[core] Fix ping ponging of job browser API between RM HA servers

Currently ping-pong can be seen in hue log.

[03/Nov/2016 14:01:26 -0700] connectionpool DEBUG "machine-2:8090 GET /ws/v1/cluster/info HTTP/1.1" 200 None
[03/Nov/2016 14:01:26 -0700] kerberos_ DEBUG handle_mutual_auth(): Handling: 200
[03/Nov/2016 14:01:26 -0700] kerberos_ DEBUG handle_response(): returning <Response [200]>
[03/Nov/2016 14:01:26 -0700] resource DEBUG GET Got response: {"clusterInfo":{"id":1478206327204,"startedOn":1478206327204,"state":"STARTED","haState":"STANDBY","rmStateStoreName":"org.apache.hadoop.yarn.server.resourcemanager.recovery.ZKRMStateStore","resourceManagerVersion":"2.6.0-cdh5.10.0-SNAPSHOT","resourceManagerBuildVersion":"2.6.0-cdh5.10.0-SNAPSHOT from 6ee762ff6775232b2e97c1644454737fe7f36ee8 by jenkins source checksum 7ed9df10ccc27e9b51c9c94cdb7c8493","resourceManagerVersionBuiltOn":"2016-11-02T12:10Z","hadoopVersion":"2.6.0-cdh5.10.0-SNAPSHOT","hadoopBuildVersion":"2.6.0-cdh5.10.0-SNAPSHOT from 6ee762ff6775232b2e97c1644454737fe7f36ee8 by jenkins source checksum a7fb82deaabdf4e32654ca44c4e7d2d5","hadoopVersionBuiltOn":"2016-11-02T12:02Z","haZooKeeperConnectionState":"TERMINATED"}}
[03/Nov/2016 14:01:26 -0700] cluster INFO RM default is not RUNNING, skipping it: {u'clusterInfo': {u'hadoopVersionBuiltOn': u'2016-11-02T12:02Z', u'startedOn': 1478206327204, u'haState': u'STANDBY', u'haZooKeeperConnectionState': u'TERMINATED', u'resourceManagerBuildVersion': u'2.6.0-cdh5.10.0-SNAPSHOT from 6ee762ff6775232b2e97c1644454737fe7f36ee8 by jenkins source checksum 7ed9df10ccc27e9b51c9c94cdb7c8493', u'resourceManagerVersion': u'2.6.0-cdh5.10.0-SNAPSHOT', u'state': u'STARTED', u'hadoopVersion': u'2.6.0-cdh5.10.0-SNAPSHOT', u'resourceManagerVersionBuiltOn': u'2016-11-02T12:10Z', u'rmStateStoreName': u'org.apache.hadoop.yarn.server.resourcemanager.recovery.ZKRMStateStore', u'hadoopBuildVersion': u'2.6.0-cdh5.10.0-SNAPSHOT from 6ee762ff6775232b2e97c1644454737fe7f36ee8 by jenkins source checksum a7fb82deaabdf4e32654ca44c4e7d2d5', u'id': 1478206327204}}
[03/Nov/2016 14:01:26 -0700] connectionpool DEBUG "machine-1:8090 GET /ws/v1/cluster/info HTTP/1.1" 200 None
[03/Nov/2016 14:01:26 -0700] kerberos_ DEBUG handle_mutual_auth(): Handling: 200
[03/Nov/2016 14:01:26 -0700] kerberos_ DEBUG handle_response(): returning <Response [200]>
[03/Nov/2016 14:01:26 -0700] resource DEBUG GET Got response: {"clusterInfo":{"id":1478206331873,"startedOn":1478206331873,"state":"STARTED","haState":"ACTIVE","rmStateStoreName":"org.apache.hadoop.yarn.server.resourcemanager.recovery.ZKRMStateStore","resourceManagerVersion":"2.6.0-cdh5.10.0-SNAPSHOT","resourceManagerBuildVersion":"2.6.0-cdh5.10.0-SNAPSHOT from 6ee762ff6775232b2e97c1644454737fe7f36ee8 by jenkins source checksum 7ed9df10ccc27e9b51c9c94cdb7c8493","resourceManagerVersionBuiltOn":"2016-11-02T12:10Z","hadoopVersion":"2.6.0-cdh5.10.0-SNAPSHOT","hadoopBuildVersion":"2.6.0-cdh5.10.0-SNAPSHOT from 6ee762ff6775232b2e97c1644454737fe7f36ee8 by jenkins source checksum a7fb82deaabdf4e32654ca44c4e7d2d5","hadoopVersionBuiltOn":"2016-11-02T12:02Z","haZooKeeperConnectionState":"CONNECTED"}}
[03/Nov/2016 14:01:26 -0700] cluster WARNING Picking RM HA: ha
[03/Nov/2016 14:01:26 -0700] connectionpool DEBUG "machine-1:8090 GET /ws/v1/cluster/apps?finalStatus=UNDEFINED&limit=1000&user=admin&doAs=admin HTTP/1.1" 200 None
[03/Nov/2016 14:01:26 -0700] kerberos_ DEBUG handle_mutual_auth(): Handling: 200
[03/Nov/2016 14:01:26 -0700] kerberos_ DEBUG handle_response(): returning <Response [200]>