@Knows NotMuch That's exactly the desired output "Ambari Server 'setup' completed successfully". Now if you start your hive 2server you shouldn't get the previous error. Please let me know after attempting the H2S startup. ... View more

Thanks Timothy !!! ... View more

@Tanu Sood Indeed this has become a very unbearable condition for anyone taking an HW exam. I am just surprised and in astonishment that despite all the complaints HW exam team have NOT resolved this damn issue. Every week there is a minimum of 4 complaints and the numbers are growing I sat for my Oracle Certified Professional (OCP), CCNA, CCNP but I never got the issues I encountered when I tried HDPCA which after 1 hour of tussling I gave up, the environment was horribly slow and unstable loss of connectivity and the Proctor was an asshole to crown it all. Panning my office desk almost 4 times,I had to remove other screens and keyboards, pens just name it ..... Thank God I passed it despite not completing the questions because for sure HW wasn't going to give me a second eCredit !!! I really don't know whether HW is aware of the frustration and how such situations can sway someone to go Cloudera....basically the underlying technology is the same save for a few components and it's not with such comments that HW can compete with others !!:-) I pray this time some HW employees in HCC take it seriously what a good product and a poor disastrous exam setup. ... View more

@sudi ts You need to copy the connector into the hadoop-client and hive-client location otherwise you will hit an error cp gcs-connector-latest-hadoop2.jar /usr/hdp/current/hadoop-client/lib/ cp gcs-connector-latest-hadoop2.jar /usr/hdp/current/hive-client/lib The below command should run successfully $ hdfs dfs -ls gs://bucket/ This should run fine, but the issue you are having is with permission for hdpuser1 you will need to correct by running $ hdfs dfs -chown hdpuser1 gs://bucket/ Now your create table should work, while logged in as hdpuser1 CREATE EXTERNAL TABLE test1256(name string,id int) LOCATION 'gs://bucket/'; Please let me know. If you found this answer addressed your question, please take a moment to log in and click the "Accept" link on the answer. ... View more

@Saurabh Srivastava I noticed you created 2 similar threads, please can you delete the other one. ... View more

@Saurabh Srivastava The maximum transmission unit (MTU) could have been set too low the default is 1500 bytes, change it to 9000 recommended see HCC document Typical HDP Cluster Network Configuration Best Practices do this for all the nodes in the cluster Check the current MTU setting with ifconfig or ip link list command under Linux look at the fourth line # /sbin/ifconfig eth1 Link encap:Ethernet HWaddr 08:00:27:D8:06:8F inet addr:192.168.0.171 Bcast:192.168.0.255 Mask:255.255.255.0 inet6 addr: fe80::a00:27ff:fed8:68f/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:786 errors:0 dropped:0 overruns:0 frame:0 TX packets:444 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:455353 (444.6 KiB) TX bytes:57860 (56.5 KiB) You can also use $ ip link list 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 2: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000 link/ether 08:00:27:d8:06:8f brd ff:ff:ff:ff:ff:ff As you see, MTU set to 1500 for eth1. You should optimize to let's say 9000 then you can use any one of the following commands to setup MTU: # ifconfig eth1 mtu 9000 or # ip link set dev eth1 mtu 9000 Verify that new MTU is setup with the following command: $ ip link list or $ /sbin/ifconfig Edit /etc/sysconfig/network-scripts/ifcfg-ethx (Red Hat Linux ) to permanently change the MTU,notice I have added MTU=9000 DEVICE=eth0 HWADDR=08:00:27:FF:AF:39 TYPE=Ethernet UUID=4df8046e-59c0-4667-9a6e-daa61b83682c ONBOOT=no MTU=9000 NM_CONTROLLED=yes BOOTPROTO=dhcp Restart the network on Redhat/Centos # service network restart Reference: https://www.pcmech.com/article/jumbo-frames/ Now retry you copy, Hadoop 2.0 should be good with big files Please let me know if the copy timing improved drastically ... View more