Specifications

Page 5–14
FOCUS System Manual
the Stream Errors dialog box. Record the
direction, frequency, and type of errors at each
chassis adjacent to the failure location.
We also recommend that you monitor the
error counts at the “down/flipped” chassis
adjacent to the problem. To do this, click on
the “Auto update” box in the Stream Errors dialog
box. With this box checked, FCS queries FOCUS
every five seconds for the real-time error counts.
To determine if the errors are consistently
occurring, you should monitor the numbers for at
least six updates (30 seconds).
Note that after 255 errors, the error counters roll
over and restart at 0. Thus, if you do not monitor
the errors through several updates, you could erro-
neously get the impression that there are a very
low number of errors, when actually they are
constantly rolling over.
Now that you have collected all the available data,
you should be better able to determine the cause of
the failure. Compare the data you have collected
with the information in Table 5-5. The table lists
many of the possible failures, along with the
supporting FOCUS targets that would be logged
and the expected type of errors.
5.1.5.4 Failure Correction
After determining the probable cause of the
failure, it is important to replace or repair the
faulty module or fiber in a way that does NOT
disrupt the current APM state of the loop.
To accomplish this, you must make sure that the
FOCUS chassis in the “down” states are always
able to communicate with the chassis remaining in
the “flipped” states. This is not possible if a
Transceiver or Maintenance module adjacent to
the break is pulled. Therefor, it is always best to
ensure that the fibers are good before attempting
to replace the FOCUS common equipment.
The main areas of caution when correcting the
failure are:
• It is critical that you NOT reset the
Maintenance Module adjacent to the
problem, unless you have determined
that this module has failed. Before
replacing a Maintenance Module, verify
that the chassis on both sides of the bad
Maintenance Module are in the “down”
state and that the directions of failure point
toward the bad Maintenance Module.
Additionally, if you are going to replace the
Maintenance Module, make sure you save
its configuration. It is also a good idea to
save the APM configuration (“flipped”
maps), in case the problem turns out to be a
broken fiber instead of a failed
Maintenance Module.
• The Transceiver Module of a chassis in the
“down” state should NOT be pulled, unless
the chassis on the other side of the break is
also in the “down” state.
• If you reset a Maintenance Module
adjacent to a fiber break, the module will
not automatically arm and will remain in
the “APM ready” state. If this occurs, the
best option is to manually enter the
alternate path map for this situation. You
can do this easily by re-mapping the
channels going out in the direction of the
break to the opposite side of the loop defi-
nition. You can leave pass-through
assignments alone. Remember, though, that
you must map PLD Modules using the next
time-slot. Finally, do NOT attempt to ARM
the loop from this chassis, as it will fail and
also cause all the other FOCUS chassis to
disarm and return to the “APM ready”
state, (with the primary map).
•When you reset a Maintenance Module, it
will always restart in the “APM ready”
state. If the chassis then receives five (5)
seconds of good XID messages from both
sides of the defined loop, it will automati-
cally switch to the armed state.
Note: XID messages are only sent out in
the “armed”, “down” and “flipped” states.
This is the only way a chassis can be armed
automatically. Therefore, it is critical that
two adjacent chassis are NOT reset within
five (5) seconds of each other when the
loop is armed. An XID message is consid-
ered valid if it comes from the expected
chassis as defined in the neighbor’s
address.