Hive queries in Ambari stay in 'running'-status

[#1]

Hi, I've installed a local HDP 2.3.2 Sandbox using VirtualBox. When I execute any query in Hive via Ambari (currently following Hello World tutorial) it most of the time keeps 'running' without results. Yarn job scheduler does not show any running jobs. I already saw a similar issue on this forum where it was solved by an Ambari restart, but in my case restarting does not solve it. What could be the problem?

[#2]

@Michel Meulpolder

VM configuration matters a lot. Hope you have enough memory and resources.

Try running the same query using hive cli

login to vm using ssh as root

su - hdfs

hive

run query from hive cli and see if its working or not.

[#3]

It might be due to missing home directories in HDFS. Depending on how you run the queries, you can try running the below:

#needed to run queries from Hive view in Ambari
sudo -u hdfs hdfs dfs -mkdir /user/admin
sudo -u hdfs hdfs dfs -chown admin /user/admin

#needed to run hive from CLI, when logged in as root
sudo -u hdfs hdfs dfs -mkdir /user/root
sudo -u hdfs hdfs dfs -chown root /user/root

#needed to run beeline from CLI
sudo -u hdfs hdfs dfs -mkdir /user/anonymous
sudo -u hdfs hdfs dfs -chown anonymous /user/anonymous

[#4]

Thanks Ali. I checked the directories and the above did not exist, but this does not seem to be the problem. When I use another browser it seems the queries run succesfully (see comment below).

[#5 · Accepted solution]

Thanks Neeraj. I did some more testing with my VM resources and also created another HDP Sandbox on Azure (A5 2 cores 14GB). It seems that the problem is browser-dependent since with all VM instances the problem randomly occurs while using Internet Explorer but not using Firefox. Anyhow, at least for now I can proceed, thanks again.

[#6]

@Michel Meulpolder

I have accepted this as best answer because you were able to proceed by changing the browser. Hope it's ok

[#7]

Thanks Neeraj