Logs for SQL Processes
This article explains how to generate logs for SQL processes.
If you are experiencing problems with a specific SQL statement, you need to provide the log files for the SQL processes.
Prerequisites
You must have enough free disk space for the logs. For more information, see EXAsupport.
Procedure
The following examples use the c4 and confd_client command-line tools in a Linux terminal. For more information about these tools, see Exasol Deployment Tool (c4) and ConfD.
-
Connect to EXAClusterOS (COS) on the cluster using
c4 connect -t <DEPLOYMENT>[.<NODE>]/cos
. For example:If you do not specify a node, c4 will connect to the first active node in the deployment.
For more information about how to use
c4 connect
, see How to use c4. -
To find the name of the database, use the ConfD job db_list. For example:
-
To collect information about SQL processes, use the EXAsupport tool. Specify the start date (
-s
) and end date (-t
) using the formatYYYY-MM-DD
.You can collect information for all SQL processes, or for a specific session identified by its session ID.
In Exasol versions prior to 8.25.0, you can only collect SQL logs for a specific session on the day that the session occurred. Later versions do not have this limitation.
For example, to collect information about all SQL processes:
To collect information about a specific SQL session:
If the log file was successfully created, you will see the following message containing the file name:
-
Use CTRL+D or type
exit
to disconnect from COS. -
Copy the log file to your local computer by connecting to COS and using the
cat
command:Example:
Verification
To verify that the log was successfully copied, use the ls
command in the target folder on your local machine.
Example
ls -lahtr ~/mylogs | grep exacluster_debuginfo
-rwxr-xr-x 1 user group 9.8K Sep 20 17:38 exacluster_debuginfo_2022_08_11-23_27_20.tar.gz