Fedora Linux Support Community & Resources Center
  #61  
Old 7th December 2008, 10:59 PM
bonzocs Offline
Registered User
 
Join Date: Dec 2008
Location: Cardiff
Posts: 9
Quote:
Originally Posted by cornelius785 View Post
bonzocs,

The only 'progress' I've made on this issue is finding http://forums.fedoraforum.org/showthread.php?t=188630 for the resolvedevice thing. I didn't find it to helpful, maybe I did something wrong with the Super Grub disk since now I only get a grub prompt and I have to load everything by commands. I'm trying to hunt down explanation of the mkinitrd script and making sense of bash scripts (never really made/read one). I have a strong feeling that finding and fixing the resolvedevice will make the computer boot successfully. I looked at /etc/fstab and /etc/fstab.anaconda, and thought I was on to something, but then mkinitrd gave a new error message (error opening /sys/block/*/: no such file or directory) with a lead(??) of http://linux.derkeiler.com/Mailing-L.../msg01861.html, that'll try tomorrow. Let me (us) know if you solve this. Have you found any good leads?

Would you, by chance, be using any sort of LVM? I'm just curious if there could be something with LVM.

The plymouth issue and the 'boot splash' issue are basically cause and effect from what I can tell. From what I've gathered, plymouth is a new fancy boot screen in fedora 10. I've also figured out that rhgb is 'red hat graphical boot', remove this from the kernel arg line along with 'quiet' makes the computer progress further (well, atleast appear to) with lots of messages. I need look up some other arguments so the screen doesn't fly by along with any possible leads.
Hi cornelius785, sorry about the delay in the response, been away for a few days.

Anyway I'm still pretty much in the same place as stated previously. Although following what you suggested in the previous post I'm no longer getting the plymouth error when running mkinitrd. Your correct in suggesting Plymouth is a fancy new boot screen, you can see it here http://www.phoronix.com/scan.php?pag...plymouth&num=3

Anyway the interesting thing is I am running LVM to my knowledge, could this me the link I wonder? The other thing which maybe denying access I was wondering could be LUKS? Are you using LUKS? The reason I ask this is because in my boot prior to the mount error I get:

Quote:
Setting up disk encryption :/dev/mapper/nvidia_cbfdcchgrp2
Command failed: Can not access device
I'm going to have a look into this LVM thing for now, I'll post back if I make any progress.
Reply With Quote
  #62  
Old 8th December 2008, 08:12 AM
saeru Offline
Registered User
 
Join Date: Dec 2008
Posts: 13
bug

After I did /sbin/mkinitrd -f --with=scsi_wait_scan
It's start's booting but I have this error..
it's can't find /dev/

Click image for larger version

Name:	%D0%A4%D0%BE%D1%82%D0%BE042.jpg
Views:	175
Size:	546.5 KB
ID:	17622
Reply With Quote
  #63  
Old 9th December 2008, 12:35 AM
pnelsonsr Offline
Registered User
 
Join Date: Jan 2007
Location: WA
Posts: 171
On HP system with adaptec 29xx scsi I did what was suggested:
1. Boot into F10 rescue mode (didn't load network)
2. Searched and mounted on /mnt/sysimage
3. chroot /mnt/sysimage
4. cd /boot
5. /sbin/mkinitrd -f --with=scsi_wait_scan /boot/initrd-2.6.27.5-117.fc10.i686.img 2.6.27.5-117.fc10.i686
6. reboot

...and all was working! Oh Yeah!
Reply With Quote
  #64  
Old 9th December 2008, 02:37 PM
cornelius785 Offline
Registered User
 
Join Date: Mar 2008
Posts: 17
As far as I can tell, I am not using LUKS. I haven't come across any encryption stuff through my journey of solving this problem yet.
Reply With Quote
  #65  
Old 10th December 2008, 01:54 PM
esilva Offline
Registered User
 
Join Date: Nov 2008
Posts: 12
I just installed the 2.6.27.7-134.fc10.i686 and the bug still exist. This is pretty pathetic. Anyhow, I also instead of booting in rescue mode from the F10 DVD. I boot off the DVD and then off local drive. This way there is no need to do chroot and you pretty much have the real system. I also added a timeout in /etc/grub.conf of 15 seconds. I dunno why they don't put one there...

/sbin/mkinitrd -f --with=scsi_wait_scan /boot/initrd-2.6.27.7-134.fc10.i686.img 2.6.27.7-134.fc10.i686

I from now own will execute that comand after updating to the latest kernel...it looks like f10 team needs to slowdown and look at issues at hand...
Reply With Quote
  #66  
Old 10th December 2008, 08:13 PM
uven Offline
Registered User
 
Join Date: Nov 2008
Posts: 6
Here's what I did to solve this problem:

After installing Fedora 10 I ran into this problem since the SCSI devices didn't get enough time to appear. I booted into rescue mode and then chrooted to the system:

Code:
chroot /mnt/sysimage /bin/bash
I then edited (or in my case, created it since it wasn't there) the file /etc/sysconfig/mkinitrd and added the following:

Code:
MODULES="scsi_wait_scan"
When running mkinitrd it would automatically include this module. I also updated the kernel and that setting seemed to stay.

Code:
mkinitrd -f -v /boot/initrd-`uname -r`.img `uname -r`
Didn't have to run mkinitrd a second time after the kernel update, this seemed to solve it. I had another machine where aacraid didn't get loaded at boot and using this approach also solved that.
Reply With Quote
  #67  
Old 10th December 2008, 08:36 PM
cornelius785 Offline
Registered User
 
Join Date: Mar 2008
Posts: 17
I'm not sure if the problem is in the kernel though, just a hunch. I think the problem is in the mkinitrd script. Just do a search for scsi_wait_scan (or similar), it is clearly set to 'NO', possibly in 2 places if I remember correctly. Basically doing what uven listed, which I also posted previously, will hopefully ensure future initrd files successfully boot.


My feeling is that Fedora 10 should have been delayed, it was brought up on bugzilla long before release and a few people have said Fedora 10 does not meet release criteria since it'll put some machines into an inoperable state. Maybe when I have some time, I can find the developer mailing list to ensure the developers are aware of this issue. This problem seems fairly popular judging on views (4000+ is quite the outlier) and how scsi/raid is not a complete rarity.
Reply With Quote
  #68  
Old 10th December 2008, 09:00 PM
Sauron Offline
Registered User
 
Join Date: Jan 2005
Posts: 258
I figured new kernels would also do this so I too added "scsi_wait_scan 3w_xxxx ext4" into my /etc/sysconfig/mkinitrd file before the latest update. Seems like an easy enough fix, but it is quite the nuisance if you're not prepared for it...
Reply With Quote
  #69  
Old 12th December 2008, 04:07 PM
daja Offline
Registered User
 
Join Date: Dec 2008
Posts: 2
Quote:
Originally Posted by cornelius785 View Post

Loading /lib/kbd/keymaps/i386/qwerty/us.map
could not start boot splash: no such device
Here is how i solved that problem (considering that your installation worked properly; in my case it was a fresh install on an adaptec controller : sata, raid5ee )

- Boot the F10 DVD in rescue mode
- While in the rescue mode, activate your network connection and allow to write on your drive
- Then chroot :
Code:
> chroot /mnt/sysimage
- Update your F10 install :
Code:
> yum update
(do all you have to do here)

- Update the initrd :
Code:
> /sbin/mkinitrd -f --with=scsi_wait_scan /boot/initrd-2.6.27.7-134.fc10.x86_64.img 2.6.27.7-134.fc10.x86_64
(replace with your own version, just do : ls /boot to know)

- Quit the chroot :
Code:
> quit
- Reboot :
Code:
> reboot
- Enjoy

Last edited by daja; 12th December 2008 at 04:15 PM.
Reply With Quote
  #70  
Old 15th December 2008, 03:58 AM
cornelius785 Offline
Registered User
 
Join Date: Mar 2008
Posts: 17
daja,

I might be doing something wrong, but that doesn't work for me. I get the same message during boot and the 'resolvedevice: device spec expected', but no plymouth stuff, when I build the initrd file. It did give me a new direction in trying to solve this. The yum update forced me to mount /proc, which gave me the idea of mounting /dev. Looks like I've got some new combinations to try, it would have been a while before the idea of mount more directories.
Reply With Quote
  #71  
Old 15th December 2008, 09:09 AM
saeru Offline
Registered User
 
Join Date: Dec 2008
Posts: 13
Upgrade to new kernel doesn't help
I did again
/sbin/mkinitrd -f --with=scsi_wait_scan /boot/initrd-2.6.27.7-134.fc10.i686.img 2.6.27.7-134.fc10.i686

It's found partition bug can't mount
/dev
/proc
/sys
/dev/console

Click image for larger version

Name:	%D0%A4%D0%BE%D1%82%D0%BE042.jpg
Views:	73
Size:	546.5 KB
ID:	17628

but kernel from FC9 working ok with this release..
what's wrong with FC10 kernels?
Reply With Quote
  #72  
Old 15th December 2008, 04:33 PM
daja Offline
Registered User
 
Join Date: Dec 2008
Posts: 2
Too bad that my trick doesn't work for you
I don't think that this problem is related to the kernel version; my update has updated about 250 packages ! so it's hard to identify the bad one.

Did you tried to install F10 without LVM ?
Is there any error during installation ?
Reply With Quote
  #73  
Old 15th December 2008, 09:10 PM
cornelius785 Offline
Registered User
 
Join Date: Mar 2008
Posts: 17
saeru,

I'm not sure if this problem is caused by the kernel or by mkinitrd and subsequent files. My guess would be mkinitrd. This is based on that mkinitrd can build a good initrd file for most people when given the appropiate options (like files in /etc/ or -with=scsi_wait_scan) and that in mkinitrd script itself it has 'wait_scsi_scan=NO' or similar in it (based on my memory).

Also, are you running a FC9 kernel (with the appropiate FC9 initrd file) on your FC10 computer? If I'm interpreting what you are saying correctly, could you explain how you did this?
Reply With Quote
  #74  
Old 15th December 2008, 09:27 PM
Insomune Offline
Registered User
 
Join Date: Nov 2008
Location: Gastown, Vancouver, BC
Posts: 42
Quote:
Originally Posted by daja View Post
Too bad that my trick doesn't work for you
I don't think that this problem is related to the kernel version; my update has updated about 250 packages ! so it's hard to identify the bad one.

Did you tried to install F10 without LVM ?
Is there any error during installation ?


I've done the install, w/out the LVM and it works fine, I would try that if you are still having problems Cornelius785...


Otherwise, just for my sake, I've formatted and am trying to set up the LVM working w/ the SCSI card and a new mkinitrd, once I've figured it out I'll post it (if you are jet set on having an LVM, jic you want to add more space)
Reply With Quote
  #75  
Old 17th December 2008, 08:23 AM
saeru Offline
Registered User
 
Join Date: Dec 2008
Posts: 13
I make yum upgrade FC9 to FC10
FC10 working fine with FC9 kernel.. but not with FC10 kernel
without SCSI_WAIT it's not working at all.. like everybody have in this post..
cannot find partition..
but with SCSI_wait it's found partition, but can't mount /dev after this like in screenshot..
I have no idea I make last yum update, with new f10 kernel.. but no result
FC9 kernel working fine..
I have dual xeon
lsmod:
sym53c8xx 68120 2
scsi_transport_spi 22912 1 sym53c8xx
sd_mod 26008 3
scsi_mod 122876 6 sr_mod,sg,libata,sym53c8xx,scsi_transport_spi,sd_m od

grub.conf
#boot=/dev/sda
default=0
timeout=5
splashimage=(hd0,0)/grub/splash.xpm.gz
#hiddenmenu
title Fedora (2.6.27.7-134.fc10.i686)
root (hd0,0)
kernel /vmlinuz-2.6.27.7-134.fc10.i686 ro root=UUID=d5409122-3676-4b53-9679-dbb6a34f65e4 rhgb quiet
initrd /initrd-2.6.27.7-134.fc10.i686.img
title Fedora (2.6.27.5-117.fc10.i686)
root (hd0,0)
kernel /vmlinuz-2.6.27.5-117.fc10.i686 ro root=UUID=d5409122-3676-4b53-9679-dbb6a34f65e4 rhgb quiet
initrd /initrd-2.6.27.5-117.fc10.i686.img
title Fedora (2.6.26.5-45.fc9.i686)
root (hd0,0)
kernel /vmlinuz-2.6.26.5-45.fc9.i686 ro root=/dev/VolGroup00/LogVol00 rhgb quiet
initrd /initrd-2.6.26.5-45.fc9.i686.img

[root@multi ~]# cat /etc/fstab
/dev/VolGroup00/LogVol00 / ext3 defaults 1 1
UUID=d5409122-3676-4b53-9679-dbb6a34f65e4 /boot ext3 defaults 1 2
tmpfs /dev/shm tmpfs defaults 0 0
devpts /dev/pts devpts gid=5,mode=620 0 0
sysfs /sys sysfs defaults 0 0
proc /proc proc defaults 0 0
/dev/VolGroup00/LogVol01 swap swap defaults 0 0
Reply With Quote
Reply

Tags
boot, fedora 10, hangs, scsi

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
aacraid trouble (SCSI hang up?) barakuda Servers & Networking 3 11th January 2010 03:02 PM
Hang: Installing F9 i386 on old AMD-K6 CPU laptop rweed Installation, Upgrades and Live Media 16 3rd September 2008 09:44 PM
Fedora 8 hang on boot Tintinosio Installation, Upgrades and Live Media 3 12th November 2007 09:31 PM
Hang at loading SCSI driver.... BlackBear Installation, Upgrades and Live Media 0 24th September 2005 07:57 AM
GRUB or LILO boot to SCSI drive in SCSI system w/ IDE card mkl838 Installation, Upgrades and Live Media 0 6th March 2005 06:44 AM


Current GMT-time: 19:39 (Wednesday, 23-04-2014)

TopSubscribe to XML RSS for all Threads in all ForumsFedoraForumDotOrg Archive
logo

All trademarks, and forum posts in this site are property of their respective owner(s).
FedoraForum.org is privately owned and is not directly sponsored by the Fedora Project or Red Hat, Inc.

Privacy Policy | Term of Use | Posting Guidelines | Archive | Contact Us | Founding Members

Powered by vBulletin® Copyright ©2000 - 2012, vBulletin Solutions, Inc.

FedoraForum is Powered by RedHat