HP StorageWorks Fabric Manager 5.x Administrator Guide (AA-RUQSF-TE, September 2005)

Fabric Manager 5.x administrator guide 81
Acknowledging alerts
When you acknowledge an alert, the alert icons are not displayed in the SAN Elements tab and
At-A-Glance windows in the Overview view. In addition, when you acknowledge an alert, your user name
is displayed in the Acknowledge column of the Alerts View. This lets other Fabric Manager users on the
same server know that you have already acknowledged the alert and are aware of the issue. Once an
alert is acknowledged, it cannot be unacknowledged.
Change Management This alert is generated only if you have elected to receive alerts through the
Change Management feature in Fabric Manager. Any alerts generated on
switches or fabrics as defined in your Change Management profile are also
displayed in the Alerts view. Note the following characteristics:
• If an alert is triggered by a Change Management profile, the alert status is
updated each time a scheduled Change Management check is run. If there
is no change in the alert status, the original alert is updated continuously in
the Last Occurrence column of the Alerts view with the latest unreachable
time until the alert is resolved.
• If you change the baseline Change Management profile that is configured
to trigger alerts, any alerts previously triggered by that defined baseline
profile remain in the Alerts view but the status is changed from No to Yes
in the Resolved column. Also, the reason for the status change is updated
in the Description column to indicate that it is resolved because of a
baseline change.
See Chapter 9, “Change management” for additional information about the
types of trigger alerts you can set using Change Management profiles.
Physical Inventory This alert is generated when the data collection for the Physical Inventory
report is completed for each fabric. Also generated if data collection fails.
FDMI This alert is generated when the FDMI data cannot be collected. If FDMI data
is collected successfully after failing the previous time, an alert is also
generated.
There must be at least one FDMI-capable switch in the fabric for FDMI
information to be collected and for FDMI failure alerts to be generated.
Performance Monitoring Two types of alerts are generated by Performance Monitoring: data collection
(by the five-minute thread), and end-to-end monitor validation (by the 24-hour
thread).
The alerts are resolved (by the five-minute and one-hour threads, respectively)
when the error condition that raised the alert is rectified. The alert is logged
the first time the error is encountered and remains until the data is recaptured
successfully or you turn Performance Monitoring off.
The following scenarios generate an alert:
• Resource limit exceeded
• Advanced Performance Monitoring license removed or otherwise
unavailable on a switch where end-to-end monitors needs to be created
• Combination of resource limit and license missing
• Authentication failure through API
• API reports a fabric access exception, causing an alert to be logged with
the API error code and error message (if available)
• Database exceptions
These alerts are at the fabric level.
Table 13 Alert types (continued)
Type Description