Member since
05-16-2016
785
Posts
114
Kudos Received
39
Solutions
My Accepted Solutions
| Title | Views | Posted |
|---|---|---|
| 2329 | 06-12-2019 09:27 AM | |
| 3590 | 05-27-2019 08:29 AM | |
| 5728 | 05-27-2018 08:49 AM | |
| 5245 | 05-05-2018 10:47 PM | |
| 3114 | 05-05-2018 07:32 AM |
08-17-2017
08:54 PM
if you have configured mysql , oracle database for hive metastore then check if you had placed the appropriate jdbc jar file inside /usr/lib/hive/lib if you didnt see any jdbc jar file please make sure you have that inside the above location also make sure you have the class entry in the hive-site.xml in the below property javax.jdo.option.ConnectionDriver let me know if that helps
... View more
08-15-2017
10:21 PM
Sqoop does not come with the JDBC driver , as you already knew that sqoop is just a client tool it uses the JDBC driver to connect to data source in our case its MYSQL , if you look under the directory /var/lib/sqoop/ should have a MYSQL jdbc driver , internally it will take care of all the connections ,query execution , result sets etc . Since we are using localhost it will connect the locally runining server , if you want to access the remote server then we use 3306 , however you can also mention the port number when refering to localhost , i dont have the vm to see their socket file configuration . moreover you have to enforce the TCP/IP connection because by default when we use localhost it will refer the socket file . Hope this helps .
... View more
08-12-2017
05:20 AM
1 . Check the policy file 2 . Check if the user "sunil " is in Impala group . if nothing helps to dig more use the safety valve to enable log4j root logger and share the logs if you can log4j.logger.org.apache.sentry=DEBUG
... View more
08-10-2017
10:33 PM
1 Kudo
The best practice is to set 8 gb aside for the os and remaining for the dameons . Are you refering to JAVA_TOOL_OPTIONS property . I would start with 8G bump it when hit outofMemory during intial setup .
... View more
08-10-2017
10:24 PM
Somtimes I prefer bucketing over Partition due to large number of files getting created . Also in bucketing actually you have the control over the number of buckets. I would suggest you test the bucketing over partition in your test env also it is a good practice to collect statistics for the table it will help in the performance side .
... View more
08-10-2017
10:16 PM
Does your impala table has statistic data being collected ? that will sure help in performance side. do you have any compression involved ?
... View more
08-10-2017
10:11 PM
Could you share any error logs or screenshot ? where are you facing the issue ? Were you able to see the CDH version in the Host summary ?
... View more
08-10-2017
12:13 AM
could you share your /etc/hosts ? status of your firewall ? what os are you runining ?
... View more
08-10-2017
12:06 AM
if you have the same error , then the suggest work around is to unmount the "tmpfs " To know the location of tmpfs path check your /etc/fstab file you will see the mount point path .
... View more
08-08-2017
11:22 PM
Yes Yarn restart is required , if you are using Cloudera manager then a Stale restart is required
... View more