ASMCA

My Env:

launching VirtualBox failed since it needs to re-compile driver:

check the logs and see the err:

Double check that Fedora Kernel 2.6.41 not provides this header again.
not like in 2.6.40(also called 3.1 and 3.0 in mainline kernel.)

Still working on a workaround….
an obvious w/r is to rollback to 2.6.40 kernel.

Thu Dec 8 23:19:43 CST 2011
seems it’s a reported issue…
upgrade to 4.1.6 and change scripts:

VirtualBox-4.1-4.1.6_74713_fedora15-1.x86_64

w/r

Following links may also help:
Default Fedora 15 kernel 2.6.41.1-1: asm/amd_iommu.h missing; bad kernel upgrade!
VirtualBox 4.1.6 on Fedora 15 with Kernel 2.6.41.1: module compilation error

I ran into troubles on my Dell box, which hosting Fedora(2.6.41.1-1.fc15.i686 ) and WinXP Dual OS.

My OS partition table looks like the following(After recovery)

 

Actually I dont especially partition for /home or /tmp,  it’s directed to the root partition.
***I think it a not smart design in Fedora using /etc/mtab
***Which may cause duplicated mount.
***If you dont like it ,just change mtab solution.

[Please Dont Copy following Steps, danger! ]

1. umount /dev/sda7
It’s good that /home umounted.

2. re-run the cmd1 , umount several times. with option ‘-l’, which may dismount root partition.

3. os in bad state although running applications keep working, but /dev/sda is no visible now.

4. fdisk or df not work.

5. reboot and try above codes…and finally…. system down.

[Time to Rescue]

1. reboot nodes doesn’t work, grub seems ok .
/boot is not affected, but cannot boot, incidentally turn to debug dracut shell.

2. use cmd fsck.ext4 and e2fsck  seems not have obvious problems.
mount all my Linux partitions one-by-one, found the problem that my root partition lost.

3. seems I find the root cause that partition table is not in good condition.
cannot find /root.
But all other partitions are work fine.

4. After a while I decide to continue.
I’ve a iso image of “Fedora-15-i686-Live-XFCE.iso” on my Windows partition.
It’s prepared for the first installation from iso.

5. so use this iso for a live linux rescue env.
editing the grub command
(my /boot is not affected, so grub still works, even if it fail, I have grub4dos ready. )

6. boot the LiveCD,  mount the disk partitions and check. seems all ok except the root partition.
TestDisk is a well-known free tool that can help repair the partition table and find the lost partitions…..
It’s however fine my LiveOS can connect internet and download TestDisk binary….
Use TestDisk Analyze to find the partition table, write back to Disk…
<Caution: the write back operation is not safe, backup current table if necessary>

7. reboot OS but ‘m still blocked on GRUB page, so I had to turned back to LiveOS as in step5

8. from my notes, noticed that the disk sequence has already changed…
for example, the /dev/sda5 windows partition is now marked as /dev/sda3
/boot which was /dev/sda8 is now /dev/sda5

need to change following files, replace the changed disk id:

for example change (hd0,8)->(hd0,4) on my env. for /boot
change /dev/sda05 mount point in fstab to /dev/sda03
those records in fstab using UUID is not necessary. if the uuid of disk partition is not changed.
<use cmd blkid to verify>

9. since the boot partition id changed and last reboot blocked on the grub page.
need to re-install grub

10. reboot  and OS can correctly restart… all partitions found.
using e2fsck to check and verify.

—————————————————
Fri Dec  9 10:36:22 CST 2011

update:
Another issue after the partition table recover.

My Oracle DB cannot startup

To Fix:

so, umount all bind mount point on tmpfs, and remount with a larger size.

Fedora already upgraded xulrunner related packages(Firefox/thunderbird) to v8.0. but chmsee package not yet catch up.

so need to pull the fresh codes and compile for myself.
https://github.com/jungleji/chmsee/tags
check the following packages installed.
xulrunner-8.0-1.fc15.x86_64
xulrunner-devel-8.0-1.fc15.x86_64

compile the src and write a short script for generated app:

It’s so long time to issue yum-update on my F15 boxes.
found the emerald packages already removed from repo…

[Following links tell everything:]
https://bugzilla.redhat.com/show_bug.cgi?id=739617
http://fedoraforum.org/forum/showthread.php?p=1516387
http://pkgs.fedoraproject.org/gitweb/?p=emerald.git;a=shortlog;h=refs/heads/master

[Workaround to continue running Emerald:]
just go back to 0.9.4~
1.  Download essential packages:
[x86_64]

[i386]

2. Remove current 0.9.5.0 compiz related packages.

3. install the RPMs downloaded above.

4. edit the /etc/yum.conf to block compiz/emerald related updates.

Well now I’ve blocked updates on compiz and libreoffice… is there any more?