Cisco Cisco Customer Voice Portal 8.0(1) Release Note
C
HAPTER
3:
A
DMINISTRATION
U
SER
G
UIDE FOR
C
ISCO
U
NIFIED
CVP
VXML
S
ERVER
AND
C
ISCO
U
NIFIED
C
ALL
S
TUDIO
81
Classes VXML
Server.
Table 3-3
VXML Server Metrics
The more information an administrator has, the better he will be in determining the health of the
system. VXML Server provides a significant amount of information on various metrics to allow
the administrator to understand what is going on within the system. Armed with this information,
the administrator will be able to react quickly to situations which could degrade the stability of
the system.
system. VXML Server provides a significant amount of information on various metrics to allow
the administrator to understand what is going on within the system. Armed with this information,
the administrator will be able to react quickly to situations which could degrade the stability of
the system.
The information falls into three categories: aggregate information, information on peaks, and
average information. Aggregate information, such as the total number of calls handled, is helpful
in determining how much work VXML Server has done so far. Peak information, such as the
maximum concurrent calls occurring in the last 10 minutes, is very helpful in understanding how
load is distributed on the system and can help the administrator understand how the volume is
changing. Average information, such as the average HTTP request completion time, helps the
administrator compare current metrics against historical averages.
average information. Aggregate information, such as the total number of calls handled, is helpful
in determining how much work VXML Server has done so far. Peak information, such as the
maximum concurrent calls occurring in the last 10 minutes, is very helpful in understanding how
load is distributed on the system and can help the administrator understand how the volume is
changing. Average information, such as the average HTTP request completion time, helps the
administrator compare current metrics against historical averages.
The metrics maintained by VXML Server is available only through the JMX administration
interface. To view the metrics, navigate to the
interface. To view the metrics, navigate to the
Global/Metrics
MBean. The Operations tab lists
15 separate functions that the administrator can call to obtain very specific information
concerning how the system is running as well as how it has performed in the past. Many of the
functions take a time duration as an input. It will display information of the specified period up
to a maximum of 60 minutes.
concerning how the system is running as well as how it has performed in the past. Many of the
functions take a time duration as an input. It will display information of the specified period up
to a maximum of 60 minutes.
The following list describes each function and the information it returns:
•
totalCallsSinceStart
– Returns the total cumulative number of calls handled by VXML Server
since it launched. This number will continually rise and only resets only when VXML Server
or the Java application server is restarted.
or the Java application server is restarted.
•
maxConcurrentCallsInLast
– Returns the most number of simultaneous callers that occurred
in the last X minutes where X is entered by the administrator (maximum of 60 minutes) and
when the maximum was reached. This is helpful in determining how close the call volume
reached the license limit on simultaneous callers. Knowing when the maximum value is
reached can be very helpful in determining if call volume is rising. For example if the peak
call volume for the last 10 minutes was achieved very close to present time, that would
indicate that call volume is rising.
when the maximum was reached. This is helpful in determining how close the call volume
reached the license limit on simultaneous callers. Knowing when the maximum value is
reached can be very helpful in determining if call volume is rising. For example if the peak
call volume for the last 10 minutes was achieved very close to present time, that would
indicate that call volume is rising.
•
avgConcurrentCallsInLast
– Returns the average number of simultaneous callers encountered
in the last X minutes where X is entered by the administrator. This data is helpful in
determining if a peak was an isolated occurrence or a sign of a trend. For example if the
maximum number of concurrent calls in the last hour was 100 but the average is 10, then
there is less to be alarmed about since the 100 peak did not last long and can be attributed to
determining if a peak was an isolated occurrence or a sign of a trend. For example if the
maximum number of concurrent calls in the last hour was 100 but the average is 10, then
there is less to be alarmed about since the 100 peak did not last long and can be attributed to