User's Manual

MC-Series System Installation & Testing
Scheduled and Unscheduled Maintenance
90 RadioFrame Networks, Inc.
Alarm Bounce Threshold (counts) Duration Threshold (minutes)
PLL 3 50 12
Tx Underrun 40 12
Tx Overflow 40 12
Rx Overflow 40 12
Slot Mismatch 40 12
CRC Errors 20 12
Packet Size Errors 20 12
Figure 28 RadioBlade fault Bounce and Duration for alarm generation.
In addition to the above-mentioned errors, if the system loses communication with a RadioBlade
or is unable to read the EEPROM from the blade, then that blade is put into an error state and if a
standby blade is available it will switch over automatically.
7.6.1 RadioBlade Policies
The RadioBlade locking policy is the action taken by the system whenever it has determined that
the RadioBlade should no longer be allowed to go active. This could be the result of an alarm,
insertion, removal, or the user specifically locking the RadioBlade.
In general, if a blade is taken out of service and a standby RB is available, then that RB will go
into service. If no standby RB is available then the BR will be locked.
7.6.1.1 Standby Blade
A Standby Blade is an extra RadioBlade that is installed in the system but does not have a
configured BR with which to register. To setup standby blades, make sure that there are more
RadioBlades for each RBS Group than Base Radios configured in the iDEN configuration page.
In all the conditions described below if RBs are present in a RBS that are in a hot standby state,
then locking the RB due to alarms, removal of a RB or administratively locking RBs, will result in
the hot Standby RB to be assigned to the BR that de-registered the locked RB.
7.6.1.2 Locking policy for RB with Errors
If the Radioblade generates enough faults such that it crosses the bounce and duration for that
particular error, an alarm will be generated. The RadioBlade controller then notifies the
associated BR, which then locks the RB.
If we determine there is a faulty RB in the system (i.e., RadioBlades that are generating errors),
we prevent them from coming up after a commanded reset/system reset, but the operator could
manually unlock the RB at which point the RB would be assigned to a BR (Note: this would not
prevent the system from locking it again if an excessive error condition occurs).