UNEXPECTED INCONSISTENCY; RUN fsck MANUALLY...help

[EricSkoglund]

:Hello,

I'm having a bit of a panic. I was trying to set up a minecraft server on my F16 machine when it froze. It didn't come back after 10 minutes so i hardrebooted it by holdin the power button. When I booted up again i get this message:

_Fedora-16-x86_6: UNEXPECTED INCONSISTENCY; RUN fsck MANUALLY
(ie without -a or -p options)

dracut warning: e2fsck returned with 4
dracut warning: _Fedora-16-x86_6 contains a file systen with errors, check forced
dracut warning: _Fedora-16-x86_6: inodes that were part of a corrupted orphaan linked list found.
dracut warning: *** An error ocurred during the file system check
dracut warning: *** dropping you to a shell; the system will try
dracut warning: *** to mount the filesystem when you leave the shell

dracut warning:

dropping to debug shell

sh:can't access tty; job control turned off
(repair filesystem);/#

when writing exit i get

Welcome to Fedora 16 (Verne)......

and it seems to stop at

[ 124.989806] EXT4-fs error (device dm-1): ext4_mb_generate_buddy:738: group1, 3421 blocks in bitmap, 3418 in gd


I can't imagine that the system will completley fail after a hardreboot and hopefully there is an easy fix and all my files will not be lost.

Any help is REALLY appreciated!

Eric

[DBelton]

Did you do as it specified in the error message, and run fsck manually?

Just run e2fsck on the filesystem that is having a problem, and it will tell you if it can fix the problems or not.

And yes, a hard reboot could really corrupt the files enough that it wouldn't reboot. It all depends on what was happening when you did the hard reboot.

[EricSkoglund]

when I do that on /dev/sda3 which is my linux partion i get the message

recovering journal

and then it says that it is clean
Then i exit the shell and it starts to boot up and it stops at the error message at the end of my first post.

When it froze the only programmes that where going were chrome and the minecraft server

Is there anyway to check the system (if its sda3,4 etc) while in the debugger shell to double check that i'm doing fsck on the right system?

[DBelton]

run it with the -f parameter to force it to check the filesystem.

journalling filesystems are really never marked as being dirty, so you need to force fsck to run and check the entire filesystem.

esfsck -f /dev/sda3

[EricSkoglund]

I did do that and the response is

Pass 1: checking inodes, blocks
...
Pass 5: checking group summary information
/dev/sda3 x/y files (1.3% non-contiguous)

[jpollard]

It looks like you have a "dm-1" based filesystem. This is not /dev/sda3....

Different filesystem.

[EricSkoglund]

hmm okay, i'm what you would call a linux noob.
How do i then know what system to fsck ?

I assumed that it was sda3 because ny windows is on sda1 and then a backup on sda2

---------- Post added at 04:30 PM ---------- Previous post was at 04:09 PM ----------

i did fsck on dm-1 and now when i boot the computer up fedora seems to load fine, but then it goes into

Loading Fedora
Loading initial ramdisk

and says things like

No caching more page present
Assuming drive cache: write through

and now i'm not even taken to a shell...

[DBelton]

Those are normal messages to get.

But what do you mean, not taken to a shell?

Does your system boot completely? or does it hang?

The messages you posted above wouldn't be related to anything causing your system to hang. So if it's hanging, you need to look a little deeper to find what's going on.

[EricSkoglund]

The bootscreen comes up with the loader then the fedora logo. Then it wont start.
I pressed ESC to see what process don't start

the service that dosn't seems to start is LSB

the message is: starts and stops login and scanning of iSCSI devises. failes, see 'systemctl status iscsi.service' for details
then it starts SYSV and the it just gets stuc

[DBelton]

Try editing your kernel line from the grub menu, and remove the parameters "quiet" and "rhgb", and then boot. That should give you a little bit more information bout what is going on.

If you still don't get any information, you can try adding "systemd.log_level=debug" to the kernel line and boot.

The iscsi service failing is probably not an error. It would only be an error if you actually have any iscsi devices. (If you don't know what an iscsi device is, then chances are you don't have any ) When you get your system booted, you can disable the iscsi service if you don't have iscsi devices to keep that message form coming up when you boot (and also to cut down the boot time)

[EricSkoglund]

Now the final message before the cursor starts to frantically blink is

tg3 0000:03:00.0: p7p1: link is down

One thing that i might add is that before it freezes the whole screen goes black and a little load circle shows up in the bottom right corner for less than a second then the services show and then it blinks like this three more times don't know if that is something.

after doing the systemd.log the tg3 thing is gone but it freezes at the same point after Started SYSV

[DBelton]

just guessing.. That is about the point in the boot process where the xserver and GDM is starting.

Since you said the little circle (I am assuming that's the "wait" cursor in Gnome/GDM) comes up, then it's trying to load GDM, and failing to do so.

One thing to try is when it freezes, hit Ctrl-Alt-F2 and see if it gets you to a console screen. If so, log in as root, and see if there is anything in your /var/log/messages file that gives you an indication of what is happening. You can alos see if there are any errors in the /var/log/Xorg.0.log that would keep your Xserver from starting.

[EricSkoglund]

I was able to change the kernel params so that i used just the command line (level 3) .

So i now have command line acces

the messages file don't make much sense to me the problem seems to be in

/usr/lib64/libgdk_pixbuf-2.0.so.0: file to short
is the message

looking inside that file it is empty, i tried to start emacs from the command line and the same message appears maybe this is where the problem is

[DBelton]

looks like at least one file got corrupted sue to your hard reboot

You can try reinstalling that package and see if it gets fixed, though. But no telling what else got corrupted.

From the command line, signed in as root:

yum reinstall gdk-pixbuf gdk-pixbuf2

Edit:
added gdk-pixbuf2 as well.

[EricSkoglund]

crap, problem is that i only have a 3g dongle...... no way to connect via the terminal i think, is there a restore option in fedora as there is in windows?