I have Sun 240 server with Solaris 8 installed. I go to OBP and try to boot in single-user mode but it is getting hung in the middle and the only way to get out of this situation is by going to the LOM and resetting the server. Here is what is happening:
{0} ok boot -s
SC Alert: Host System has Reset
SC Alert: CRITICAL ALARM is set
Probing system devices
Probing memory
Probing I/O buses
Netra 240, No Keyboard
Copyright 2006 Sun Microsystems, Inc. All rights reserved.
OpenBoot 4.22.19, 2048 MB memory installed, Serial #75049468.
Ethernet address 0:14:4f:79:29:fc, Host ID: 847929fc.
Rebooting with command: boot -s
Boot device: /pci@1c,600000/scsi@2/disk@0,0 File and args: -s
SunOS Release 5.8 Version Generic_117350-45 64-bit
Copyright 1983-2003 Sun Microsystems, Inc. All rights reserved.
WARNING: forceload of misc/md_trans failed
WARNING: forceload of misc/md_raid failed
WARNING: forceload of misc/md_hotspares failed
WARNING: forceload of misc/md_sp failed
Hardware watchdog enabled
Restoring uplink0 configuration
Linking uplink0 to bge0 as link A
upLink v1.2.5r00
upLink instance 0 ether 00:02:bb:aa:aa:00
upLink instance 1 ether 00:02:bb:aa:aa:01
NOTICE: Recognized a 'bge' interface
Linking uplink0 to ce0 as link B
Restoring uplink1 configuration
Linking uplink1 to bge1 as link A
NOTICE: Recognized a 'bge' interface
Linking uplink1 to ce1 as link B
configuring IPv4 interfaces: uplink0 uplink1.
Hostname: greenwood
Nothing comes up after that. It is then when I go to the LOM (#.) and reset the server. Can anyone help troubleshoot this?
Freezing after hostname is displayed usually happens when a box is configured as part of a domain and it is not reaching the domain controller for whatever reason. If that's the case, you'll have to boot from cd or net, then remove the contents of /etc/defaultdomain.
Set auto-boot to false.Try booting from cdrom to single user, then maye fsck ?
Then reset-all from ok> prompt and boot again.
If still fails, your system nvram could have some "corruption". save a copy of the nvram settings and do a "set-defaluts" at the ok> prompt, reset-all once more than boot up.
I do not find /etc/defaultdomain when in multi-user mode. Are you saying it should be there when I boot from CD? I am using Solaris 8
Tried fsck and fixed all reported errors. Then did "set-defaluts" at the ok> prompt, reset-all once more then boot up in single-user mode but failed again!
Is it a Solaris 10? new setup?
By the way, any changes made to /etc/hosts and /etc/nsswitch.conf files?
whats the patch level of the system? You have have configured for any DNS, remove it and reboot and see again.. looks familiar :rolleyes:
Ok, now I'm a little confused...
You can boot the box to multi-user mode to look at /etc/defaultdomain but not on single user mode? It wouldn't be on the CD, it would be on your hard drive.
what I was saying is that, in my experience of a box booting up and stopping forever after the hostname is displayed in the boot process, the reason why that happened was because the box was part of a NIS domain, which is declared in /etc/defaultdomain, and the server could not see the domain master, therefore it hung forever waiting for the master to respond.
You are right! I can boot to multi-user mode no problem. I am having the problem only when I try to boot to single-user mode using the command "boot -s" as shown on the capture above.
#uname -a
SunOS greenwood 5.8 Generic_117350-45 sun4u sparc SUNW,Netra-240
# cat /etc/release
Solaris 8 2/04 s28s_hw4wos_05a SPARC
Copyright 2004 Sun Microsystems, Inc. All Rights Reserved.
Assembled 08 January 2004
Nothing has changed to /etc/hosts and /etc/nsswitch.conf and no DNS configured.
Well, that's a new one... to get to multi-user mode, single-user has to load...
Out of curiosity, if you are in multi-user, can you do 'init S" and drop to single user, instead of rebooting into single user?
IVe had a similar problem. ITs where the home directory of the root user has been altered and relocated to a drive that is mounted in multiuser mode.
Dont get me wrong, I never configured this rubbish, but I noticed it on a server the other day when I tried to boot -s myself. The previous sa wally had pointed root to /u01/, so single user mode wouldnt work, and went straight to multiuser mode.
Check the root entry in /etc/passwd. Make sure the home directory is / and if the shell has been changed, makes sure its /sbin/sh Ive seen it where ksh has been the choice of shell, but located on a different mount, mounted at multi user mode.
