PDA

View Full Version : Is Build 4744 SP2 a card killer?



eracer
01-14-2004, 09:49 AM
Has anyone else experienced this?

I built a VT(2) (with SX-8 option) system with the following computer:

Supermicro 4U SC742S-420 XEON Server case (420W single power supply with redundant cooling w/ PFC)
Tyan S2665UANF motherboard (i7505 chipset)
2 x INTEL XEON 2.66 533 FSB
4 x CORSAIR 512 DDR ECC NON REG
1 x SEAGATE 80 GIG 7200 RPM Barracuda (system drive)
4 x SEAGATE ST373307LC 73GB (RAID-0 on onboard controller)
1x ASUS CDRW-5224A 52X24X52 E-IDE
1 x FLOPPY DRIVE MITSUMI 35W
1 x GAINWARD GF FX5600 256MB DDR
WINDOWS XP PROFESSIONAL

The Toaster card is in the last 32-bit slot.
The system worked fine.

I upgraded the system to VT(3) Build 4679 and it worked flawlessly.

I then upgraded the system to Build 4744 SP2 (downloaded from the Newtek website). On starting VT(3), I got an error message "No video signal present - Potential driver lockup." The DDR's did not work. Neither killing WINRTME nor rebooting changed anything. I uninstalled Build 4744. No help. I uninstalled all Toaster software and reinstalled Build 4679. No help.

Here's the crux of the problem. I took the Toaster board out and put it into another good VT(3) system (running Build 4679) and IT DID NOT WORK. I then took the Toaster card out of the good system and put it into the problem system that was running Build 4679. It worked perfectly.

Newtek sent me a replacement board. I put it into the problem system (running 4679) and everything was back to normal.

Then I installed 4744 and the exact same thing happened. The board died, and after rolling back to Build 4679 my other good board worked fine. Again the dead board would not work on Build 4679.

I am left with the conclusion that installing Build 4744 fried two Toaster boards.

How can this be?

Is there something here I'm missing?

Any help would be greatly appreciated.

Paul Lara
01-14-2004, 10:51 AM
This likely is tied to a hardware configuration in your system, as we have only had 3 or 4 such reports. Not sure of the root cause, though. :(

eracer
01-14-2004, 11:34 AM
How can a hardware configuration be right for Build 4679, and wrong for Build 4744? And how could this "hardware configuration problem" cause a board to stop working only AFTER a software upgrade?

I appreciate the feedback, but it doesn't really make sense...does it?

Both instances of this problem happened after installation from the same CD. Could there be some weird corruption on that download? The problem is, I'm afraid to try again with a different CD.

Max Steel
01-14-2004, 07:13 PM
I have come across this issue before.

I suspect your VT card is not initialising correctly. A similar thing would happen with certain revision Toaster cards, and was reffered to as 'the no video signal error'

With VT[3] a similar error message to what you mention comes up in the VT Visions and flashes in red writing.

For whatever reason.. when this happens... the Toaster drivers or configuration, or whatever seem to get scrambled up, and then somehow freaks out the Toaster card.

I have built up many Toaster systems, and I have seen this problem happen a couple of times with certain revision boards. (The Toaster revison can be found by pulling out the memory module on the Toaster card.. the revision number is underneath)

I suggest you do the following.

Boot up your system from a cold start.
Go to the start button, find the VT program group, and under tools (from memory) run the core updater.
When it's finished.. power down your system.
Wait a minute.. then start back up.
Once Windows has loaded.. check and see if a red light has lit up on Input one of the SX-8.

If you see no light... go to the run command of the START button of Windows and type 'newtekrtme' (without quote marks obviously)
Wait a few seconds then enter the identical command once more.
A Dos prompt window should open up with all sorts of tech speak, but essentially saying it's downloading the firmware for the SX-8.
If successful.. the red light should light up on your SX-8.

Now.. try running VT[3] and see if video has come back.

Let me know how you get on.

..steel..

Jim_C
01-14-2004, 07:44 PM
That is all very excellent ideas from steel.
If I could I would like to add an ! to the....

>>'When it's finished.. power down your system.'

Not only power down but UNPLUG your system for a solid minute.
Its a tough minute, like waiting for the conditioner in your hair to work or your car to warm on cold days, but it is of utmost importance.


The urgency of the situation was passed along to me from Newtek tech help during a reconfig.

Max Steel
01-14-2004, 08:45 PM
Thats a good point Jim...

Some of the SuperMicro PSU's don't have switches.
(mine doesn't) and on many modern mobo's.. power still gets to the motherboard.

I don't know how dire completely severing the power is for Toaster core updating, but I do know pulling the plug is good practice when card installing/removing.

I once saw a chap (sales person).. and he was installing an SDI card into an uncompressed edit system at a TV studio.

The computer was switched off.. but must have still had the power cord plugged in. Well.. somehow he must have bumped the front panel and accidently switched the PC back on, and he hadn't noticed.

So.. there's me walking into the studio.. semi-dark.. and I see this kinda weird electric light show. I think 'smeg.. my eyes have gone stuffed'

I walk up to this chap whom I was supposed to meet, but he must have got impatient waiting for the tech (me) to arrive.

I said 'do you know that computer is on?' and he goes
'Arghhhhhhhhhh' and just about starts crying.

He then grabs his stuff and says 'see ya!'
And I say. Dude.. you have to go tell the clients' and he says 'Your the tech...fix it'

To cut the story short..... I did get the system working again.. but I had to confront the clients and tell them to put something in writing as a statement, and that I'd sign it as a witness to what happened.

An honest mistake that can happen to anyone. (well not me! )(grin) but to just try and pretend it didn't happen.. that sucked.

Obviously I didn't contract my technical services to that reseller for long...
ie) morons!


So... good point Jim... the above is an example of what can happen. heh heh heh

Sorry for going off topic.....

..steel..

Jim_C
01-14-2004, 09:36 PM
Originally posted by Max Steel
I don't know how dire completely severing the power is for Toaster core updating..<snip>


Tech support might have been a little over cautious in the importance level, but it can't hurt..


http://groups.yahoo.com/group/VTNT/message/32658


nice story

jim

dml
01-15-2004, 01:35 PM
I always remove the power cord when working on computers simply to prevent doing something stupid like hiting the power button by mistake, etc. You can never be too careful. Same thing with wearing a wrist strap . . . it is such a temptation not to wear one. But when you get tired after a long day, or night, (and become careless), and say to yourself I don't need to remove the power cord or wear a wrist strap, that is when you make mistakes.

Matt Drabick,
DigiTek Systems
Raleigh, NC

eracer
01-16-2004, 10:34 AM
Thanks to all for the great advice. Good to see that someone else has seen this before. It proves (well...suggests anyway) that I'm not crazy. The next time I come across this problem I'll post a reply, or a new thread to confirm the fix.