F12 & "No Drives Found"

Place to discuss Fedora and/or Red Hat
Post Reply
User avatar
Basher52
guru
guru
Posts: 912
Joined: Wed Oct 22, 2003 5:57 am
Location: .SE

F12 & "No Drives Found"

Post by Basher52 » Fri Apr 09, 2010 4:04 pm

Do anyone know any problems with this?
The gateway I'm upgrading has the same H/W as when I upgraded to F10 and I can't remember any problems at that point, at least this of this kind.

Have they removed some SATA drivers in F12 or...?


Update:
chipset is: ICH5/ICH5R and M/B: MSI 865PE Neo2-PFS
http://eu.msi.com/index.php?func=prodde ... rod_no=569

User avatar
Basher52
guru
guru
Posts: 912
Joined: Wed Oct 22, 2003 5:57 am
Location: .SE

Post by Basher52 » Fri Apr 09, 2010 4:38 pm

I used the kernel parameter 'nodmraid' and that did it :)
I didn't get the install-page that I told you about but probably the "true" one that really asked for what system to upgrade.
and this I really can't understand since the ONE drive is in no RAID system...

thank God/Devil for google :P


(Installing by net-install right now)

User avatar
Basher52
guru
guru
Posts: 912
Joined: Wed Oct 22, 2003 5:57 am
Location: .SE

Post by Basher52 » Fri Apr 09, 2010 6:54 pm

THIS!!! is why I do a partition backup first...

I waited and the screen's power saver came on...
I hit a key and looked that it kept installing...
but waaaay later I hit a key again, and the install had just got to about 50% installing some package for SE-Linux :(
I waited for about 15 minutes but still no progress and the file was just 2M.
I did a RESET and just waited for the text telling me it can't boot.

It booted and it said Fedora 12 and I thought, OK.
I let it boot up and it even got into the "final stage" and I could use SSH to connect to it.
But when doing uname -a it says:
Linux unknown.host 2.6.27.41-170.2.117.fc10.i686 #1 SMP Thu Dec 10 11:00:29 EST 2009 i686 i686 i386 GNU/Linux
so it gotta be a F10 anyways... somehow...

so... wth so I do now? do I reghost it and try again or can I just 'yum update'
Think I'm gonna try the latest thing, updating the "almost" installed F12 over the night if it gets bad, I have a partition backup so I can, hopefully, get F10 back. (I did a backup of both '/boot' and '/' so I hope I can fix it).

I'll get back with info...

User avatar
Void Main
Site Admin
Site Admin
Posts: 5712
Joined: Wed Jan 08, 2003 5:24 am
Location: Tuxville, USA
Contact:

Post by Void Main » Fri Apr 09, 2010 7:10 pm

I have had installs hang occasionally on many versions of Red Hat/Fedora (not on F12 though). It usually hangs on some specific package upgrade and all I do to fix it rather than cancelling the upgrade is to hit CTRL+ALT+F1 (or F2, F3, etc) to get to a virtual console where I can do a "ps auxww | grep rpm" and take note of which RPM is hanging, then killing whatever script/command it's hung up on and it then continues on with the upgrade. If you backed up /boot and / you should have no problem restoring unless you also have /usr on a separate partition that you didn't back up. Worst case I would boot with the rescue disk and chroot to the hard drive and upgrade the rest of the packages manually but I've rarely ever had to do something that drastic. I can't recall ever having an upgrade mess up so badly I couldn't fix/finish the upgrade.

User avatar
Basher52
guru
guru
Posts: 912
Joined: Wed Oct 22, 2003 5:57 am
Location: .SE

Post by Basher52 » Fri Apr 09, 2010 7:38 pm

What you say sure makes me feel good. When dealing with winblows(as I work with that but losing my job after next week) when some stuff goes bad, you get a BSOD and you gotta do it all from the start.

Since I didn't know and did what you said pushing those keys, I'll try a 'yum update' and if it don't seem to work correctly, I'll need your help or reinstall all programs from start.


... to be continued ...

User avatar
Basher52
guru
guru
Posts: 912
Joined: Wed Oct 22, 2003 5:57 am
Location: .SE

Post by Basher52 » Fri Apr 09, 2010 8:55 pm

Just did what I believed would fix it, but as always... no luck :(
it still boots :) ... and I saw that some packages was named for F12, but... :( :(


after all this and a restart;

[basher52@unknown ~]$ uname -a
Linux unknown.host 2.6.27.41-170.2.117.fc10.i686 #1 SMP Thu Dec 10 11:00:29 EST 2009 i686 i686 i386 GNU/Linux


what can I do now without reGhosting?
if anyone have any clue... what so ever...

User avatar
Void Main
Site Admin
Site Admin
Posts: 5712
Joined: Wed Jan 08, 2003 5:24 am
Location: Tuxville, USA
Contact:

Post by Void Main » Sat Apr 10, 2010 6:21 am

If you can't boot the upgrade disk and continue the upgrade I would restore your backup and try the upgrade again (use the DVD if at all possible). If it hangs at the same spot switch to a virtual console and figure out why it is hung (while it is hung) by looking at the process list. Usually it will be hung trying to execute some scriptlet associated with a specific RPM being installed. Take note of which RPM it is and kill the command that it is hanging up on and the upgrade should continue. Then switch back to the graphical console and watch the upgrade continue. Take care of that one package when the upgrade finishes.

User avatar
Basher52
guru
guru
Posts: 912
Joined: Wed Oct 22, 2003 5:57 am
Location: .SE

Post by Basher52 » Sat Apr 10, 2010 12:02 pm

I got some error when laying the partitions back in the ghost program but I tried and it seems like it's back to F10 again.
I'm DLing the DVD and gonna try that.

btw, do you have some knowledge about a program that can do the same? like backup and restoring partitions/disks.
It should be able to handle FAT, NTFS, ext3 and even ext4 since that is what I wanna use from now on.

This version of ghost I used is an old "Solution Suite 2.5" that I used at my old job and that can handle both FAT, NTFS and ext2, ext3 but it seems that it's not fully working for ext file systems :P

User avatar
Basher52
guru
guru
Posts: 912
Joined: Wed Oct 22, 2003 5:57 am
Location: .SE

Post by Basher52 » Sat Apr 10, 2010 2:50 pm

This is what I was afraid of :(
The update to F12 was OK but after reboot it can't start.

This is some of the first errors that comes up:

Code: Select all

dracut: Scanning for dmraid devices
dracut: Found dmraid sets:
dracut: ERROR: sil: wrong # of devices in RAID set "sil_aeajeidfffce" [1/2] on /dev/sda
ERROR: removing inconsistent RAID set "sil_aeajeidfffce"
ERROR: no RAID set found
...


then it seems to try different names like...
(this text if just copied from a google search, I dont have the energy to write it all, but you get the idea. The only thing different is the sil_xxx name)

Code: Select all

dracut: Activating no
dracut: ERROR: sil: wrong # of devices in RAID set "sil_afabacdcaidd" [1/0] on /dev/sdb
dracut: ERROR: either the required RAID set not found or more options required
dracut: no raid sets and with names: "no"
dracut: Activating raid
dracut: ERROR: sil: wrong # of devices in RAID set "sil_afabacdcaidd" [1/0] on /dev/sdb
dracut: ERROR: either the required RAID set not found or more options required
dracut: no raid sets and with names: "raid"
dracut: Activating sets
dracut: ERROR: sil: wrong # of devices in RAID set "sil_afabacdcaidd" [1/0] on /dev/sdb
dracut: ERROR: either the required RAID set not found or more options required
dracut: no raid sets and with names: "sets" 

and the final text is:

Code: Select all

No boot device found

Boot has failed, sleeping forever.


UPDATE:
I found a one search with this error for F12 and I tried the kernel command: rd_NO_DM.
That made only one error to come up and the system booted up.

What is this new dracut? It can't be a system for finding SATA drives with some sort of default drivers, since this worked perfect in F10.
It says something with RAID so it has to do with that, but it has to be able to see that the boot drive ain't no RAID... right?

Getting updates and we'll see what that does...

User avatar
Basher52
guru
guru
Posts: 912
Joined: Wed Oct 22, 2003 5:57 am
Location: .SE

Post by Basher52 » Sun Apr 11, 2010 8:00 am

Updates added but I still need to use the 'rd_NO_DM' command which I now added to the kernel parameters. Cant understand why tho, can you?

I just mkfs.ext4 /dev/sdc and asked me if I wanted to use the entire disk and I said yes.
Later when I moved all data to this new disk, removed the faulty ones and adds the 2 other 1Tb disks into the raid set, how do I expand the disk 'cos that wont just happen right?

User avatar
Void Main
Site Admin
Site Admin
Posts: 5712
Joined: Wed Jan 08, 2003 5:24 am
Location: Tuxville, USA
Contact:

Post by Void Main » Sun Apr 11, 2010 8:49 am

I'm not sure why it would be detecting dmraid unless you had tried to set it up previously and something is left over making the boot process believe you have a dmraid setup. I looked at the kernel patch that added that "rd_NO_DM" flag and all it does on boot when you add that to the boot line is this:

Code: Select all

 
rm /etc/udev/rules.d/61-dmraid*.rules
So, you must have those files (I don't) or one is getting created during boot and I don't know off the top of my head how that would happen. I would have to dig into it a little deeper.

User avatar
Basher52
guru
guru
Posts: 912
Joined: Wed Oct 22, 2003 5:57 am
Location: .SE

Post by Basher52 » Sun Apr 11, 2010 9:03 am

OK, I'll have a look around too.
Also, since I don't use it I should be able to uninstall it right?

And I'm gonna use resize2fs -p /dev/sdc to grow the ext4 since that seems to handle ext4 too

User avatar
Void Main
Site Admin
Site Admin
Posts: 5712
Joined: Wed Jan 08, 2003 5:24 am
Location: Tuxville, USA
Contact:

Post by Void Main » Sun Apr 11, 2010 2:56 pm

Found a page where a guy ran into a similar problem you had:

http://paul.vox.com/library/post/fedora ... oh-my.html

Post Reply