This read-only counter tracks the number of egress cells dropped specifically because the port was in an alarm condition, such as a line failure, rather than because a queue was full. An administrator uses this counter to distinguish alarm-driven traffic loss from congestion-driven loss (egrQfullDiscardCells) when diagnosing why a port's outbound traffic dropped. A sudden jump here that coincides with an ausmPortState change to a failure value confirms the cell loss was caused by the physical or signalling alarm, not oversubscription.