[c-nsp] strange npe-225 and 7206 non-vxr crashes

[Chris Cappuccio]

The OS on your bootflash cannot handle the NPE-225, which has some
significant differences from the NPE-200.  The "Bad CPU ID" message is
from the boot IOS not the ROM AFAIK.  If you have a 4MB bootflash,
then there are no images you can download which will will fit in that
flash and support this NPE (well, that are currently available from CCO.)

You need at least 12.0(9)S or 12.1(1) to support NPE-225 according to
Cisco's web site.  I think I have a boot image of 12.0(16)S that fits
in the 3600000 or so bytes of space on the "4MB" flash.  The current 12.0S 
boot images on CCO are like 3800000-something bytes so they are just a hair
too big for the older I/O controllers.

Ed Ravin [eravin at panix.com] wrote:
> I just purchased an NPE-225 with 256 MB memory from someone on Ebay.
> I yanked the NPE-200 from my existing 7206 router (non-VXR), and fired
> it up with the new board.  The first message after the bootstrap was
> rather inausipicious:
> 
>    Bad CPU ID 000028A0
> 
> But the router kept going, and worked for five minutes.  Then it died
> with this error:
> 
>    Error: (rsvd) instr cache, fields: data, SysAD
>    virtual addr 0x60109CE6, physical addr(21:3) 0x109CE0, vAddr(14:12) 0x1000
>    virtual address corresponds to main:text, cache word 0
> 
>                 Low Data   High Data  Par    Low Data   High Data  Par
>    L1 Data  : 0:0x10A00003 0x00000000 0x03 1:0x0C041217 0x0240202D 0x03
>               2:0x32050021 0x10A00003 0x03 3:0x00000000 0x0C041E39 0x01
> 
>     Invalid CPU type 28
> 
>     Interrupt exception, CPU signal 20, PC = 0x60109CE0
> 
> I researched the "Bad CPU ID" message, and upgraded the bootflash.  Now,
> I get something else upon powerup, just after the bootflash is loaded:
> 
>    *** Emulating mis-aligned load at 0x87ff1adf
>        PC = 0x87ff7820 ... succeeded
> 
>    *** Trap Exception ***
>    PC = 0x87ff7824, Cause = 0x34, Status Reg = 0x3040d003
> 
>    monitor: command "boot" aborted due to exception
> 
> The router then reloads the bootstrap, reloads the bootflash, and successfully
> boots into IOS.  But then the router crashes again, with this info in the
> crashinfo file:
> 
>  Error: (rsvd) instr cache, fields: data, SysAD
>  virtual addr 0x60104881, physical addr(21:3) 0x104880, vAddr(14:12) 0x0000
>  virtual address corresponds to main:text, cache word 0
> 
>               Low Data   High Data  Par    Low Data   High Data  Par
>  L1 Data  : 0:0xAFB10044 0xAFB00040 0x02 1:0xAFA40068 0x8C9E0194 0x03
>             2:0x8FC20068 0x8FC30060 0x01 3:0x8FD20078 0x00021100 0x00
> 
>               Low Data   High Data  Par    Low Data   High Data  Par
>  DRAM Data: 0:0xAFB10044 0xAFB00040 0x02 1:0xAFA40068 0x8C9E0194 0x03
>             2:0x8FC20068 0x8FC30060 0x01 3:0x8FD20078 0x00021100 0x00
> 
>   Interrupt exception, CPU signal 20, PC = 0x60104880
> 
> 
> So what might be going on here?  If it's a hardware error, why does
> it even boot all the way up on the second try?  I see that updating
> the bootflash got rid of the "Invalid CPU Type" error in the crashinfo,
> but there's clearly something else going on...
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/

-- 
"Attacks always get better; they never get worse."
  -- "Old NSA saying"