How to boot Solaris OS temporarily when both root disk mirrors are in maintenance mode ?

 Lets see a small KB/steps to boot Solaris OS system when both root disks are in maintenance mode and is not coming up in multi-user mode even after reboot.  
  
	   
  
	The system is again and again falls into maintenance(single user) mode.  
  
	Both the disks of root partition/mirror are in faulty state.  
  
	   
  
	1. First check the root device and its mirrors. For example:- In our case, d0 is the root mirror with two disks c1t1d0s0 and c1t0d0s0.  
  
	#metastat -p  
  
	d0 -m d20 d10 1  
  
	d20 1 1 c1t1d0s0  
  
	d10 1 1 c1t0d0s0  
  
	   
  
	   
  
	2. Both mirror disks of root are in "needs maintenance".  
  
	-bash-3.00$ metastat d0  
  
	d0: Mirror  
  
	    Submirror 0: d20  
  
	      State: Needs maintenance  
  
	    Submirror 1: d10  
  
	      State: Needs maintenance  
  
	    Pass: 1  
  
	    Read option: roundrobin (default)  
  
	    Write option: parallel (default)  
  
	    Size: 273551232 blocks (130 GB)  
  
	   
  
	d20: Submirror of d0  
  
	    State: Needs maintenance  
  
	    Invoke: metareplace d0 c1t1d0s0 <new device>  
  
	    Size: 273551232 blocks (130 GB)  
  
	    Stripe 0:  
  
	        Device     Start Block  Dbase        State Reloc Hot Spare  
  
	        c1t1d0s0          0     No     Maintenance   Yes  
  
	   
  
	d10: Submirror of d0  
  
	    State: Needs maintenance  
  
	    Invoke: after replacing "Maintenance" components:  
  
	                metareplace d0 c1t0d0s0 <new device>  
  
	    Size: 273551232 blocks (130 GB)  
  
	    Stripe 0:  
  
	        Device     Start Block  Dbase        State Reloc Hot Spare  
  
	        c1t0d0s0          0     No      Last Erred   Yes  
  
	   
  
	Device Relocation Information:  
  
	Device   Reloc  Device ID  
  
	c1t1d0   Yes    id1,sd@n500186200203df40  
  
	c1t0d0   Yes    id1,sd@n5001862003d36b04  
  
	   
  
	   
  
	3. Check /var/log/messages and dmesg, you should see many hard disk bad sector messages, which clearly shows the fault in disks and they need replacement.  
  
	Check other hardware errors with below command.  
  
	#fmadm faulty  
  
	   
  
	4. Ran fsck on both root mirror devices.  
  
	#fsck -F ufs -y /dev/rdsk/c1t0d0s0  
  
	#fsck -F ufs -y /dev/rdsk/c1t1d0s0  
  
	and on metadevice.  
  
	#fsck -F ufs -y /dev/md/dsk/d0  
  
	   
  
	5. Now try to reboot the server, if there are few bad sectors which get repaired in fsck then it will boot otherwise it will again drop into system maintenance mode.  
  
	Now again run fsck on first root mirror.  
  
	#fsck -F ufs -y /dev/rdsk/c1t0d0s0  
  
	Mount the root filesystem rw.   
  
	#mount -o remount,rw /  
  
	   
  
	6. Start essential system services.  
  
	#svcadm enable svc:/system/filesystem/usr:default  
  
	   
  
	7. Now the server and all its services are up, as the partition is in rw mode and services are started. Fire init command to change runlevel to multi-user mode.  
  
	#init 3  
  
	   
  
	The server is now in multi-user mode and will serve temporarily.  
  
	Now avoid rebooting the server again and chase with Oracle to replace the disks asap  
  
	  

Tags: cdsalinuxoopsprogrammingunix