Server disk issue need help

Hello all,

Our Solaris 9? Sun Fire 480R backup server(in another city) is throwing disk errors such as these repeatedly.

WARNING: vxvm:vxio: Subdisk rootdisk-02 block 24037056: Uncorrectable read error
WARNING: vxvm:vxio: Subdisk rootdisk-02 block 7767072: Uncorrectable write error

The server is only accessible via a console port and I don't know if the cmd line is even available... and unfortunately this server was dropped from a service contract.

I'm not the actual 'Unix' admin but it seems that I need to fix this myself and I'm not sure where to begin although it seems that fschk could possibly be the first step?

The server has been rebooted with no change.

I'd really appreciate any direction on this one..as I need to get the Netbackup environment up and running again.

Thanks in advance,
R-

---------- Post updated at 12:25 PM ---------- Previous update was at 12:09 PM ----------

I was able to access the console, but it appears to be hung or asleep?? not sure

When I telnet to the console via putty, the command prompt never comes up...

---------- Post updated at 12:32 PM ---------- Previous update was at 12:25 PM ----------

Apparently the server is not coming back after the reboot, only the banner is showing.

Thanks

Post the output you get from the console after the reboot.

I dont think there is any output at this point. It boots up and gets to the banner screen where it hangs and the prompt never comes up.

Is there an interup?during the boot where I can stop it and try some diagnostic tool or chdsk?

Thanks!

Which prompt? is the "ok" prompt visible?

This appears to be the banner login prompt window, i think maybe the Open Boot? banner screen?

---------- Post updated at 02:57 PM ---------- Previous update was at 02:56 PM ----------

The prompt is not visible at this time.

Thanks!

---------- Post updated at 03:04 PM ---------- Previous update was at 02:57 PM ----------

we tried breaking the banner screen (waking the server up with) shift # (release) .

The system is still unresponsive...

I think we might have to hard boot it and see what happens?

---------- Post updated 04-11-13 at 10:23 AM ---------- Previous update was 04-10-13 at 03:04 PM ----------

The system was power cycled and is back to the banner login screen and frozen...

Have you tried to send a "break" signal from the RSC to the console to drop to the ok prompt? if you can reach the ok prompt check the devalias entries for a mirror disk and try to boot from that disk...

Also, please attach the boot process as you see it from the console.

Some information:
You have Veritas Volume Manager. This is optional software and replaces the LVM shipped with Solaris. A separate license is needed.
Sector errors! One disk is broken. A broken disk cannot be repaired with a file system check (fsck)! But needs to be replaced.
The controller of a broken disk can "hang". This is likely causing the freeze that you encounter.
Some ideas:
A power down (e.g. from the RSC) let the disk stop; then a power up can bring the broken disk to live for some time.
Assuming the disks are mirrored, it will work to pull the defective disk, so the booting must happen from the working disk.

thats what Ive been hoping I could do! thanks!

How do I send the break cmd from a remote console? is it (CTRL something??) CTRL-

the only way I can access the server right now is via a terminal console

---------- Post updated at 01:12 PM ---------- Previous update was at 01:12 PM ----------

you guys rock...

You are connected to the RSC (Remote Control Board).
There are commands like

help
poweron
poweroff
reset
break
console

Obviously the "console" command was run, that connects to the system console (Ok> prompt etc.). At the very beginning it says how to escape from the connection.
Try #. (two key strokes within 1 second).

1 Like

I get no response when I try shift# .

Thanks!
Ryan

---------- Post updated at 01:35 PM ---------- Previous update was at 01:35 PM ----------

Dallas Terminal Server Hosts UNIX 2
------------------oOO-----
|| ||
| | | |
mgmt-unix2 line 15
C
*****************************************************************

  • --- --- *
  • LOCAL MODE *
    *****************************************************************

just enter "break" from the rsc (#.) prompt and go back to console after that. if that worked you should be at the "ok" prompt. set auto-boot to "false" for further troubleshooting.

ok auto-boot? false

A # and a . within one second!
Maybe hit the Enter key before.
The default RSC prompt is sc>
and there is "help" command
Then give "break" command and go back to console with "console" command.

Well whatever it's stuck for right now, I can't break out with shift# . enter

not shift#. just #.

That's not doing anything. We actually tried it yesterday on another server and it did work.

---------- Post updated at 03:52 PM ---------- Previous update was at 03:50 PM ----------

i cant PM until I have more than 10 posts, lol.

If you're not able to get to the OBP level prompt, you'll need physical access to the machine. Shut the machine down completly (remove power cords) and pull the internal (boot) disks. If no boot medium is found, you should drop to the ok prompt.

Looks like I'm making a trip to that data center next week...thanks for the help on this.

Also a serial connection to the server would be nice to capture the bootup of the system... A max diag run will show if the hardware is ok... Keep the keyswitch in the off position when you return power. Get the serial connection to work with the RSC card. Turn the keyswitch to diag and press the power button after connecting to the console from RSC.

1 Like

I will most definitely try to do that.

Thanks!