Topic: Raid problem or sme hacked ?

[davibou]

hi all , this night my server won't respond and when i reboot the server i have this error

Code: [Select]


          --- rd:3 wd:1 fd:2
          disk2 , 0:1, dev:sdc2
         raid5: failed to run raid set md2
         md: pers->run() returned -22
         md :do_mdrun() failed ...
         md: md2 stopped.
         md: unbind<sdc2>
         md: export_rdev(sdc2)
         md: ... autorun DONE.
         Making device-mapper control node
         Scanning logical volumes
             Reading all physical volumes. this may take a while...
             No volume groups found
         Activating logical volumes
             Unable to find volume group "main"
         ERROR: /bin/lvm exited abnormally! (pid 240)
         Creating root device
         Mounting root filesystem
         mount: error 6 mounting ext3
         mount: error 2 mounting none
         Switching to new root
         switchroot: mount failled: 22
         umount /initrd/dev failed: 2
         Kernel panic - not syncing: Attempted to kill init!
         

Can someone help me please.

My config
Sme 7 final whith last updates ; no contribs installed
bi p3 1G slot one
1024 of sdram pc 100 ecc kingston
3 disks scsi  U160 18 gigas seagate in raid5 (raid is configured by sme at install )

I have many data to recorver on the server if someone have the trick for fix this error please tell me .

Regards  David

[byte]

What Kernel are you using to boot up ? it should be one of the SME Sever (depending on cpu) kernel's NOT the Centos kernel's

[davibou]

i boot with SME Server ( 2.6.9-34.ELsmp) kernel.
I have test over kernel boot ( SME and centos ) same error ...
The server worked fine before this night ...
Don't know what apened

[brianr]

Try the non SMP kernel - some motherboards are sensitive to SMP kernels when they are not SMP themselves.

[davibou]

i have try  it same problem ...
I don't know what i can do  ..

[davibou]

Anyone ??

[Gaston94]

Hi David,

[code]
          --- rd:3 wd:1 fd:2
          disk2 , 0:1, dev:sdc2
         raid5: failed to run raid set md2
         md: pers->run() returned -22
         md :do_mdrun() failed ...
         md: md2 stopped.
         md: unbind<sdc2>
         md: export_rdev(sdc2)

looks like you have (at least) a faulty partition : sdc2
The system was not able to rebuilt the raid , it should have be OK we 2 disks from a RAID5 installation, may be there was some other problems earlier you did not noticed.
As Murphy helps, this partition (disk) was part of the main "data" array (/dev/main/root) and if you really loose two disks (partitions) from the raid, things are done
if you loose only one, you might try to boot from a live CD (with raid and LVM support) and look further accessing a degrated array (I won't be of any help for this)

G.

[RvLardin]

 Kernel panic - not syncing: Attempted to kill init!

This line is becoming my nightmare !
We have encounterred it so many times that I'm now loosing my logic.
Last time (2 days ago) on a server on wich, for the same raison, we had broken the Raid before, and was just working on a single disk.
Have a look in the Biugtracker ...

After reading *a lot* it seems (from CentOS forums) that the pb will disapear with the next version of the kernel ...

Soonly, I hope.

RV.

[jonic]

Kernel panic - not syncing: Attempted to kill init!


This line is becoming my nightmare !
We have encounterred it so many times that I'm now loosing my logic.

You give me the creeps, I've just finished converting 4 servers to SME Server from different distributions (slackware, debian, ubuntu). They were running fine, but having a unified platform would make the administration tasks a lot easier, and SME Server seemed like the perfect solution. I wonder now if it was a good decision.

[RvLardin]

Be sure it is.
I've made the same decision about 4 years ago. SME has cross difficulties period during that time but never deceive me. Be confident we will find the trick, the main pb is that it is not possible to provoke this pb so, we just can describe the specific situation when it arrives. But we gain experience and the light will come.

RV.

[duncan]

You give me the creeps, I've just finished converting 4 servers to SME Server from different distributions (slackware, debian, ubuntu). They were running fine, but having a unified platform would make the administration tasks a lot easier, and SME Server seemed like the perfect solution. I wonder now if it was a good decision.

Meh - I have been running these servers from way back on all sorts of hardware in all sorts of configurations with no issues. So have many many others.

[p-jones]

Davibou, did you find a solution. I have the same problem.

My server had been just fine for a long time, I came out one morning, it was down and this is what greeted me.