Suggested maintenance?
28 February, 2010
Hi,
Just wondering if you have any suggested maintenance that should be done to the Yellowfin application. We do normal maintenance on our servers, SQL tables, etc.
Are there any built-in tools/scripts/etc that we can or should run?
Just wondering if you have any suggested maintenance that should be done to the Yellowfin application. We do normal maintenance on our servers, SQL tables, etc.
Are there any built-in tools/scripts/etc that we can or should run?
Generally speaking, the Yellowfin application will handle most required maintenance tasks itself, via background tasks when the server is running. There are a few things you can do that will help ensure Yellowfin is running at optimal performance.
Database backups - for production systems it is recommended to take regular backups of the Yellowfin database. This gives you recovery options if the database becomes corrupted, or you need to restore content from a previous date.
Trim the audit log. In the Yellowfin database, the Event table contains log items for many events within Yellowfin, from users logging in and running reports, to administration items such as creating data sources and views. This table can get very large, and as it grows it takes longer to query and to insert records. If you don't need a full audit history, you can delete old records from this table to increase performance. We are currently looking at ways to improve the performance in this area, and will likely make changes in a future release.
Log file maintenance. Log files are stored in the Yellowfin/appserver/logs/ directory. There are a number of different log files for different parts of the application. Some are handled by log4j (a Java logging library), which allows you to set up rotating log files (eg files won't be larger than a certain size, and only the latest 10 files will be retained). Some are not handled by log4j, and can get very large. The main one to be aware of is jdbc.log - the log file for the Yellowfin configuration database's connection pool. You can trim this file (or delete it) periodically. This should be done while Yellowfin is not running. Log files for source database connections can also get quite large, depending on the log level settings. In a future release these file will be managed by log4j as well, so this will not be a problem.
Thanks,
Steve
Database backups - for production systems it is recommended to take regular backups of the Yellowfin database. This gives you recovery options if the database becomes corrupted, or you need to restore content from a previous date.
Trim the audit log. In the Yellowfin database, the Event table contains log items for many events within Yellowfin, from users logging in and running reports, to administration items such as creating data sources and views. This table can get very large, and as it grows it takes longer to query and to insert records. If you don't need a full audit history, you can delete old records from this table to increase performance. We are currently looking at ways to improve the performance in this area, and will likely make changes in a future release.
Log file maintenance. Log files are stored in the Yellowfin/appserver/logs/ directory. There are a number of different log files for different parts of the application. Some are handled by log4j (a Java logging library), which allows you to set up rotating log files (eg files won't be larger than a certain size, and only the latest 10 files will be retained). Some are not handled by log4j, and can get very large. The main one to be aware of is jdbc.log - the log file for the Yellowfin configuration database's connection pool. You can trim this file (or delete it) periodically. This should be done while Yellowfin is not running. Log files for source database connections can also get quite large, depending on the log level settings. In a future release these file will be managed by log4j as well, so this will not be a problem.
Thanks,
Steve