Database Health Monitoring

Cisco UCS Manager Database Health Monitoring

Cisco UCS Manager uses a SQLite database stored on the Fabric Interconnects to persist configuration and inventory. Data corruption on both the Flash and NVRAM storage devices can cause failures and loss of customer configuration data. Cisco UCS Manager provides several proactive health check and recovery mechanisms to improve the integrity of the Cisco UCS Manager database. These mechanisms enable active monitoring of the database health.

Changing Internal Backup Interval

You can change the interval at which the internal backup is done. To disable the backup the value can be set to 0.

Procedure

  Command or Action Purpose

Step 1

UCS-A# scope system

Enters the system.

Step 2

UCS-A /system# set mgmt-db-check-policy internal-backup-interval days

Specifies the time interval (in days) at which the integrity backup is done.

Step 3

UCS-A /system* # commit-buffer

Commits the transaction.

Example

This example changes the time interval at which the check runs to two days, and commits the transaction.

UCS-A# scope system
UCS-A /system # set mgmt-db-check-policy health-check-interval 2 
UCS-A /system* # commit-buffer
UCS-A /system #

Triggering Health Check

Use the following commands to trigger an immediate full database integrity check.

Procedure

  Command or Action Purpose

Step 1

UCS-A# scope system

Enters the system.

Step 2

UCS-A /system # start-db-check

Triggers health check.

Step 3

UCS-A /system # commit-buffer

Commits the transaction.

Changing Health Check Interval

You can change the interval at which the integrity check runs. To disable the periodic check entirely set the value for to 0.

Procedure

  Command or Action Purpose

Step 1

UCS-A# scope system

Enters the system.

Step 2

UCS-A /system# set mgmt-db-check-policy health-check-interval hours

Specifies the time interval (in hours) at which the integrity check runs.

Step 3

UCS-A /system* # commit-buffer

Commits the transaction.

Example

This example changes the time interval at which the check runs to two hours , and commits the transaction.


UCS-A# scope system
UCS-A /system # set mgmt-db-check-policy health-check-interval 2 
UCS-A /system* # commit-buffer
UCS-A /system #