Nortel DMS-100 Series Maintenance Manual page 416

9-52 Trouble isolation and correction
A single test failure can report one or more diagnostic failures and zero or
more cards that are defective. A diagnostic can run in one unit to report
cards in that unit and the mate unit. When a diagnostic fails, the separate
diagnostic routine sends the failure information to the history database. The
following table is a summary of how each type of counter increases and the
differences necessary on NT-40 platforms.
Counter increments
Counter
Diag
Card
Diag and card
Resets and timestamps
The history database stores five timestamps for every peripheral. These
timestamps are:
297-8231-550 Standard 09.01 August 1998
for the node for the time when
— long term failure counters are last reset
for unit 0 for the time when
— short term failure counters for unit 0 are last reset
— the last diagnostic failure occurred on unit 0
for unit 1
— the time when short term failure counters for unit 1 are last reset
SuperNode, BRISC
Increases short and
long term failure counts
for all diagnostics on the
unit in which the
diagnostic runs.
Increases short and
long term failure counts
for all cards in the unit
or units that contains
the card.
Increases short and
long-term failure counts
for all combinations of
diagnostics and cards.
This counter is for cards
in the same unit in
which the diagnostic
runs.
NT–40
Increases short and
long term failure counts
for diagnostics that the
SWACT controller
requires.
Increases short and
long term failure counts
for cards associated
with diagnostics the
SWACT controller
requires.
The NT-40 does not
store diagnostic and
card combinations
because of data store
limits.