rp8400-core I/O replaced, tape recovery, and panic on boot.

[stephen peng]

Guys,

For hardware reason, I replaced an rp8400's master and slave core I/O cards and upgrade firmware from V5.0 to V6.5, directly. Then I tried to boot the os, but it panicked:

panic: all VFS_MOUNTROOTs failed: NEED DRIVERS ?????

I recovered the os with a backup tape, and there were some warning in the recovery

WARNING: Call to io_insert_node failed
io_instert_node failed: Node already exists or parameter error detected.

And when the box tried to reboot:

: disk(0/0/0/2/0.6.0.0.0.0.0;0)/stand/vmunix

10416128 + 1851392 + 5009712 start 0x1fbee8

WARNING: /stand/ioconfig entries corrupted.

............

Initialization finished successfully... slot is 9
    Swap device table:  (start & size given in 512-byte blocks)
        entry 0 - auto-configured on root device; ignored - can't open device
WARNING: no swap device configured, so dump cannot be defaulted to primary swap.
WARNING: No dump devices are configured.  Dump is disabled.
mountfs:opend

linkstamp:          Fri Dec 9 20:58:23 EAT 2005
_release_version:   @(#)              $Revision: vmunix:    vw: -proj    selectors: CUPI80_BL2000_1108 -c 'Vw for CUPI80_BL2000_1108 build' -- cupi80_bl2000_1108 'CUPI80_BL2000_1108'  Wed Nov  8 19:24:56 PST 2000 $
panic: all VFS_MOUNTROOTs failed: NEED DRIVERS ?????

...............

System Panic:

linkstamp:          Fri Dec 9 20:58:23 EAT 2005

_release_version:   @(#)              $Revision: vmunix:    vw: -proj    selectors: CUPI80_BL2000_1108 -c 'Vw for CUPI80_BL2000_1108 build' -- cupi80_bl2000_1108 'CUPI80_BL2000_1108'  Wed Nov  8 19:24:56 PST 2000 $

panic: all VFS_MOUNTROOTs failed: NEED DRIVERS ?????

PC-Offset Stack Trace (read across, top of stack is 1st):

  0x0020e5d8  0x00306540  0x00358c0c

  0x00470974  0x00358ee8  0x0017731c

  0x001fc4e4

End Of Stack

I could not boot to single-user mode and maintenance mode. So my questions are:

1, what does "all VFS_MOUNTROOTs failed: NEED DRIVERS ?????" mean?

2, would this cause by upgrade the firmware from V5.0 to V6.5 directly.

3, Is there a way to bring the system?

There were two rp8400s with their core I/O cards replace, I would describe the other's issue in the thread.

regards

Stephen

[stephen peng]

Guys,

As discribed in http://h30499.www3.hp.com/t5/System-Administration/rp8400-core-I-O-replaced-tape-recovery-and-panic-on-boot/td-p/5615257, there still other issue with another rp8400.

Master and slave core I/O replaced, and firmware updated from V5.0 to V6.5, then panicker on reboot:

panic: swapconf - no primary swap space configured

And I recovered os from a backup tape and reboot, panicked again:

Boot
: disk(0/0/0/2/0.6.0.0.0.0.0;0)/stand/vmunix

10272768 + 1830912 + 5011944 start 0x1fcb68

WARNING: /stand/ioconfig entries corrupted. (these did not show before the recovery when it panicked)

.....................

linkstamp:          Fri Dec 9 20:10:55 EAT 2005
_release_version:   @(#)         $Revision: vmunix:    vw: -proj    selectors: CUPI80_BL2000_1108 -c 'Vw for CUPI80_BL2000_1108 build' -- cupi80_bl2000_1108 'CUPI80_BL2000_1108'  Wed Nov  8 19:24:56 PST 2000 $
panic: swapconf - no primary swap space configured

PC-Offset Stack Trace (read across, top of stack is 1st):
  0x00237850  0x0049b924  0x0049a2a0
  0x00493f04  0x00382b48  0x001773ac
  0x001fd164
End Of Stack

Like the other one, I could not boot to single-user mode or maintenance mode.

1, "no primary swap space configured" caused the panic? How to fix it, there was swap configuration in the recovery.

2, what is "/stand/ioconfig entries corrupted" standing for?

3, how to bring up the system?

regards

Stephen

[stephen peng]

PS: WARNING: /stand/ioconfig entries corrupted.

the above warning did not show up in the panic before tape recovery.

[Dennis Handly]

>As discribed in ...  there still other issue with another rp8400.

Instead of creating multiple related topics, please continue in your above previous topic.

[Torsten. · Accepted solution]

I don't know why you replaced both MPs, 2 failed MPs are IMHO very unlikely.

However, there is still a problem. The scsi controllers for the internal disks are on these cards, so consider to swap the upper with the lower MP and do a cold install just to test if all works. Looks like your system has a problem to access the disks properly ...

[whiz_kid911]

if you have a mirrored disk in this server, i would try and install OS in one of the bootdisks, and mount the other disk and chek for ioconfig file....

[Ajin_1]

Hi
using setboot or lvlnboot to identify the boot disk in SU mode.
Then activate your root volumegroup (vgchange -a y /dev/vg00) and then run lvlnboot -s /dev/vg03 to configure your swap.

[stephen peng]

Torsten，

The problem was done and I thought I found out what happened.

I recovered the os with another rp8400 with A7109A installed, and the os booted ok. The difference between was scsiU320 driver, with this driver, os could boot after replacing A6193A with A7109A.

So, scsi controllers inside A6193A and A7109A were different？

regards

Stephen