[j-nsp] J/SRX and ip fragmentation

[Richard A Steenbergen]

On Mon, Jan 25, 2010 at 09:20:31PM -0600, Richard A Steenbergen wrote:
> Oh good lord. I tried to downgrade the SRX from 10.0R2 to 9.6R2 after a 
> few people suggested off-list that there might be some sw bugs, but the 
> SRX never came back from the upgrade:
> 
> FreeBSD/MIPS U-Boot bootstrap loader, Revision 1.6
> (builder at ormonth.juniper.net, Sat Dec 12 15:59:41 UTC 2009)
> Memory: 1024MB
> [1]Booting from nand-flash slice 1
> Un-Protected 1 sectors
> writing to flash...
> Protected 1 sectors
> \
> can't load '/kernel'
> can't load '/kernel.old'
> Press Enter to stop auto bootsequencing and to enter loader prompt.
> 
> 
> Type '?' for a list of commands, 'help' for more detailed help.
> loader> ls
> open '/' failed: no such file or directory

Well I guess I'll go ahead and stick this one in the archives incase
anybody else hits the same problem. Here is what happens when you try to
install 9.6R2 (the website still recommends ver 9.6R1, go figure :P)
from a 10.0R2 box on an SRX210:

da0: <ST ST72682 2.10> Removable Direct Access SCSI-2 device 
da0: 40.000MB/s transfers
da0: 1000MB (2048000 512 byte sectors: 64H 32S/T 1000C)
if_pfe_open: listener socket opened, listening on ...
Trying to mount root from cd9660:/dev/md0
WARNING: preposterous time in file system
WARNING: clock 8243 days greater than file system time
fstab: /etc/fstab:0: No such file or directory
/dev/md1: 50.0MB (102400 sectors) block size 16384, fragment size 2048
        using 4 cylinder groups of 12.52MB, 801 blks, 1664 inodes.
super-block backups (for fsck -b #) at:
 32, 25664, 51296, 76928
******* Working on device /dev/da0 *******
fdisk: Geom not found
fstab: /etc/fstab:0: No such file or directory
/dev/da0s1a: 913.0MB (1869772 sectors) block size 16384, fragment size 2048
        using 5 cylinder groups of 183.62MB, 11752 blks, 23552 inodes.
super-block backups (for fsck -b #) at:
 32, 376096, 752160, 1128224, 1504288

Exception: Cause NMI.
Trapframe Register Dump:
        zero: 00000000  at: fffffffffffffffe    v0: 8001180000001010    v1: 0000
        a0: ffffffff80011800    a1: 00001010    a2: 00000c10    a3: 00000003
        t0: ffffffffc1bfba40    t1: 00000000    t2: 04000009    t3: 00000020
        t4: 00000001    t5: 00060000    t6: ffffffff80060000    t7: 00000004
        t8: ffffffff801180cc    t9: 00000002    s0: 6e656c6e616d6520    s1: 2f6a
        s2: 00000000    s3: 7f454c4601020100    s4: 00000000    s5: 200080000001
        s6: 801000d800000034    s7: 83bd44808b1101      k0: 3480100034  k1: 8010
        gp: ffffffff8092a5a0    sp: ffffffffecdb9a80    s8: 75c4d08085c4d0     8
        sr: 50c8ff84    mullo: 0000148f mulhi: 00000015 badvaddr: ffffffffecde18
        cause: 00000000 pc: ffffffff805f4500
Resetting the  system now...
cpu_reset: Stopping other CPUs
timeout stopping cpus

Which does a wonderful job of wiping out the entire box. The only way to 
recover is to load a new image from the boot loader via console, like 
so:

set ipaddr=x.x.x.x
set netmask=x.x.x.x
set gateway=x.x.x.x
set serverip=x.x.x.x
install tftp://x.x.x.x/newjunosimage.bin

-- 
Richard A Steenbergen <ras at e-gerbil.net>       http://www.e-gerbil.net/ras
GPG Key ID: 0xF8B12CBC (7535 7F59 8204 ED1F CC1C 53AF 4C41 5ECA F8B1 2CBC)