Home.... | log in.... | Pub my question! | Sign-up....

Java-index >> Solaris Operating System >> Solaris Essentials - General Technical Questions

After Solaris live upgarde disks unavailable

Hi All,

we have two SUN Fire 6800 - cluster node. OS and Kernel Version: Solaris 9 9/05 s9s_u8wos_05 SPARC. After Solaris live upgrade the new root disks are continueously failing. During boot with new boot environment Solaris_9_905 wait and then boot disks become unavailable, cluster will crash and start with old boot environment. If the failed devices are removed then inserted, it makes the device available again.

Could anybody help us? Any idea?

regards

Josef

[497 byte] By [Joesef] at [2007-11-26 11:55:58]

«« SQL server licence error
»» how to total number of threads in system

# 1

I'd need to have more information about why the boot process is failing. What are the console messages, for example?

Having said that, Live Upgrade is not supported with Sun Cluster until version 3.2 (as far as I know) so you may have run into one of the issues that lead to this restriction.

Regards,

Tim

TimRead at 2007-7-7 12:14:57 >

top of Java-index,Solaris Operating System,Solaris Essentials - General Technical Questions...

# 2

Hi Tim,

We have Sun Cluster 3.1. Is it possible error? I'll seek for description of LU.

However: it seems a simple hardware failure. it lost just the disks.

Messages:

root@mcl01:~#metastat -p

d60 -m d61 d62 d63 1

d61 1 1 c1t1d0s6

d62 1 1 c1t5d0s6

d63 1 1 c1t4d0s6

d50 -m d51 d52 d53 1

d51 1 1 c1t1d0s5

d52 1 1 c1t5d0s5

d53 1 1 c1t4d0s5

d30 -m d31 d32 d33 1

d31 1 1 c1t1d0s3

d32 1 1 c1t5d0s3

d33 1 1 c1t4d0s3

d0 -m d1 d2 d3 1

d1 1 1 c1t1d0s0

d2 1 1 c1t5d0s0

d3 1 1 c1t4d0s0

0. c1t0d0

/pci@8,600000/SUNW,qlc@2/fp@0,0/ssd@w21000000870e3de0,0

!!!

1. c1t2d0

/pci@8,600000/SUNW,qlc@2/fp@0,0/ssd@w21000000871421d4,0

2. c1t3d0

/pci@8,600000/SUNW,qlc@2/fp@0,0/ssd@w21000000871432fc,0

!!!

3. c1t5d0

/pci@8,600000/SUNW,qlc@2/fp@0,0/ssd@w21000000871421a1,0

root@mcl02:~#metastat -p

d60 -m d61 d62 d63 1

d61 1 1 c0t1d0s6

d62 1 1 c4t1d0s6

d63 1 1 c4t2d0s6

d50 -m d51 d52 d53 1

d51 1 1 c0t1d0s5

d52 1 1 c4t1d0s5

d53 1 1 c4t2d0s5

d30 -m d31 d32 d33 1

d31 1 1 c0t1d0s3

d32 1 1 c4t1d0s3

d33 1 1 c4t2d0s3

d0 -m d1 d2 d3 1

d1 1 1 c0t1d0s0

d2 1 1 c4t1d0s0

d3 1 1 c4t2d0s0

0. c0t0d0

/ssm@0,0/pci@18,700000/pci@1/scsi@2/sd@0,0

!!! 1. c0t1d0

/ssm@0,0/pci@18,700000/pci@1/scsi@2/sd@1,0

2. c0t2d0

/ssm@0,0/pci@18,700000/pci@1/scsi@2/sd@2,0

297. c4t0d0

/ssm@0,0/pci@1c,700000/pci@1/scsi@2/sd@0,0

!!! 298. c4t1d0

/ssm@0,0/pci@1c,700000/pci@1/scsi@2/sd@1,0

!!! 299. c4t2d0

/ssm@0,0/pci@1c,700000/pci@1/scsi@2/sd@2,0

And

"format" doesn't show some of these disks or says "disk type undefined" (- something like this)

thx

Joesef at 2007-7-7 12:14:57 >

# 3

WARNING - I have not tried. If this is a production cluster I'd try it out on a test cluster first!Try booting the nodes outside cluster first with -rsx. When they come up, run:devfsadm -CvThen try rebooting into cluster mode.Tim

TimRead at 2007-7-7 12:14:57 >

# 4

Ive seen instances where format didnt show disks. And a reboot didnt bring them back.But a full poweroff and on made them visible again.So that might provide a stopgap measure.But I'd consider that a sign of flakey hardware and be looking at replacing.

robertcohen at 2007-7-7 12:14:57 >

Java programming resources: FAQs, tutorials, compiler and browser download sites, documentation, books lists, IDEs, etc.

After Solaris live upgarde disks unavailable

Solaris Operating System New Topic

Solaris Operating System Hot Topic

Solaris Operating System

All Classified