By My Oracle Support Team-Oracle on Jan 11, 2013
Leverage New High Availability Checks In Oracle RACcheck Utility
Oracle originally designed the Oracle RACcheck as a RAC Configuration Audit tool to review important configuration settings within a Real Application Clusters (RAC). Over time, we enhanced Oracle RACcheck to cover additional areas. These include single instance (non-rac) databases, Exadata, and now high availability (HA) checks.
This article will walk you through executing the new high availability (HA) checks in the RACcheck utility and reading the new report it generates.
Executing RACcheck with high availability checks
To execute RACcheck and include HA checks, you have two options:
$ ./raccheck -r (for regular healthcheck, including the new HA best practices)
$ ./raccheck -c hacheck (for only HA best practices)
Once you execute the command, RACcheck will detect and report all instances running on your server and give you the option to run all databases or a specific database. The list of databases will include any standby databases running on the server.
Once the execution completes, the RACcheck utility will create a report and a zip file containing all necessary files. This zip file can submitted to support for review. The results from our RACcheck example are:
Reading the report
Now we can look more closely at our results. Remember, we used the new flag to show only the high availability (HA) checks.
The report gives a general summary followed by a high-level list of items needing attention and/or items that passed the HA checks.
For each item, you are able to click on “view” to show details on the check and the settings the check recommends. In our example, if we look more closely at the failed check:
|SQL Check||The recommended minimum number of archive processes on PRIMARY DB is not set||All Databases||View|
We see both the current setting and the recommended change.
There are details on the instance, as well as a link to allow for additional information.
This is the first set of High Availability checks in the RACcheck tool, but we will add more in the future. If you have any input on these checks, suggestions on which checks we should add, or questions on the new HA checks, please join us in the discussion we created for this conversation in My Oracle Support Community within the Database Backup and Recovery Community.
RACcheck - RAC Configuration Audit Tool [ID 1268927.1]
Resolve - Troubleshooting and Preventing RAC issues with RACCheck [recorded Advisor Webcast; ID 1456176.1; Archived 2012 tab; Sept 12, 2012]
The new HA checks are:
- Check that database is running in archivelog mode
- Check that log_archive_config is set
- Check archive processes, log_archive_max_processes parameter, setting.
- Check that standby redo log files are configured.
- Check Status of the log transport services for destination
- Checks for a current gap for destination
- Check for at least one active RFS-Process when log transport services to a standby database are active
- Check current status of the managed recovery process
- Check that FAL is configured
- Check that data guard broker is running
- Check for any unrecoverable datafile with no backup following the nologging command
- Check for any offline datafiles.
- Check that controlfile autobackup is configured to ON
- Check that control_file_record_keep_time is not set too high or too low
- Check that archivelog deletion policy is configured
- For RAC database, check snapshot controlfile is configured and suggest it be placed on a on a shared device
- Check if flashback database is set to ON
- Check that recycle bin is ON
- When FRA is configured, check that archivelog destination is set to FRA and not directory directly