Trying to Cancel Long Running Impala Queries Fails with "Error communicating with the Impala Daemon"
(Doc ID 2154506.1)
Last updated on JANUARY 27, 2020
Applies to:Big Data Appliance Integrated Software - Version 2.5.0 and later
When attempting to cancel an Impala query through the Cloudera Manager Web Interface, an internal error is reported and the query is not canceled.
Note: Other issues such as SSL/TLS setup can also cause this message in the web UI.
In this case, clicking cancel on the query reports:
Clicking again reports:
The cloudera-scm-server.log shows an error similar to the following immediately after attempting to cancel a query:
java.io.IOException: Login failure for impala/<CLUSTER_NAME>.internal@<CLUSTER_NAME>.<DOMAIN> from keytab /tmp/<DIR>/hadoop<ID>.keytab
The keytab file and its enclosing directory listed in the error message do not exist at the time the error is generated.
After restarting the cloudera-scm-server, the keytab file is recreated, and queries can be canceled as expected. However, the issue will eventually recur. After the first recurrence, it will persist until the next cloudera-scm-server restart
To view full details, sign in with your My Oracle Support account.
Don't have a My Oracle Support account? Click to get started!
In this Document