Try /dev/rdsk/c0d0p0, or ...p1, ...p2, ...p4.
Getting "drive type unknown" on the output of format is pretty bad news. That disk is pinin' for the fjords...
I ran fdisk
#fdisk /dev/rdsk/c0d0p0
here is the output of it
Total disk size is 9729 cylinders
Cylinder size is 16065 (512 byte) blocks
Cylinders
Partition Status Type Start End Length %
========= ====== ======== ===== === ====== ===
1 EXT LBA 1 329 329 3
SELECT ONE OF THE FOLLOWING:
1. Create a partition
2. Specify the active partition
3. Delete a partition
4. Change between Solaris and Solaris2 Partition IDs
5. Exit (update disk configuration and exit)
6. Cancel (exit without updating disk configuration)
Enter Selection: 6
#
I believe the disk is healthy , if you have any further checks please give it to me
this is bad news... i don't know if you can rescue your solaris installation... if the fdisk information is missing, solaris can't work with the disk. so i hope you have a backup of your data!
so this is really bad news... i don't know if you can create the fdisk partition new and access the data which should be still on the disk. i've never done something like that before. maybe one of the other guys has ideas about that.
if not, your data is lost and you need a new installation and maybe a new disk because the old might be the bad part in this...
1 Like
@DukeNuke2
Thanks for your efforts
i don't care for the disk the data is the most precious i'm looking for any suggestions from the guys out there
@achenle
i tried to dd c0d0p0 and c0d0p1
#dd if=/dev/rdsk/c0d0p0 of=/dev/null bs=1024k
^C11553+0 records in
11553+0 records out
slices 2,3,4,5,6,7, don't work
---------- Post updated at 03:43 AM ---------- Previous update was at 12:16 AM ----------
does anybody knows any software for solaris x86 partition recovery capable commercial or open source
!!!!
There was some hope for you to recover your data before but I'm afraid you somewhat compromise them with such a questionable command.
You overwrote the first partition by the full disk first 11 MB.
Achenle suggested you to output the dd command to /dev/null. That was to check if the disk was readable. He never suggested to output anything to the disk itself, which is something you should never do when trying to recover data.
In my opinion your best bet now is a data recovery service. As you probably know data recovery can be very expensive. jilliagre, your math is off, 11553 records * 1024k = 11GB rather than 11MB.
Its too bad you have to learn the hard way how important backups are. Having important data with no backup is like owning a home with no insurance. You will rarely file an insurance claim, but you don't want to be the person whose house burned and has no insurance.
Thanks for poiniting my attention to that typo
sorry it was a typing typo , i wasn't capturing the text came from shell
i did'nt do that
acutally i meant
#dd if=/dev/rdsk/c0d0p0 of=/dev/null bs=1024k
#dd if=/dev/rdsk/c0d0p1 of=/dev/null bs=1024k
this acutally what i entered in shell
is there still any hope after all
@ilikecows
i can't affored the cost of recovery service the very right now
i'm looking for a cheap software for recovery
@ilikecows: I soon realized my MB vs GB mistake but was too lazy to correct it as had my laptop turned off. I was sure someone would quickly catch it anyway. Glad to know the OP didn't actually ran that command.
@h: please let's restart from the beginning.
What precise drive model is this ?
How did you partition it at installation time ?
Is this a multi-boot configuration or Solaris only ?
Your partition table is severely bogus. Have you an idea about what was done for this to happen ?
yeah... but would you use something like this to recover lost data on a solaris system? why not use the real thing instead of a (maybe lousy) linux implementation?
1- It's a Western Digital - Caviar 80.0 GB black - IDE interface ( if you want i can give the exact paramteres )
2- Default install options - no modification at install time root file system on ZFS
3- No multiboot - Only Solaris 10
4- I wanted more capacity so i powered off the server and add a new WD HDD IDE sized 160g to same IDE cable with slave jumper setting , then i powered on the server again , wired msg appeared at the bottom of screen highlighted in red
Bad PBR Sig
at first i thought its comming from BIOS then i found that it's from Solaris
then i powered off the server unplugged the second drive without change anything , i powered on again and since that server won't boot anymore that's all
Not only FUSE is in your hand there is a promising project for ZFS on Linux regardless the difference between SUN CDDL and GNU Legislations or even the deep technical details between Linux and Solaris
The point is not to discount something offhandedly, especially when the problem remains ill-defined.
I have no idea what the problem actually is, or how it can be fixed, since we don't know what the issue is, really.
It is not accurate to just completely discount using Linux as a potential "part of the solution", because, you said:
Your post (quoted above) strongly says that you can't mount a ZFS filesystem on Linux... which is not accurate.
---------- Post updated at 15:31 ---------- Previous update was at 15:05 ----------
If you properly powered off your server and simply added a new hard drive to the server; why not simply disconnect the new hard drive and go back to the original configuration and reboot?
---------- Post updated at 15:33 ---------- Previous update was at 15:31 ----------
You might also consider swapping the boot order of the hard disks in BIOS....
After re-reading the posts, it sounds like your original disk may be OK, and your OS is simply trying to boot off the wrong hard disk (the new one you just installed).
@Neo
I already did disconnect the new HDD and powered on ,swaped the boot order of HDD's
with no luck
Can you boot from your install disk?
---------- Post updated at 16:05 ---------- Previous update was at 16:00 ----------
I assume you can either boot from a CD or from another system. Then why not do that and then try mounting the offending filesystem with zfs mount ?
What do you mean by boot from your install disk?
if you mean HDD , the answer is no i can't boot off HDD .
if you mean dvd , the answer is yes i can boot off dvd in single user shell.
i have no solaris x86 servers live right now , only SPARC , but i can install fresh system on alternate disk and attach the dead disk to that system , so what do i need to do zfs mount from either booting off dvd or from within another solaris 10 x86 machine .
what exact steps do i need to do a sucessful zfs mount , deatiled steps if you please
I think you should not get ahead of yourself.
First, you should mount the offending filesystem from a working system and confirm that the filesystem is mountable (working).
Then post back ....
i have no clue how to do this exactly
if you mean to attach the courrpted disk to live system
and issue command like this without creating new pool
#zfs mount somepool
#zfs unmount somepool
the only way i konw is to create a new pool from scratch
with a command like this
#zpool create somepool c0d0
then how can this be done more explainition required please
Do you have another spare disk like that one available, with the same reference (manufacturer, size) ?
That doesn't make sense. Pluging/unpluging a disk cannot cause a mbr to be rewritten anywhere. What else did you or someone else do on that Solaris disk ? Was there another disk plugged in where the second disk was inserted ?
What says
iostat -En
and
fdisk -G /dev/rdsk/c0d0p0