We have a DS Pro appliance - P4 running v4.2 r3 - when we try to connect to a matrix switch that communicates at 2400 baud the communication fails. The same COM1 port of that DS Pro platform will conenctect successfully to a Cisco switch through a console cable and if we switch the cable over to the matrix switch the Hyperterminal session fails.
Keep in mind that we've been using switches like this (with 2400 baud rate) in conjunction with Sniffers for years, all the way back to the days of DOS DSS. We have not been able to replicate this symptom when we test with PII or PIII platforms. I tried a P4 platform at a customer site that had been reimaged with a NT4.0 sp6a clean image (from about 18 months ago - none of the new MS security patches applied and no OS hardening done). We installed v4.2 r3 and set up the switch - all worked perfectly as usual.
The box that was sent to us for testing and problem assessment was set up originally with v4.3 but we have also found same symptoms when running v4.2 r3.
Some insist that it must be a chipset issue that's casuignt he COM port to fail when it tries to slow down to 2400 baud to talk to the switch. I'm inclined to look at OS hardening as the source of the problem but don't know where to start looking. WE have one custmer who reported this probelm on a lab test box that was a PIII (he thinks) but that box has since ben reimaged and sent elsewhere for other tests - he can't replicate his results. Thus... at present it appears to be limited a few isolated P4 boxes (yet it doesn't occur on all of them as I have already successfully proven).
Hoping some of you understand the serial communication relationship to the hardware and OS and can shed light on this. Thanks!
Owen O'Neill
Datacom Systems Inc.
Northeastern SE
Keep in mind that we've been using switches like this (with 2400 baud rate) in conjunction with Sniffers for years, all the way back to the days of DOS DSS. We have not been able to replicate this symptom when we test with PII or PIII platforms. I tried a P4 platform at a customer site that had been reimaged with a NT4.0 sp6a clean image (from about 18 months ago - none of the new MS security patches applied and no OS hardening done). We installed v4.2 r3 and set up the switch - all worked perfectly as usual.
The box that was sent to us for testing and problem assessment was set up originally with v4.3 but we have also found same symptoms when running v4.2 r3.
Some insist that it must be a chipset issue that's casuignt he COM port to fail when it tries to slow down to 2400 baud to talk to the switch. I'm inclined to look at OS hardening as the source of the problem but don't know where to start looking. WE have one custmer who reported this probelm on a lab test box that was a PIII (he thinks) but that box has since ben reimaged and sent elsewhere for other tests - he can't replicate his results. Thus... at present it appears to be limited a few isolated P4 boxes (yet it doesn't occur on all of them as I have already successfully proven).
Hoping some of you understand the serial communication relationship to the hardware and OS and can shed light on this. Thanks!
Owen O'Neill
Datacom Systems Inc.
Northeastern SE