K570 warning codes

[bill vandenbosch_1]

I have a K570 that fails boot up. It shows a C5F0 warning on the LCD. On the console display it shows: Booting
boot IO dependent code(IODC)revision 152
Failed I/O operation with Entry_IO
status= -3
It then shows 4lines of registers (I can send those if needed). Then
Error reading IPL. Then it shows the main menu.

[Eugeny Brychkov]

Try booting fro CD. Insert Core OS or Support CD, then get to PDC:
sea (will find all devices)
sea ipl (should list all bootable devices, note CD index here)
bo pX (where X is a CD index)
and see if it will succeed. If yes, then we'll troubleshoot root/boot hard drives, if not then those registers and PIM
BTW, check if 'sea ipl' lists disk you're originally trying to boot from.
C5F0 means 'primary IPL warning'
Eugeny

[Sanjay_6]

Hi,

try this link,

http://forums.itrc.hp.com/cm/QuestionAnswer/1,,0x1e8e84534efbd5118ff40090279cd0f9,00.html

check out paul's attachment.

Hope this helps.

Regds

[bill vandenbosch_1]

Tried booting from Core CD. I get the same type of failure, except it also says "cannot find ENTRY_TEST" after is says "boot IO dependent code"

[Michael Tully]

Try doing a 'SEA'. I suspect that the SCSI controller that runs the internal disks/CD has gone.

[Michael Steele_2]

C7F0 equates to "An error occurred reading IPL".

It really looks like a boot disk problem that will be indicated with the SEA IPL command from the BCH.

GSP>SEA IPL

But your saying you can't even boot to your CD.

The cd device and the interal boot disks will be on different SCSI controllers so the common HW device is going to be your core I/O board.

Was there a "...FLT..." message on the LED panel?

Did all CPU's light up on LED panel?

LED = 0123 verses LED= 023, etc.

Can you get into the service menu please?

interupt boot cycle
ser
pim_info (* check for HPMC and note the timestamp *)
PR (* note any processor's deconfigured *)

If you perform the following where does it fail?

bo alt
interact with IPL = yes
ISL>hpux -is

if this fails then:

ISL>hpux -lm -is

[bill vandenbosch_1]

When a sea is done it does find all devices in the system that should be there. I have installed a new set of hard drives, and tried to reload HPUX. I get an error "error writing VGDA [1] to disk H/W path 10/0.5.0(error=5). It does create some of the file system, then it fails randomly somewhere else in the load.

[Michael Steele_2]

Intermittent errors like this indicate a HW problem and I would open up a HW call with HP post haste. In guessing, a SCSI controller sound flaky.

[Eugeny Brychkov]

Bill,
just a guess... does server has HVD terminator missing at the Core I/O built-in SCSI connector? Remember: this connector on the same bus as internal SCSI disks, so not terminating it you leave whole internal bus unterminated from one end
Eugeny