System hung up

[bschaly]

Hello, I have very little experience with computer maintenance, especially Unix but have been charged with maintaining a server running SCO 5.0.5 that stores a legacy database so I begin by apologizing for my ignorance in this area. We have been having issues with tapping into the system through our terminal emulator program. Typically we have the server running but have no user interface hardware connected. Today we tried to log on through the emulator and the connection was never made. I connected a monitor to the server and it displayed the following messages: Open event driver failed. Fatal server error: Check mouse configuration. NOTICE: cha: SCSI bus has been rest ha=0 Attached SCSI peripherals will return to power up state (chaN04) NOTICE: cha: Unexpected SCSI phase ha=0 id=3 lun=0 cmd=12 blk=0 len=36 A SCSI periperal behaved unexpectedly. Use Compaq Diagnostics to confirm a hardware failure (chaN02) %Stp-0 - - - Vendor =HP Product=C1537A

At the end of these messages there was no command prompt and the cursor was rapidly blinking. I reconnected the mouse and keyboard which were not recognized. At some point as I went about researching an explanation (of which getting on this website was a part) the command prompt came back and I was able to shutdown the system I am currently waiting for it to come back up.

Can anyone explain to me what is happening? The server is a Compaq Proliant800 running SCO Unix OpenServer 5.0.5. The terminal emulator program we run is called JSB Multiview 2000 and we use PCs running Windows XP. We used a digiboard to connect dumb terminals but now that the database is legacy those terminals have been eliminated (a collegue disconnected and then reconnected the digi at a time that coincides with the fact that we have begun to have these connection issues - unrelated perhaps.)

Sincere thanks to anyone who has made it to the end of this tome! And eternal gratitude for any information!!

 

[motoslide]

It would be helpful to know which peripheral on the SCSI channel is set for ID=3.
You should probably have the mouse attached when the system is booted. Some BIOS routines would give the mouse an interrupt if it were present, but make that same interrupt available (and dynamically assigned to) another device (network adapter of SCSI controller) if the mouse were missing. This can cause problems with SCO, since the interrupt is configured into the device driver and is not dynamic.
(whew).
The DigiBoard (if it's an Xem) should not be removed or reconnected if the system is powered up. It draws power for the external box from the motherboard. The C/X has externally powered modules, so those can be removed while the system is running.



"Proof that there is intelligent life in Oregon. Well, Life anyway.

 

[bschaly]

motoslide - Thanks for the response. After running hwconfig I believe a tape drive is the peripheral in question because one of the resulting lines was
name=tape vec=- dma=- type=S ha=0 id=3 lun=0 bus=0 ht=cha

I also ran hw and got the line
%tape - - - type=S ha=0 id=3 lun=0 bus=0 ht=cha

While running these commands I also noticed that what I assume are the digiboard entries indicate it is an X(e)m. Since you said it should not be removed/reconnected with the system powered up and we did exactly that - I am wondering what damage we have done and what I need to do to repair it? Is this the source of my issues?

The digiboard is marked as a PC 8em DB25. The hwconfig and hw provided this: 8 PCI X(e)m v7.0.5

Thanks for any and all responses!

 

[motoslide]

The Digiboard 8em shouldn't be attached (or disconnected) while the system is powered up. That's hindsight at this point, though. It's unlikely that it caused a hardware failure, but I suppose that is possible. If you no longer need the DigiBoard, you can disable the ports, delete any printer spooler entries which might have been configured on those ports, and remove the card from the O/S with "mpi". Then yank the card (with power off, of couse).

What are your current symptoms? I'm a bit confused.

You are correct in that the tape drive is apparently configured onto the "cha" adapter as SCSI ID=3. That device reported a problem, but it might not have caused the system to become unresponsive to your MultiView connections. The other error you mention:
"Open event driver failed. Fatal server error: Check mouse configuration"
Might always appear on the screen when the system is booted. It sounds like the X-Windows GUI interface is enabled, but the configuration is incorrect. Again, this shouldn't cause the system to be unresponsive to a telnet session.
If the system is up, you can look in /usr/adm/syslog and /usr/adm/messages for clues. Both of these files can be very large.


"Proof that there is intelligent life in Oregon. Well, Life anyway.