[lugm.org] scan disk on linux system.

Thu Jun 28 09:21:57 UTC 2012

Hello,

Well they told me it's not possible on OpenVZ

From: discuss-bounces at discuss.lugm.org
[mailto:discuss-bounces at discuss.lugm.org] On Behalf Of selven
Sent: Thursday, June 28, 2012 1:02 PM
To: LUGM Discuss Mailing List
Subject: Re: [lugm.org] scan disk on linux system.

your datacenter should normally be fine in helping you do that... 

On Thu, Jun 28, 2012 at 5:00 AM, selven <pcthegreat at gmail.com> wrote:

Hmm, maybe this is too late, but couldn't you have booted off a bootable
disk and the mounted your partitions and then chrooted to that and then try
to rebuild your initdrd using mkinitrd? 

+selven

On Thu, Jun 28, 2012 at 4:56 AM, Sebastien <david20 at intnet.mu> wrote:

Hello,

Thanks for your input. It was a bad initrd but there no way to boot off the
OS. I had to reload at the end.

Better to use VPS based kvm next time. At least I wll be able to boot on
knoppix or custom recovery when the OS crashed.

Thanks

Seb

-----Original Message-----
From: discuss-bounces at discuss.lugm.org
[mailto:discuss-bounces at discuss.lugm.org] On Behalf Of Keshwarsingh Nadan
Sent: Wednesday, June 27, 2012 8:40 PM
To: 'LUGM Discuss Mailing List'
Subject: Re: [lugm.org] scan disk on linux system.

Hello,

Assuming the virtual hard drive isn't having a problem (and it may very well
be) the issue you're seeing is a bad initrd, the kernel is almost certainly
going to need a new initrd.

You may be able to get into a repair shell using 'rdshell' kernel argument,
but only if this is a newer version of CentOS (6 is when that came in? I
think?) If you can get into rdshell then you can manually mount the LVM and
get a working root filesystem.

Easiest way is if the 'rdshell' kernel option gives you a dracut repair
shell when it fails to mount the LVM. Try another boot and when Grub comes
up edit your kernel options to add that at the end, then proceed with
booting. When the mounting fails, you should drop into a repair shell rather
than a kernel panic. If you get into dracut's repair shell, then your google
fodder is 'manually mount LVM'. If you get the file system mounted
successfully you're golden, as at that point you can build a new initrd to
replace the failing, reboot, and you're done. (Google fodder mkinitrd)

lvm vgscan -v
lvm vgchange -a y
lvm lvs -all
mount /dev/volumegroup/logicalvolume /mountpoint

Second option would be to boot the VM from a LiveCD or DVD image, preferably
one that is very close (in CentOS versions and kernel versions) to your VM
image's OS - then mount the filesystem, chroot to it and rebuild your
initrd. Google fodder for that process is 'bad initrd'.

The last option (and not guaranteed to work first try) if you have another
virtual image of an operable VM. Go back to that operable VM and make a new
initrd there. Then boot the non-operable VM using a LiveCD boot disk, mount
the LVM drive, and copy the new initrd to the VM's /boot directory.

The new initrd must incorporate the kernel modules needed to function on the
non-operable VM, so it isn't as simple as just running mkinitrd on the
operable VM - you first need to make certain all the modules needed for the
virtual hardware are in your non-operable VM's config or you'll run into the
exact same problem you're having now.

Regards,
kn

-----Original Message-----
From: discuss-bounces at discuss.lugm.org
[mailto:discuss-bounces at discuss.lugm.org] On Behalf Of Ajay R Ramjatan
Sent: Wednesday, June 27, 2012 6:14 PM
To: LUGM Discuss Mailing List
Subject: Re: [lugm.org] scan disk on linux system.

Hi there,

Sorry to hear about your distress. If you have an IP-KVM, please use this to
boot with a boot-cd (ask your datacenter for help) and rescue your system.
If not, ask the datacenter to do it for you and give them precise
instructions on what you want done. E.g., do you want a rescue or a
completely new install? Do you want to preserve data on certain devices such
as /home?

Now is the time you will be testing how good your backup mechanism is
assuming something ReallyBad(TM) happened. I wish you can solve your problem
with just a simple reboot and fsck with the help of your datacenter tech
guys.

Let us know how it works out!

On Wed, Jun 27, 2012 at 5:12 PM, Sebastien <david20 at intnet.mu> wrote:
>
> Of course the lazy option is to reload. . .I am just curious on this one.
>
> -----Original Message-----
> From: discuss-bounces at discuss.lugm.org
> [mailto:discuss-bounces at discuss.lugm.org] On Behalf Of Chris Wilson
> Sent: Wednesday, June 27, 2012 4:01 PM
> To: LUGM Discuss Mailing List
> Subject: Re: [lugm.org] scan disk on linux system.
>
> Hi Sebastien,
>
> On Wed, 27 Jun 2012, Sebastien wrote:
>
>> I was working on a remote system installing cpanel and stuff like
>> that. I run scandisk while the file system was mounted.
>>
>> Now the system is not able to boot. Getting kernel panic at boot.
>>
>> I am a bit lazy to reconfigured the server again. I am using Xen and
> centos.
>> Any ideas what can be done?
>
> You're a bit lazy in providing details about the problem too. Maybe if
> you imagine the problem really hard, we might be able to psychically
> determine what it is and fix it for you using telekinesis.
>
> If you're as lazy as you say, then reinstalling will be easier than
> recovering the server, which sounds like it might be seriously hard
> work, depending on how many "errors" you "fixed" in fsck. By the way,
> if you really ran scandisk and not fsck against the filesystem, you
> might as well give up now.
>
> If you're as lazy as you say, you might find that having good backups
> helps you to get away with it.
>
> Cheers, Chris.
> --
> Aptivate | http://www.aptivate.org | Phone: +44 1223 967 838
<tel:%2B44%201223%20967%20838>  Future
> Business, Cam City FC, Milton Rd, Cambridge, CB4 1UY, UK
>
> Aptivate is a not-for-profit company registered in England and Wales
> with company number 04980791.
>
>
> __________________________________________________________
> Linux User Group of Mauritius (LUGM) Discuss mailing list
> Website: http://lugm.org
> Mailing list archive:
> http://discuss.lugm.org/pipermail/discuss_lugm.org/
> Forum: http://lugm.org/forum/
> IRC: #linux.mu on Freenode
>
>
> __________________________________________________________
> Linux User Group of Mauritius (LUGM) Discuss mailing list
> Website: http://lugm.org
> Mailing list archive:
> http://discuss.lugm.org/pipermail/discuss_lugm.org/
> Forum: http://lugm.org/forum/
> IRC: #linux.mu on Freenode

__________________________________________________________
Linux User Group of Mauritius (LUGM) Discuss mailing list
Website: http://lugm.org
Mailing list archive: http://discuss.lugm.org/pipermail/discuss_lugm.org/
Forum: http://lugm.org/forum/
IRC: #linux.mu on Freenode

__________________________________________________________
Linux User Group of Mauritius (LUGM) Discuss mailing list
Website: http://lugm.org
Mailing list archive: http://discuss.lugm.org/pipermail/discuss_lugm.org/
Forum: http://lugm.org/forum/
IRC: #linux.mu on Freenode

-- 

Pirabarlen Cheenaramen | $3|v3n 

L'escalier

mobile: +230 49 24 918 <tel:%2B230%2049%2024%20918> 

email: pcthegreat at gmail.com || god at hackers.mu

contact: http://godifiy.me <http://godifiy.me/> 

/*memory is like prison*/
(user==selven)?free(user):user=malloc(sizeof(brain));

P Save electricity & disk space. Cat this mail to >/dev/null 2>&1 after use.

-- 

Pirabarlen Cheenaramen | $3|v3n 

L'escalier

mobile: +230 49 24 918

email: pcthegreat at gmail.com || god at hackers.mu

contact: http://godifiy.me <http://godifiy.me/> 

/*memory is like prison*/
(user==selven)?free(user):user=malloc(sizeof(brain));

P Save electricity & disk space. Cat this mail to >/dev/null 2>&1 after use.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://discuss.lugm.org/pipermail/discuss_discuss.lugm.org/attachments/20120628/c51b21fb/attachment.html>