| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 1311.1 | ... and here's the diagrams missing from base-note, sorry | JAYJAY::KORNS | | Mon Aug 15 1994 15:24 | 52 |
|
ORIGINAL CONFIGURATION: Using DC500/DB620s
------------------------------------------
.--------------------------------------------.
| |
| .--------------------------------------. |
| | | |
A B A B
| | | |
+-+--+--------+ +-+--+--------+
| DC500 | | DC500 |
+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+
| | | | | | | | | | | |
M M M M M M M M M M M M
| | / |
.-|--------------------------------------' |
| | | |
| | `------------------------------. |
A B A B
| | | |
+-+-+---------+ +-+-+---------+
| DB620 | | DB620 |
+-------------+ +-------------+
PROPOSED CONFIGURATION: Using DH900/DC900MX/DB900MX
---------------------------------------------------
.---------------------------------------------------.
| |
| .-------------------------------. |
| | | |
| | | |
| .-------|-------------------------------|---------|--------.
| | | .----------------------|-. | |
| | | | | | | |
A M M M M B | A M M M M B |
1 2 3 4 5 6 A/S 1 2 3 4 5 6 A/S
| | | | | | | | | | | | | |
+-+---------+-+ +-+-----------+ +-+---------+-+ +-+-----------+
| CONC 1-1 | | BRIDGE 1-1 | | CONC 2-1 | | BRIDGE 2-1 |
+-+---------+-+ +---+---------+ +-+---------+-+ +---+---------+
| | | | | |
n/a +----------+ n/a +----------+
M B/S M B/S
8 8
|----------- DEChub900-1 -------------| |------------ DEChub900-2 -----------|
|
| 1311.2 | | NACAD2::ANIL | | Mon Aug 15 1994 19:04 | 6 |
| The original config should work fine - the fact that two hubs are
involved and the devices are different doesn't change anything.
(In fact, the modified proposal is not correct since it shows the
DB's FDDI ports as A/S and B/S rather than A and B.)
Anil
|
| 1311.3 | Legal and equivalent but not supported (yet) | LEVERS::PARISEAU | Luc Pariseau | Tue Aug 16 1994 10:04 | 5 |
|
Trees on the backplane are not supported yet. But
otherwise your proposed configuration is fine.
Luc
|
| 1311.4 | So what is the correct configuration ? | JAYJAY::KORNS | | Tue Aug 16 1994 13:06 | 66 |
| I labeled the BRIDGE ports A/S and B/S as a way to indicate
they are the A and B ports, which when plugged into an M-port,
would begin operating as S-ports.
It's my understanding of FDDI configuration rules that when
an A or B port is plugged into an M port, it automagically
goes into S mode (as in the orginal configuration). Further,
if both A & B ports are plugged into M-ports, this is dual
homing and the B-port becomes the primarly and the A is
the fall back or secondary. Is this a correct understanding
for FDDI config rules?
Secondly, back to the PROPOSED DH900/DC900MX/DB900MX, I quote
from Karl Pieper's Jan'94 FDDI CONFIG RULES FOR for DECHUB,
top of page 2, last sentence about DECbridge900MX:
"The port-type assignments are also completely
"software-selectable" where any two ports can act
as A or B ports (Dual Attachement Station - DAS)
or either M and S ports in a tree configuration."
It's that statement which lead me to believe the PROPOSED
CONFIGURATION is valid, even though I couldn't find an example
configuration in Karl's memo like the one we're trying to
build.
Sounds like, from .-1 comment, and re-reading Karl's document,
that the current (?) microcode is limiting backplane connections
to DAS only. I think that means I need to use "Stump" building
blocks for both the DC900MC and DB900MX and do our dual homing
connections on the front-panel ports like so ... which uses (wastes?)
an M-port on each concentrator.
PROPOSED CONFIGURATION: Using DH900/DC900MX/DB900MX, DUAL-HOME ON FRONT-PANELS
------------------------------------------------------------------------------
.---------------------------------------------------.
| |
| .-------------------------------. |
| | | |
| | | |
| .-------|-------------------------------|---------|--------.
| | | | | |
| | | .----------------------|-. .-|--------|-.
| | .-|--------|-. | | | | | |
| | | | | | | | | | | |
A M M M M B | | A M M M M B | |
1 2 3 4 5 6 A B 1 2 3 4 5 6 A B
| | | | | | | | | | | | | | | |
+-+---------+-+ +-+-+---------+ +-+---------+-+ +-+-+---------+
| CONC 1-1 | | BRIDGE 1-1 | | CONC 2-1 | | BRIDGE 2-1 |
+-+---------+-+ +-------------+ +-+---------+-+ +-------------+
|----------- DEChub900-1 -------------| |------------ DEChub900-2 -----------|
QUESTIONS:
1) Am I forced to use this configuration with the current microcode?
2) What version of DB900MX/DC900MX microcode will support the
previous (treed/chained) configurations on the backplane. Karl's
memo is now siz months old and speaks of summer release. What
is the current target date?
|
| 1311.5 | dual homing... | NACAD::PARISEAU | Luc Pariseau | Tue Aug 16 1994 14:36 | 12 |
|
First, it is not a good idea to say that an A (or B) port
connected to an M "begin operating as S-ports". They are
NOT S ports. They are A (or B) ports connected in a TREE
configuration (vs a PEER configuration.) (nit pick...)
Karl Pieper's statement is correct but...that is NOT
supported in the current products. I believe that March
is the time frame when trees will be supported in the hub.
Luc
|
| 1311.6 | Thanks | JAYJAY::KORNS | | Tue Aug 16 1994 15:23 | 10 |
| Thanks Luc,
You are no doubt correct on refering to A/B ports plugging into
M ports, somewhere I picked up that habit.
I'm going to recommend the customer do the dual-home connections
out the front for now, with the hope it can be done over the BP
in the future.
Thanks again, Dave
|
| 1311.7 | another dual homing config | MINNY::SCHUMACHER | More than cows & chocolate... | Wed Aug 24 1994 16:49 | 34 |
| I have a slightly modified configuration. I have six DEChub900 with a
DECbridge900 in each one. We want to dual-home the bridges using 2
DECconcentrator900MX in the same hub. The A port of each bridge
connects to an M-Port of CONC-1, while the B-Port of each Bridge
connects to an M-port of CONC-2. Is this a supported config today?
PROPOSED CONFIGURATION: Using DH900/DC900MX/DB900MX, DUAL-HOME ON FRONT-PANELS
------------------------------------------------------------------------------
.-----------------------------------------.
| |
| .---------------------------------------|------------------.
| | | |
| | .----------------------|-. |
| | | | | |
| | | .--------------------|-|------------------.
| | | | | | | |
| | | | | | | |
M M M M M M M M M M M M | | | |
1 2 3 4 5 6 1 2 3 4 5 6 A B A B
| | | | | | | | | | | | | | | |
+-+---------+-+ +-------------+ +-+-+---------+ +-+-+---------+
| CONC 1-1 | | CONC 1-2 | | Bridge 2-1 | | BRIDGE 3-1 |
+-+---------+-+ +-+---------+-+ +-+---------+-+ +-------------+
A B--------A B
| |
+----------------------------+
|----------- DEChub900-1 ----------| |-DEChub900-2 | |-DEChub900-3-|
Thanks for clarification
Peter
|
| 1311.8 | yes...1311.7 is fine | NACAD2::PARISEAU | Luc Pariseau | Wed Aug 24 1994 17:43 | 0 |
| 1311.9 | Another invalid V3 config ??? | BACHUS::VANLOOCK | | Wed Sep 07 1994 10:39 | 34 |
|
Is .7 a valid configuration??
According to Karl Pieper's wonderful 651.0, a completely self
contained config (with all backplane ports of the modules to
the back) will only be supported in DEChub 900 Firmware V4.0
release (of (next?) summer timeframe ...)
Please tell me if I'm wrong: because we just told a customer that
following config is currently (V3.*) NOT supported!!!!
Patrick
.-------------------------------------------.
| |
| |
| |
| |
M M M M M M |
1 2 3 4 5 6 S
| | | | | | |
+-+-+-+-+-+-+-+ +-------------+ +---+---------+
| Concen 900MX| | Bridge 900MX| | DECnis 600 |
+-+---------+-+ +-+---------+-+ +-------------+
A B--------A B
| |
+----------------------------+
|----------- DEChub900 ------------|
|
| 1311.10 | config is OK. | NACAD2::PARISEAU | Luc Pariseau | Wed Sep 07 1994 18:48 | 6 |
|
As and Bs on the back (Trunk) IS SUPPORTED NOW.
Trees (Ss and Ms) are not.
Luc
|