Infrastructure Software

After upgrading from Solaris 11 Express (snv_151a) to Solaris 11 (snv_175), snv_175 fails to boot. Within seconds of loading the kernel, the console displays "NMI received" and the system hangs. Sometimes I get a core-dump, but this is irregular.

The hardware is a Dell Precision T5400, 4 core Xeon (X5450) @ 3.00GHz, 16 GB DDR2.

I have reverted to the old snv_151a boot environment, which runs perfectly fine. As of yet I have not yet had the time to run with the kernel debugger, but I was wondering if anyone else has run into the same problem.


More information available now, after booting the kernel with "-k"

After the NMI received appears, the system seems stalled for about a minute then the following messages appear:

WARNING: /pci@0,0/pci1028,21e@1f,2/disk@0,0 (sd1):
SYNCHRONIZE CACHE command failed (5)


WARNING: /pci@0,0/pci1028,21e@1f,2/disk@0,0 (sd1):
drive offline


WARNING: /pci@0,0/pci1028,21e@1f,2/disk@0,0 (sd1):
drive offline

NOTICE: zfs_parse_bootfs: error 6
Cannot mount root on rpool/251 fstype zfs


panic[cpu0]/thread=fffffffffbc36de0: vfs_mountroot: cannot mount root

Warning - stack not written to the dump buffer
fffffffffbc7fa30 genunix:vfs_mountroot+33a ()
fffffffffbc7fa30 genunix:main+171 ()
fffffffffbc7fa30 unix:_locore_start+90 ()

panic: entering debugger (no dump device, continue to reboot)

Welcome to kmdb
Loaded modules [<list >]

[0]>


(I skipped the kmdb list, since I am typing this from my console)

What's next? Any suggestions on what kmdb commands are most useful now to debug this issue? I will be searching for any possible solutions regarding the vfs_mountroot.



Booting with -asv leads to limited information,

After asking for the name of the system file (I hit enter), the kernel seems to boot, dumps a whole list of x86_features. At the question of the Retire Store, the process hangs, ie, no characters are echoed and the system is unresponsive.



Edited by: user7778697 on Nov 14, 2011 7:01 AM

Edited by: Bones on Nov 14, 2011 7:27 AM