Service Availability Status Determined During Scanning¶
During a scan cycle, multiple services are monitored to detect the fault status and availability of each service. An encoder samples a particular service periodically as illustrated by the blue and orange columns in the next figure.
Service faults and equipment and link errors such as video lost, STB fault, encoder stopped, and service stopped determine the service state of the availability computations during sampling. See “Media faults” and “Equipment and link errors” of section “Fault Report Operation” @10.1.2 for details.
In between scans (while the RPM scans other services on the same encoder), the availability of a service is assumed to persist during the whole scan cycle until the beginning of the next scan cycle. Therefore the availability duration for one scan cycle starts at its beginning and ends at the beginning of the next scan cycle for the monitored service. RPM sums up the availability durations of the service for the whole day to determine the service’s daily availability.
Service Availability Computation Example¶
Fig: Service Availability for Sampled States¶
In figure above a service is monitored during scanning (of multiple services) to detect if it is good or faulty. Then the scan results are converted to a service availability computation as follows:
Service availability status begins from the beginning of the first monitored good scan up to the beginning of a first faulty scan as illustrated by the blue timeline above
The “Not Available” service status begins at the beginning of the first faulty scan and continues to the beginning of a first good state as illustrated by the orange timeline above
If a service is free of faults – i.e., “Good” at 12:00 PM and at 12:05 PM but goes faulty at 12:10 PM, then it is considered to be available for two scan cycles: 12:10 – 12:00 => 10 minutes as illustrated in figure above. If the faulty service at 12:10 is followed by other faulty service scans at 12:15 and 12:20, but recovers after 3 service visits at 12:25 PM, then the overall downtime would be 12:25 – 12:10 => 15 minutes.