Member since
08-16-2016
642
Posts
131
Kudos Received
68
Solutions
My Accepted Solutions
| Title | Views | Posted |
|---|---|---|
| 3976 | 10-13-2017 09:42 PM | |
| 7472 | 09-14-2017 11:15 AM | |
| 3797 | 09-13-2017 10:35 PM | |
| 6031 | 09-13-2017 10:25 PM | |
| 6600 | 09-13-2017 10:05 PM |
08-14-2017
11:04 AM
Don't know the exact cause that it was not working even i restarted the Yarn, when i had another chance to restart the Yarn, the issue was resolved.
... View more
08-11-2017
03:16 AM
So, summarizing: When starting Spark with --deploy-mode client one must distribute Jaas file and keytab to every machine. When starting Spark with --deploy-mode cluster one must distribute Jaas file and keytab with --files and/or --keytab.
... View more
08-07-2017
02:58 AM
After adding the "classpath" when I execute, hadoop classpath, it doesn't return the path that I have added. Any ideas?
... View more
08-01-2017
09:53 AM
Ok @mbigelow thanks. I also reasearched more and found this hortonworks link, https://community.hortonworks.com/questions/45962/dataxceiver-error-processing-write-block-operation.html where they say we can ignore this error. This issue has been already fixed in 2.3 version of ambari but of course I am using CDH. Thanks, Shilpa
... View more
08-01-2017
08:04 AM
Thanks i was doing something for short term, worked.
... View more
07-31-2017
02:21 AM
Hi , Idle session works indeed . I have set it to 1 hour but can not wait 1 hour untill killed queries will be dead . Work arround that I found is disconnecting from the session after query has been cancelled in the UI . It should work better . Thanks Alon
... View more
07-28-2017
09:58 AM
@mbigelow Thank you, that solved the issue. The /opt/cloudera/csd dir was empty, I downloaded CSD and configured it and then re-added service. The above exception is resolved.
... View more
07-26-2017
12:51 PM
Hive has the limitation that it cannot set these values at runtime. They have to either be in the core-site.xml or the table definition (not positive on this one though). The former will require admin access to the CM UI. The latter can be done by the person creating the table, you. Set the location as such, s3a://ACCESS_KEY:SECRET_KEY/path/. This will not work if there is a '/' unless you have this patch. https://issues.apache.org/jira/browse/HADOOP-3733
... View more
07-26-2017
10:41 AM
I am going to repost this as an HBASE SNAPSHOT question.
... View more
07-25-2017
11:00 PM
I've explored idle query timeout and idle session out before itself and using in prod cluster as well. Those timeouts are not related to idle connections. Idle session timeout is taking care of only session, but still connection is established and being counted under active connections in use. On reading about SocketTimeout option in jdbc driver, it clearly said it will take care of idle connections. Hence tried it out, but doesn't seem to work. Most of users here, using sql workbench through jdbc driver keeps opening new tab for every new query and don't close their tab for the whole day. I assumed sockettimeout will solve this problem.
... View more