| Title: | FDDI - The Next Generation |
| Moderator: | NETCAD::STEFANI |
| Created: | Thu Apr 27 1989 |
| Last Modified: | Thu Jun 05 1997 |
| Last Successful Update: | Fri Jun 06 1997 |
| Number of topics: | 2259 |
| Total number of notes: | 8590 |
Can anyone help me with the following errors I am getting on the FDDI ring
at Pfizer. These have been occuring in the locations noted and have
continued following reboots and complete site resets.
I have not been able to receive any info through the "old" remote
support channels. Needless-to-say, the FDDI backbone and it's integrity
is critical to Pfizer's operation.
Pfizer FDDI Errors 3-24-95
================================================================================
Errors are being logged in the following locations (**)
gvax19 gbr036 gbr033 gvax02 gbr034
| gvax05 | | | |
| | gbr032 | | (**) demfa line 1
demfas| | | | |
2B 3B 1B 1B 2B 2B 4B
| | | | | | |
______ ______ ______ (**)
|GCN001|--1A-IN---2A-OUT|GCN002|--1A-IN---------OUT--2A--|GCN003|--1A--in--+
------ ------ ------ |
| |
2A OUT ______ |
+-----------------------------------------------1A-IN-|GCN004|--2A--out-+
------
|
1B
|
gbr035 (line 1)
Errors:
------
(**)
GCN003: (DECconcentrator 500)
------
PORT 1A (ring in)
Lem Rejects 0-2 per day
Lem Link Errors 225-250 per day
(**)
GVAX02: (9410)
------
DEMFA
MAC error count 5-20 per day
MAC lost count 200-400 per day
Revision info:
-------------
GBR0nn = DECBRIDGE610's: V1.3
GCN0nn = DECconcentrator500's: V3.2.1
GVAX05 = VAX7630, VMS A5.5-2, DEMFA V1.4
GVAX19 = VAX9210, " " " V2.0
GVAX02 = VAX9410, " A5.5-1 " V1.4
| T.R | Title | User | Personal Name | Date | Lines |
|---|---|---|---|---|---|
| 1638.1 | NETCAD::B_CRONIN | Mon Mar 27 1995 13:32 | 25 | ||
It looks like the receiver of GCN003 thinks that the link between it
and CGN004 is noisy. I am concluding that based on the 225-250 LEM
errors per day (these are PHY level errors counted by the PHY level
error counting mechanism known as the Link Error Monmitopr (LEM)).
Between 0 and 2 times per day the PHY automatically disconnects the
link and retests it until the error rate is believed low enough to
reconnect the link. My guess is that each outage lasts about 50
seconds, which is the time it takes for 1 of these tests to run.
Port 2B is the next station downstream from port 1A of GCN003. It is
counting MAC LEVEL errors seen between itself and the next upstream
MAC (GCN004). These errors are probably being caused by the same
error process that is causing the LEM errors on GCN003. The first
opportunity to count them is at GVAX02, which is the MAC immediately
downstream from CGN004's MAC.
So, the problem is probably in the link between the transmitter of
GCN004 and the receiver of GCN003. Could be either of those two
boxes, or the fiber between them. It could also be CGN003 has 2
problems, but that is not as likely, since there don't seem to be any
LEM errors on GVAX02.
| |||||