| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 5970.1 | | DECWET::VOBA | | Wed Apr 30 1997 08:28 | 7 |
| Re .0, could you tell us more about the Madge adapter itself?
Of course, i'm assuming that you've ensured the symptoms were isolated
to only when the traffic indeed made it over the Madge TR link. What
was the file size when you noticed the problem?
--svb
|
| 5970.2 | | DECWET::VOBA | | Wed Apr 30 1997 10:15 | 4 |
| Re .0, could you tell us what is the buffer count values of the Madge
adapter from the network control panel adapter tab?
--svb
|
| 5970.3 | What should be the buffer count values?? | ZPOVC::LAICHAN | | Thu May 01 1997 09:11 | 21 |
| Hi,
Thanks for the replies.
The file sizes that gave problems are bigger than 12 MB with some of
files bigger than 50 MB.
I do not remember the buffer count of the Madge adapter. I'll check for
on this when I go to the customer site tomorrow (S'pore time). What
should the buffer count values be so I can change the parameters
tomorrow.
We've checked the traffic handled by the Madge adapter to be around
30%.
Appreciate your assistance.
Thanks.
Regards,
Lee
|
| 5970.4 | We are still looking for the basic information | DECWET::VOBA | | Fri May 02 1997 10:04 | 7 |
| Re .0 & .3, could you tell us ...
What model is the Madge adapter? What (PCI/EISA) and which (0/1/2) bus
it was installed on? What version of the Madge driver was configured
for the card?
--svb
|
| 5970.5 | Any more parameters for such a problem? | ZPOVC::LAICHAN | | Sun May 18 1997 20:29 | 16 |
| Hi,
The card is Madge Smart 16/4 PCI Ringnode (BM) 2. We've tried the
various buffer sizes, 4, 6, 8, 10, for testing and also change the frame
size from the default 4096 to 2168. The customer told us that they've
tried frame size 2168 in other Compaq servers and it seems to be working
fine. However, after changing all these parameters we still can get the
problem resolved.
What could possibly be the cause to just a simply file copy or moving
of files be?
Thanks.
Regards,
Lee
|
| 5970.6 | | DECWET::VOBA | | Mon May 19 1997 08:04 | 12 |
| Re .5, i'm sorry it took this long to establish the Madge card in
question to be a PCI BM adapter.
The only cause that i've seen for the behavior(s) the customer has been
experiencing is due to the fact that they may have purchased an early
spin of this adapter model. This spin of the adapter has a PCI bug,
discovered at DECwest and has been fixed since then.
Madge has respun and released a newer version of this adapter. Please
have the customer contact Madge for a replacement unit.
--svb
|
| 5970.7 | Recommended rev? | MSE1::MASTRANGELO | | Fri May 23 1997 07:15 | 3 |
|
What is the new rev of the card that should be recommended to customers
who are also having problems with the Madge card?
|
| 5970.8 | rev 151-100-04 | KOKY::PEREZ | | Mon Jun 02 1997 08:29 | 25 |
| A note that I recieved from David Fieldhouse at MADGE says that the
current version is 151-100-04. This node dates Feb 20, 1997.
.......
FYI, here's a summary of what we're currently recommending:
PCI - board rev. 151-100-04. This is now shipping in bulk, and has
replaced the previous -03 revision.
EISA - standard product (52-08). PCI is recommended in preference to
EISA, where possible.
Driver - mdgmport.sys v2.16 (Alpha)
Regards,
Dave
--------
Dave Fieldhouse
External Test Mgr
Madge Networks Ltd.
Tel. +44 1753 661283
[email protected]
Ozzie Perez
|
| 5970.9 | | DECWET::VOBA | | Mon Jun 02 1997 11:05 | 8 |
| Re .8, Ozzie - the latest one is at higher rev.
Re .7, the best way to tell the difference between an old board and a
fixed board is to look at the Board Number etched into the PCB blank on
the back of the adapter. New fixed cards are 150-101-05 and older
boards are 150-101-02.
--svb
|