdown.
2013-07-15 08:57:40,182 INFO  com.mapr.warden.WardenServer [main-EventThread]: Process path: /servers/ip-10-140-5-194.ec2.internal. Event state: SyncConnected. Event type: NodeDeleted
</pre>
Do I need to tweak any other configuration to get the nodes running ?

Cheers,
Subroto Sanyal

Outcomes