Cisco MDS 9000 Series Configuration Manual page 57

Configuring Interfaces
Note The value of the check interval is common across counters and policies.
• Check interval should be less than the poll interval.
• Check interval is applicable to all the active port monitor policies configured.
• Users should deactivate all the active port monitor policies before enabling, modifying, or disabling the
check interval functionality.
• Check interval cannot be enabled when an active policy is configured.
• Software downgrade to a version that does not support the check interval functionality is restricted when
the check interval functionality is enabled.
• We recommend that you do not have a portguard action set to the state-change counter when an interface
state is changed from down state to up state.
• We recommend that you do not use the default policy when the check interval is configured.
Check Interval
Let us consider a scenario where the poll interval, rising threshold and check interval are configured
with the following values:
• Poll interval is 100 seconds
• Rising threshold is 30
• Check interval is 20 seconds
The check interval starts its interval, C1, along with the poll interval at P1. If an error occurs between
the check intervals C2 and C3, the check intervals C2 and C3 are higher than the configured rising
threshold value of 30, an alert (syslog or trap or both) is generated at C3, alerting the user that an
error has occurred at that particular port.
Note
You can configure longer poll intervals to capture events across poll intervals. For example, configure
a poll interval of 24 hours with a check interval of 30 seconds, with the rising threshold value being
checked cumulatively every 30 seconds.
Guidelines for Configuring Port Monitor Check Interval
Cisco MDS 9000 Series Interfaces Configuration Guide, Release 8.x
39