Solved: du locked up and crashed hardrive FreeBSD 7.0 | Experts Exchange

asked on

du locked up and crashed hardrive FreeBSD 7.0

I have an ASUS laptop

http://www.monkeyview.net/id/965/fsck/dmesg/PB12001901.vhtml

with three hardrives in various state of software and hardware distress, but I get by. Right now, I am
running

# uname -a
FreeBSD kayve-PC 6.2-RELEASE FreeBSD 6.2-RELEASE #0: Fri Jan 12 11:05:30 UTC 2007 root@dessler.cse.buffalo.edu:/usr/obj/usr/src/sys/SMP i386
# killall moused
# moused -p /dev/ums0

on a hardrive that makes scary sounds when it over heats.

my shell happens to be pointing to

# pwd
/mnt/usr/home/kayve/cs162

which is one of the four partitions (that was actually a boo boo.. I should have made only one partitiion)
of a newer hardrive that is the subject of this question. I have mounted it on /mnt/usr

I was in a frantic session of svn-ing and java-ing and decided to do a du command. it scrolled and scrolled
and then suddenly everything froze. after that, I haven't been really able to boot that hard drive. I did
a vipw and got my user back, and the way it was booting I would get a character based prompt (no
X) but then after logging in and getting my prompt I would type "startx" and I would get KDE. after
rebooting multiple times it seems I can usually get a prompt (although it rebooted between
the username and password once) .

Here is the interesting output of fsck_ffs -y.. I did this instead of just fsck because fsck "couldn't find
the filesystem type" and I did the -y "constitutive yes" because of what you see...

# moused -p /dev/ums0
# fsck_ffs -y /mnt/usr
** /dev/da0s4 (NO WRITE)
** Last Mounted on /mnt/usr
** Phase 1 - Check Blocks and Sizes
INCORRECT BLOCK COUNT I=2779162 (4 should be 0)
CORRECT? no

An interesting thing about the whole du thing is that it seems like it fails in the exact same directory, and
I have done the du command while booting THIS harddrive and it deadlocks in a very similar way but this
disk is still able to reboot after that.

I have yet to try the obvious deletion of the nasty directory.. I'll probably get around to doing that when I
have timeto make a backup. {:}

If that's the solution.. EASY POINTS for someone!

TeRReF🇳🇱

Kayvey, don't tell me you STILL have that 'about to breakdown months ago' HD in your laptop?

rid🇸🇱

Sounds like an unhealthy hard drive, indeed.... You may want to get your important stuff off of it and get a new one.
/RID

ASKER

Here is another snippet, waiting for it to finish:

# mount /dev/da0s4 /mnt/usr
# fsck_ufs -y /mnt/usr
** /dev/da0s4 (NO WRITE)
** Last Mounted on /mnt/usr
** Phase 1 - Check Blocks and Sizes
INCORRECT BLOCK COUNT I=2779162 (4 should be 0)
CORRECT? no

fsck_ufs: cannot alloc 871186332 bytes for inoinfo
#

I am wondering if this error could be related to something in my laptop other than the hard drive. also,
the du fails on that same directory. when I get home I could take a picture of that. the fact that
it is looking for 0.87TBytes reminds me of something that was happening on a THIRD hard drive..
I have some pictures of that online...

http://www.monkeyview.net/id/965/fsck/vistat/p7170051.vhtml
http://www.monkeyview.net/id/965/fsck/vistat/p7170052.vhtml
http://www.monkeyview.net/id/965/fsck/vistat/p7170053.vhtml
http://www.monkeyview.net/id/965/fsck/vistat/p7170054.vhtml
http://www.monkeyview.net/id/965/fsck/vistat/p7170055.vhtml
. . . .
(click on the right arrow if you want.. there are a bunch more.. after
a while the predicted centuries of load time evaporated over just a few
minutes until...)

http://www.monkeyview.net/id/965/fsck/vistat/p7170068.vhtml

weird.. just before it was done, it was predicting 44000 more days.. it
DID eventually finish, if I remember correctly....

ASKER

this is the hard drive I am running now. It is my oldest of the three..

http://www.monkeyview.net/id/965/fsck/inst_world/p5240182.vhtml

it has a broken vista partion and

$ uname -a
FreeBSD 6.2-RELEASE FreeBSD 6.2-RELEASE #0: Fri Jan 12 11:05:30 UTC 2007 root@dessler.cse.buffalo.edu:/usr/obj/usr/src/sys/SMP i386
$

this is my second hard drive:

http://www.monkeyview.net/id/965/fsck/newhd/PC010317.vhtml

that's the one that is running vista. it has a broken FreeBSD on the other partition

I don't know if I have a picture of the newest hard drive, it has 160GB, all devoted to FreeBSD 7.0..
let me search a bit more....

ASKER

here is the newest hard drive, the one that had this du error of which I speak

http://www.monkeyview.net/id/965/fsck/torrent/p5120227.vhtml

here I am swapping

http://www.monkeyview.net/id/965/fsck/torrent/p5120213.vhtml

ASKER

TeRReF:
The "about to breakdown months ago hardrive" I never threw out! right now it runs my only good
freeBSD! I feel like I have correlated it's clunking sounds well with time of use. It gets hot, and starts
clunking. Last night I forgot to hold down the power button and it stayed on in my backpack all the
way home accross the Bay on BART (I guess that takes a total of around an hour to walk down
the hill thru the Berzerkeley campus.. wait for a train. 26 minutes to Civic Center SF, I don't know another
20 minutes on the Haight Street buses.. I got home and it still had that freek weird pixelated gnome shutdown
screen..

http://www.monkeyview.net/id/965/fsck/ubcd/pb220076.vhtml

right now, I am in the SFSU puter lab, with the new 160GB drive with the du error mounted on
/mnt/usr.. I just updated /etc/fstab so maybe I can boot up with it plugged in?
# cat /etc/fstab
# Device Mountpoint FStype Options Dump Pass#
/dev/ad0s2b none swap sw 0 0
/dev/ad0s2a / ufs rw 1 1
/dev/ad0s2e /tmp ufs rw 2 2
/dev/da0s4 /mnt/usr ufs rw 2 2
/dev/ad0s2f /usr ufs rw 2 2
/dev/ad0s2d /var ufs rw 2 2
/dev/acd0 /cdrom cd9660 ro,noauto 0 0
#

right now I am using my NexStar-3 hard drive enclosure to USB mount (/mnt/usr) the 160GB drive
and I can access my files (and run fsck just fine)

http://www.monkeyview.net/id/965/fsck/torrent/p5120215.vhtml

http://www.monkeyview.net/id/965/fsck/torrent/p5120226.vhtml

ASKER

some advice about fixing and/or diagnosing this bad hard drive was given on this question:

https://www.experts-exchange.com/questions/23798042/can't-svn-commit-to-https-repository.html

ASKER

jeeee hoo see fat!

I just had to reboot multiple times.. I finally took the /etc/fstab entry out.. I HOPE that I don't crash
again..

I just wanted to say.. umm. you DO realize I sorta messed up on that hard drive in terms of I
accidently made 4 partitions instead of 4 slices so I have a different partition for swap, / /usr and
what, /var? when it boots it asks me what to boot on, F1-F4 all Freebsd..

anyway..

::phew:: gotta go!

ASKER

I think one of these files is the culprit

# ls -l /mnt/drive
total 476148
-r-xr-xr-x 1 root wheel 27925130 Jun 22 07:13 478866af.flv
-r-xr-xr-x 1 root wheel 232005014 Jun 22 07:33 478e4be9.flv
-r-xr-xr-x 1 root wheel 1418007 Jun 22 06:48 479f5612.flv
-r-xr-xr-x 1 root wheel 23029293 Jun 22 06:54 47a5013f.flv
-r-xr-xr-x 1 root wheel 3888443 Jun 22 07:05 47c3879c.flv
-r-xr-xr-x 1 root wheel 18870536 Jun 22 06:46 47cb02d5.flv
-r-xr-xr-x 1 root wheel 4217050 Jun 22 07:06 47d5a28c.flv
-r-xr-xr-x 1 root wheel 6906826 Jun 22 06:49 481c989d.flv
-r-xr-xr-x 1 root wheel 11711233 Jun 22 06:43 481db000.flv
-r-xr-xr-x 1 root wheel 11711233 Jun 22 06:35 481dbd34.flv
-r-xr-xr-x 1 root wheel 11887764 Jun 22 07:01 481e23b9.flv
-r-xr-xr-x 1 root wheel 768283 Jun 22 06:45 4820ef2b.flv
-r-xr-xr-x 1 root wheel 8018036 Jun 22 06:45 4820f0bd.flv
-r-xr-xr-x 1 root wheel 12155968 Jun 22 06:44 482127ca.flv
-r-xr-xr-x 1 root wheel 12236880 Jun 22 06:32 48223d72.flv
-r-xr-xr-x 1 root wheel 19432579 Jun 22 06:49 4828d3be.flv
-r-xr-xr-x 1 root wheel 6589062 Jun 22 06:27 483a4b5a.flv
-r-xr-xr-x 1 root wheel 68716529 Jun 22 06:59 483d0958.flv
-r-xr-xr-x 1 root wheel 6091887 Jun 22 07:02 484cd937.flv
#

could that 266MByte file be too big for my hardware?

SOLUTION

TeRReF🇳🇱

membership

Create a free account to see this answer

Signing up is free and takes 30 seconds. No credit card required.

ASKER

I think I did that. I came up with a new idea.

you can't see it on this link:

http://www.monkeyview.net/id/965/fsck/torrent/p5120215.vhtml

but the new hard drive is marked "600mA" while an older one I have is
marked "1A"

Did I fry my new hard drive with too many amps? {:(

SOLUTION

TeRReF🇳🇱

membership

Create a free account to see this answer

Signing up is free and takes 30 seconds. No credit card required.

ASKER

I will double check doing your advice when I am logged back into that computer.

ASKER

Okay, you are saying that fsck should work, but I have done it multiple times and it always says this:

fsck_ufs: cannot alloc 871186332 bytes for inoinfo

I tried deleting the directory du froze on, and it just froze on another directory. My FEELING
is that I DID do an fsck on it unmounted at least once, but I will try at least one more time.

My idea is that the amperage incompatibility caused me to ruin my hard drive, but you are
saying that doesn't make sense. When I bought the second hard drive (120GB), the guy
at central computers was all insecure about advising me since I was using FreeBSD I was
supposed to know. The whole experience for me was unpleasant enough I decided I
would rather keep my mouth shut, so the next time when I bought the newest (160GB)
hard drive, the one that needs the fsck for this question, I didn't bother asking anybody,
I just went in there and pointed, "I'll have THAT one," and they sold it to me say no more.

I should have RTFM the library of congress that is computers, I guess.

ASKER CERTIFIED SOLUTION

TeRReF🇳🇱

membership

Create a free account to see this answer

Signing up is free and takes 30 seconds. No credit card required.

rid🇸🇱

"My idea is that the amperage incompatibility caused me to ruin my hard drive, but you are
saying that doesn't make sense."

Right! This doesn't make sense; you can't push Amperes into something that doesn't want it unless you use a too high voltage, which is nonsensical in this situation. Computers have pretty good power management generally.

You may have to realise that you're dealing with HDs that are on the edge and not totally reliable. Why waste your time on this? Copy important data off those thingies, get new drives as necessary and get on with a normal computer usage.
/RID

ASKER

I am not rich like you. I will not buy any more hard drives from central computers.

ASKER

The old clanky hard drive is in the machine now. You are insulting my hardware I bought for $89. Great.
Is this because I was clinging to IDE technology (that's all I have for now).

Sugar momma promises to buy me a new puter for XMas. Do you have recommendations for better
hardware? I want to keep doing this hard drive swap thing. She doesn't want to mess with the
No OS nonsense we did this time. I guess I will get something running vista because I don't think
I want impaired performance with wine running ChessBase. The number of chess positions is
estimated at 10^120. ChessBase might be the reason this hard drive is clunky. I was running
it all night to try to assess a chess position when it started making noises. I don't feel like messing
with the dual boot crap any more. I like carrying around my handy screwdriver. Umm.. long story
short I want to get an extra hard drive and now they have been giving us openSolaris. I feel like
taking a crack at that. I'll be putting that in an additional hard drive the sugar momma is buying.
Also, do they have laptops that have swappable motherboards nowadays? That would be cool.

ASKER

btw, I accidently put the FreeBSD 7.0 160GB IDE hard drive we are diagnosing into the boot position just
before. It allowed me to log me into the character based part (you know, like with the beastie). Startled by
my mistake, I paused, then went ahead and typed "startx." I thought it was going to go up for a while,
but then it rebooted. I hit "4" and went into single user mode. Typed "fsck."

looked exactly like the output of the below "fsck_ufs" command:

# killall moused
m# # moused -p /dev/ums0
# fsck /dev/da0s4
fsck: Could not determine filesystem type
# fsck_ufs /dev/da0s4
** /dev/da0s4
** Last Mounted on /usr
** Phase 1 - Check Blocks and Sizes
INCORRECT BLOCK COUNT I=2779162 (8 should be 0)
CORRECT? [yn] y

fsck_ufs: cannot alloc 871186332 bytes for inoinfo
#

ASKER

err.. correction.. I think it checked /dev/da0s1 first and it was fine.

ASKER

# /usr/src/tools/tools/recoverdisk
zsh: permission denied: /usr/src/tools/tools/recoverdisk
# set -o vi
# ls /usr/src/tools/tools/recoverdisk
Makefile recoverdisk.1 recoverdisk.c
# cd /usr/src/tools/tools/recoverdisk
# make
Warning: Object directory not changed from original /usr/src/tools/tools/recoverdisk
cc -O2 -fno-strict-aliasing -pipe -Wsystem-headers -Werror -Wall -Wno-format-y2k -W -Wno-unused-parameter -Wstrict-prototypes -Wmissing-prototypes -Wpointer-arith -Wreturn-type -Wcast-qual -Wwrite-strings -Wswitch -Wshadow -Wcast-align -Wunused-parameter -Wchar-subscripts -Winline -Wnested-externs -Wredundant-decls -c recoverdisk.c
cc -O2 -fno-strict-aliasing -pipe -Wsystem-headers -Werror -Wall -Wno-format-y2k -W -Wno-unused-parameter -Wstrict-prototypes -Wmissing-prototypes -Wpointer-arith -Wreturn-type -Wcast-qual -Wwrite-strings -Wswitch -Wshadow -Wcast-align -Wunused-parameter -Wchar-subscripts -Winline -Wnested-externs -Wredundant-decls -o recoverdisk recoverdisk.o
gzip -cn recoverdisk.1 > recoverdisk.1.gz
# ls
Makefile recoverdisk.1 recoverdisk.c
recoverdisk recoverdisk.1.gz recoverdisk.o
# ./recoverdisk
usage: recoverdisk [-r worklist] [-w worklist] source-drive [destination]
# ./recoverdisk /dev/da0s4
start size len state done remaining % done
590348288 1048576 134551002112 0 590348288 134551002112 0.0043684

ASKER

cuil

ASKER

Oh jeez. This is going to take quite some time, huh?

ASKER

# date
Thu Oct 9 18:50:16 PDT 2008
#

# ./recoverdisk /dev/da0s4
start size len state done remaining % done
5075107840 1048576 130066242560 0 5075107840 130066242560 0.0375541
# date
Thu Oct 9 18:50:30 PDT 2008
#
# ./recoverdisk /dev/da0s4
start size len state done remaining % done
5438963712 1048576 129702386688 0 5438963712 129702386688 0.0402465

ASKER

I forgot to put the question mark on "Is this because I am clining to IDE technology (the problems I am having with my 'iffy' hard drive)?"

ASKER

clining = clinging

ASKER

# date
Thu Oct 9 19:08:01 PDT 2008
#
usage: recoverdisk [-r worklist] [-w worklist] source-drive [destination]
# ./recoverdisk /dev/da0s4
start size len state done remaining % done
34712059904 1048576 100429290496 0 34712059904 100429290496 0.2568574

ASKER

oh no. I hope this isn't a disaster

===> jdk-1.5.0p3_5 :
Due to licensing restrictions, certain files must be fetched manually.

Please open http://www.sun.com/software/java2/download.html
in a web browser and follow the "Download" link for the
"JDK 5.0". You will be required to log in and register,
but you can create an account on this page. After registration and
accepting the Sun Community Source License, download the
SCSL Source file, jdk-1_5_0-src-scsl.zip and the
SCSL Binaries file, jdk-1_5_0-bin-scsl.zip .

In addition, please download the patchset, bsd-jdk15-patches-3.tar.bz2, from
http://www.eyesbeyond.com/freebsddom/java/jdk15.html.

Please place the downloaded file(s) in /usr/ports/distfiles
and restart the build.

.*** Error code 1

Stop in /usr/ports/java/jdk15.
# pushd /usr/ports/distfiles
/usr/ports/distfiles /usr/ports/java/jdk15
# uname -a
FreeBSD 6.2-RELEASE FreeBSD 6.2-RELEASE #0: Fri Jan 12 11:05:30 UTC 2007 root@dessler.cse.buffalo.edu:/usr/obj/usr/src/sys/SMP i386
#

ASKER

thought about running recoverdisk..

would it be a bad idea to keep my puter in the fridge while i do it?

oh crap.. how am I going to connect the wires.

ASKER

I was going to put it in the freezer for a bit and when I tried to do that I physically unmounted the
hard disk (in the nexstar) {:(

kilt that recoverdisk.. hmm.. maybe I should try the other thing then

ASKER

no such animal ffs2recov {:(

# ls -d *ffs*
scan_ffs
# pwd
/usr/ports/sysutils
#

ASKER

I have a bad feeling about this: {:(
Attempting to fetch from http://sunsite.mff.cuni.cz/pub/fedora/4/i386/os/Fedora/RPMS/.
fetch: http://sunsite.mff.cuni.cz/pub/fedora/4/i386/os/Fedora/RPMS/elfutils-libelf-0.108-1.i386.rpm: Not Found
=> Attempting to fetch from http://sunsite.mff.cuni.cz/pub/fedora/updates/4/i386/.
fetch: http://sunsite.mff.cuni.cz/pub/fedora/updates/4/i386/elfutils-libelf-0.108-1.i386.rpm: Not Found
=> Attempting to fetch from http://sunsite.mff.cuni.cz/pub/fedora/4/SRPMS/.
fetch: http://sunsite.mff.cuni.cz/pub/fedora/4/SRPMS/elfutils-libelf-0.108-1.i386.rpm: Not Found
=> Attempting to fetch from http://sunsite.mff.cuni.cz/pub/fedora/updates/4/SRPMS/.
fetch: http://sunsite.mff.cuni.cz/pub/fedora/updates/4/SRPMS/elfutils-libelf-0.108-1.i386.rpm: Not Found
=> Attempting to fetch from ftp://sunsite.mff.cuni.cz/pub/fedora/4/i386/os/Fedora/RPMS/.
fetch: ftp://sunsite.mff.cuni.cz/pub/fedora/4/i386/os/Fedora/RPMS/elfutils-libelf-0.108-1.i386.rpm: File unavailable (e.g., file not found, no access)
=> Attempting to fetch from ftp://sunsite.mff.cuni.cz/pub/fedora/updates/4/i386/.

ASKER

I think I saw a little progress.. {:P my ole FreeBSD 6.2 clunker is going to start clunking soon

TeRReF🇳🇱

Putting a computer in the freezer?? If you do that the condensation will ruin it for sure!

ASKER

I wasn't going to leave it in for long.. umm.. heh {:}

well.. I pulled the nexStar out of the puter getting up to go to the fridge. Maybe I
would have put it in the fridge. Big computer facilities have really power AC units,
I know..

/usr/src/tools/tools/recoverdisk was running a reaall long time. Could you make out
the output data? The tabular format got a little messed up by my cut and paste operation.
Do those numbers look right for a 160GB HD? Here, let me fix it for you:

# ./recoverdisk /dev/da0s4
start size len state done remaining % done
34712059904 1048576 100429290496 0 34712059904 100429290496 0.2568574

Also this didn't exist:

/usr/ports/sysutils/ffs2recov

I was figuring I would try again, but right now I am running the intermediary (120GB) HD..
let me reacquaint you with my disks:

relative age capacity dual boot? partition 1 partition 2 notes
oldest 60GB yes FreeBSD 6.2 Vista makes the
gnome (operational) (broken) clunking noises

intermediate 120GB yes Vista FreeBSD (?)
(running now) (broken)

newest 160GB no FreeBSD 7.0 n/a subject of this
KDE (operational) question

ASKER

umm.. I made a boo boo I guess. the FreeBSD 7.0 KDE HD is not _really_ operational, of course..
I HAVE been getting a shell but X won't start.

gheist🇧🇪

My old friend kayve!

In your initial post you did irrepairable damage by mounting file systems and fscking it while mounted.
It is up to you to decide if you reinstall or restore from backup. There are no other options.
Do not get mad at me - it is now your fault, not coincidence.

ASKER

I don't believe you. You have been wrongly accusing me of using less than perfect English. I don't understand why you hate me, but you obviously hate me. My professor said I might be able to fix it.

ASKER

Okay maybe you are not lying.

http://en.wikipedia.org/wiki/Fsck

However, I believe you have the skills to fix "severe data corruption/loss," but for unknown reasons
you hate me, just like many other computer scientists aaaalllllll hate each other.

ASKER

I'll just run my clunky hard disk for now and let this question listlessly float away then.

SOLUTION

gheist🇧🇪

membership

Create a free account to see this answer

Signing up is free and takes 30 seconds. No credit card required.

ASKER

Point of info: the clunky hard drive is master. Okay. Thank you very much for advice. I will plug in the
non clunky fsck abused 160GB now to take your advice.

ASKER

clunky old booting harddrive is ad0 right now. Nonclunky but fsck abused drive is da0 right now. That
could chnage if I swap them.

boy this is taking long:

# dd if=/dev/da0 bs=65536 of=/dev/null

ASKER

# date
Sun Oct 12 13:13:57 PDT 2008
# ps -aux | grep dd
root 1083 0.2 0.1 1400 728 p0 DL+ 1:01PM 0:11.42 dd if=/dev/da0 bs
# top
last pid: 1116; load averages: 0.64, 0.54, 0.37 up 0+00:40:56 13:15:20
74 processes: 2 running, 72 sleeping
CPU states: 9.3% user, 0.0% nice, 6.6% system, 6.2% interrupt, 77.8% idle
Mem: 158M Active, 42M Inact, 81M Wired, 748K Cache, 77M Buf, 714M Free
Swap: 2222M Total, 2222M Free

PID USERNAME THR PRI NICE SIZE RES STATE TIME WCPU COMMAND
818 root 1 96 0 98M 53628K select 1:14 4.88% Xorg
939 kayve 1 96 0 13108K 9860K select 0:14 1.56% metacity
1083 root 1 -8 0 1400K 728K physrd 0:12 0.54% dd

ASKER

last pid: 1116; load averages: 0.44, 0.51, 0.37 up 0+00:42:40 13:17:04
74 processes: 1 running, 73 sleeping
CPU states: 4.3% user, 0.0% nice, 3.1% system, 6.6% interrupt, 85.9% idle
Mem: 159M Active, 42M Inact, 81M Wired, 748K Cache, 77M Buf, 713M Free
Swap: 2222M Total, 2222M Free

PID USERNAME THR PRI NICE SIZE RES STATE TIME WCPU COMMAND
1083 root 1 -8 0 1400K 728K physrd 0:14 0.68% dd

ASKER

last pid: 1140; load averages: 0.75, 0.58, 0.45 up 0+00:49:21 13:23:45
74 processes: 1 running, 73 sleeping
CPU states: 19.5% user, 0.0% nice, 6.2% system, 6.6% interrupt, 67.7% idle
Mem: 162M Active, 42M Inact, 81M Wired, 748K Cache, 77M Buf, 710M Free
Swap: 2222M Total, 2222M Free

PID USERNAME THR PRI NICE SIZE RES STATE TIME WCPU COMMAND
818 root 1 97 0 98M 53628K select 1:31 2.59% Xorg
1083 root 1 -8 0 1400K 728K physrd 0:20 0.49% dd

ASKER

last pid: 1150; load averages: 0.40, 0.41, 0.40 up 0+00:55:02 13:29:26
72 processes: 2 running, 70 sleeping
CPU states: 30.4% user, 0.0% nice, 5.8% system, 5.1% interrupt, 58.8% idle
Mem: 163M Active, 42M Inact, 82M Wired, 748K Cache, 77M Buf, 709M Free
Swap: 2222M Total, 2222M Free

PID USERNAME THR PRI NICE SIZE RES STATE TIME WCPU COMMAND
818 root 1 96 0 98M 53628K select 1:39 0.34% Xorg
1083 root 1 -8 0 1400K 728K physrd 0:25 0.15% dd

ASKER

kayve 1 5 0 2948K 2796K ttyin 0:00 0.00% zsh
last pid: 1166; load averages: 0.30, 0.42, 0.41 up 0+00:59:50 13:34:14
73 processes: 1 running, 72 sleeping
CPU states: 2.3% user, 0.0% nice, 3.5% system, 3.9% interrupt, 90.2% idle
Mem: 163M Active, 42M Inact, 82M Wired, 748K Cache, 77M Buf, 709M Free
Swap: 2222M Total, 2222M Free

PID USERNAME THR PRI NICE SIZE RES STATE TIME WCPU COMMAND
1083 root 1 -8 0 1400K 728K physrd 0:29 0.39% dd
818 root 1 96 0 98M 53628K select 1:48 0.15% Xorg

ASKER

last pid: 1174; load averages: 0.30, 0.35, 0.38 up 0+01:05:49 13:40:13
73 processes: 1 running, 72 sleeping
CPU states: 13.6% user, 0.0% nice, 2.7% system, 5.1% interrupt, 78.6% idle
Mem: 163M Active, 42M Inact, 82M Wired, 748K Cache, 78M Buf, 709M Free
Swap: 2222M Total, 2222M Free

PID USERNAME THR PRI NICE SIZE RES STATE TIME WCPU COMMAND
1083 root 1 -8 0 1400K 728K physrd 0:34 0.20% dd

ASKER

989 kayve 1 5 0 2948K 2796K ttyin 0:00 0.00% zsh
last pid: 1281; load averages: 0.13, 0.23, 0.20 up 0+01:57:19 14:31:43
72 processes: 2 running, 70 sleeping
CPU states: 10.1% user, 0.0% nice, 1.9% system, 5.1% interrupt, 82.9% idle
Mem: 163M Active, 42M Inact, 83M Wired, 748K Cache, 80M Buf, 707M Free
Swap: 2222M Total, 2222M Free

PID USERNAME THR PRI NICE SIZE RES STATE TIME WCPU COMMAND
1083 root 1 -8 0 1400K 728K physrd 1:19 0.49% dd
818 root 1 96 0 98M 53628K select 2:22 0.10% Xorg

ASKER

dd has been running 1h 15m now. Is that normal?

ASKER

YAAAAY! {:D

(oops. Sorry. That wasn't a proper English word.)

# dd if=/dev/da0 bs=65536 of=/dev/null
2442045+1 records in
2442045+1 records out
160041885696 bytes transferred in 5718.122211 secs (27988539 bytes/sec)
# echo $?
0
#

ASKER

"yay" is an expression of joy in English.

SOLUTION

gheist🇧🇪

membership

Create a free account to see this answer

Signing up is free and takes 30 seconds. No credit card required.

ASKER

2442045 records of 6536 bytes each iso only 149GB. Is that a problem?

gheist🇧🇪

Drives are sold using decimal terabytes.
it is 160GB in marketing catalog
that yields 149 binary GB

Problems are at 128GB, 32GB,8GB,2GB,1TB (binary xB)

ASKER

I have to get out my screwdriver and put the 160GB hard disk in the computer instead of in
the nexSTAR?

http://www.monkeyview.net/id/965/fsck/torrent/p5120226.vhtml

gheist🇧🇪

Make sure you have only target disk and install cdrom when installing. otherwise you may lose more data than you wish.

ASKER

I should burn this disk:

ftp://ftp.freebsd.org/pub/FreeBSD/releases/i386/ISO-IMAGES/7.0/7.0-RELEASE-i386-livefs.iso

or this one?

ftp://ftp.freebsd.org/pub/FreeBSD/releases/i386/ISO-IMAGES/7.0/7.0-RELEASE-i386-bootonly.iso

ASKER

I can boot to single user mode with no disk. I have not run fsck -f (although I did use -y). I need
cd "target?" to fix hard disk is what you are saying?

ASKER

When you say "target disk" you mean the abused 160GB hard drive I guess?

ASKER

So "only target disk" means I get out my screwdriver and swap the "clunker" 60GB FreeBSD 6.2 that I
am currently running out of the master slot

http://www.monkeyview.net/id/965/fsck/torrent/p5120212.vhtml

and put the 160GB FreeBSD 7.0 abused by fsck disk into that slot. I need "install cd" which I think
is one of two options above. I boot up on cd instead of hardrive 160GB FreeBSD 7.0 and do the
fsck's as directed? Or boot up on 160GB FreeBSD abused by fsck and then put the cd in or using
my MBR direct booting to master hard drive?

gheist🇧🇪

CD1 is smallest that does without network connection.

ASKER

So I want bootonly, and I should screwdriver in the "target disk" into the master slot?

SOLUTION

gheist🇧🇪

membership

Create a free account to see this answer

Signing up is free and takes 30 seconds. No credit card required.

ASKER

Maybe I shouldn't be using the term "master" because I do not fully understand it. The thing I felt like
clarifying is that in this picture:

http://www.monkeyview.net/id/965/fsck/torrent/p5120212.vhtml

you seem me unscrewing the bottom of my laptop where the hard drive goes. Right now,
the disk that is screwed in there is the 60GB running FreeBSD 6.2 disk that is "clunky" but
not "abused by fsck."

My understanding is that I should unscrew this and take the disk out of the "nexStar"

http://www.monkeyview.net/id/965/fsck/torrent/p5120226.vhtml

which IS the 160GB disk running FreeBSD 7.0 that is abused by fsck and screw that into
the actual laptop frame which is depicted in the first URL on this post.

When a disk is screwed into "nexStar" it appears as /dev/da0
When a disk is screwed into the laptop frame, it appears as /dev/ad0

Just a review of what I posted before:

# cat /etc/fstab
# Device Mountpoint FStype Options Dump Pass#
/dev/ad0s2b none swap sw 0 0
/dev/ad0s2a / ufs rw 1 1
/dev/ad0s2e /tmp ufs rw 2 2
/dev/da0s4 /mnt/usr ufs rw 2 2 <<----took out this line
/dev/ad0s2f /usr ufs rw 2 2
/dev/ad0s2d /var ufs rw 2 2
/dev/acd0 /cdrom cd9660 ro,noauto 0 0

You see the /etc/fstab that I actually have edited I think. I think I had to take out the line
with the /dev/da0s4 which is currently the 160GB FreeBSD 7.0 disk that is abused by
fsck, but by using my screwdriver, I can make that disk appear in my /dev ls as ad0 instead.
That is what I am assuming I should do.

gheist🇧🇪

What are you talking about?

ASKER

I "abused" my 160GB hard drive. I am running a _DIFFERENT_ hard drive. I used a purple USB linked
hard drive enclosure called a nexSTAR to run a bus from one hard drive to the other. I am operating
under the assumption that I am to reinstall the fsck abused 160GB hard drive into my laptop case

(have you looked at the two pictures in the URLs I gave you above? that would be simplest)

and then I can boot on the 160GB hard drive running FreeBSD 7.0 that needs an fsck instead
of running the 60GB hard drive running FreeBSD 6.2 that is in the laptop case now. I guess you
said it didn't matter what I boot, so perhaps I could do either. I just don't want to make any mistakes
and abuse my hardware more. Single user mode was mentioned, and the only way I know how
to do single user mode is to get my screwdriver out and put the right hard drive in the laptop...although
I guess I could envision mounting a /dev/da0s4, er.. I mean NOT mounting it but following the
instructions you gave very carefully that is listed above.

[::clip from above::]
gheist:
Drives are sold using decimal terabytes.
it is 160GB in marketing catalog
that yields 149 binary GB

Problems are at 128GB, 32GB,8GB,2GB,1TB (binary xB)
[::end clip::]

thanks for the clarifcation! ACK! {:) I should have realized...

I felt like commenting on this. I am taking a few courses at UC-Berkeley lately, and
this summer they told us a better system of doing this stuff:

Kilo = 1000 KiBI = 1024
Mega =1000 000 MeBI = 1024 * 1024
(etc all the way up to "yobi")
People should adopt this system! {:) It would avoid these problems!

Okay. I just realized I made a typo:

you seem me unscrewing the bottom of my laptop

seem = see

Sorry about that.

ASKER

When you told me to mount my hard drive with a screwdriver, I thought this is what you meant!

ASKER

okay. the procedure has been followed. Newest hard drive screwed into laptop. Bootonly disk provided
confusing irritations with sysinstall. Tried fixit mode in sysinstall. Was I supposed to let it install?

# fsck -f -y /usr
** /dev/da0s4 (NO WRITE)
** Last Mounted on /mnt/usr
** Phase 1 - Check Blocks and Sizes
INCORRECT BLOCK COUNT I=2779162 (4 should be 0)
CORRECT? no

fsck_ufs: cannot alloc 871186332 bytes for inoinfo
#

This is what it looked like. It takes maybe 20 seconds to run. I type the command fsck -f -y /usr
and that exact thing happened again. It didn't run for 1h15m it gave me my prompt back. All
other filesystems (/, and /var) were okay. I didn't do the swap partition. Btw.. I accidently
configured this hard drive to have 4 partitions instead of 4 slices so MBR comes up and says

FreeBSD press f1
FreeBSD press f2
FreeBSD press f3
FreeBSD press f4

Also, I didn't allocate one area of disk to any partitions, approximately 10% thinking that would
help things. All this configuration happened many months ago before I managed to install
KDE and run anything. I am just clarifying the configuration.

ASKER

The guy downstairs said I need to kernel debug core. I found this page:

http://www.freebsd.org/doc/en/books/developers-handbook/kerneldebug-gdb.html

He mounted my /var partition from the USB nexstar and found the crash directory. I mounted
my /usr partition on /mnt/nexstar and found that path they were talking about in the handbook..
so I managed to make the kdgb work.

# kgdb /mnt/nexstar/obj/usr/src/sys/KV_KERN/kernel.debug vmcore.12
[GDB will not be able to debug user-mode threads: /usr/lib/libthread_db.so: Undefined symbol "ps_pglobal_lookup"]
GNU gdb 6.1.1 [FreeBSD]
Copyright 2004 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB. Type "show warranty" for details.
This GDB was configured as "i386-marcel-freebsd".

Unread portion of the kernel message buffer:
dev = ad0s4d, block = 1, fs = /usr
panic: ffs_blkfree: freeing free block
cpuid = 0
Uptime: 8m2s
Physical memory: 1010 MB
Dumping 78 MB: 63 47 31 15

#0 0x00000000 in ?? ()
(kgdb)

ASKER

# kgdb /mnt/nexstar/obj/usr/src/sys/KV_KERN/kernel.debug vmcore.0
[GDB will not be able to debug user-mode threads: /usr/lib/libthread_db.so: Undefined symbol "ps_pglobal_lookup"]
GNU gdb 6.1.1 [FreeBSD]
Copyright 2004 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB. Type "show warranty" for details.
This GDB was configured as "i386-marcel-freebsd".

Unread portion of the kernel message buffer:
panic: ufs_dirbad: /usr: bad dir ino 9750624 at offset 28160: mangled entry
cpuid = 0
Uptime: 5m6s
Physical memory: 1010 MB
Dumping 149 MB: 134 118 102 86 70 54 38 22 6

#0 0x00000000 in ?? ()
(kgdb)

ASKER

(kgdb) quit
# pwd
/mnt/tmp/crash
#

SOLUTION

gheist🇧🇪

membership

Create a free account to see this answer

Signing up is free and takes 30 seconds. No credit card required.

ASKER

I think my hardware is bad. It rebooted last night with no prompting. I need a whole
new puter. I found a website about mangled inodes that was talking about bad hardware.

http://phaq.phunsites.net/2007/07/01/ufs_dirbad-panic-with-mangled-entries-in-ufs/

They distributed free openSolaris disks at San Francisco State University where I am taking
one course and auditing another. I think I will try running that on the new puter my sugar
momma promises to get me for Xmas.

SOLUTION

gheist🇧🇪

membership

Create a free account to see this answer

Signing up is free and takes 30 seconds. No credit card required.

ASKER

The user's guide is not going to admit that their product is substandard.

ASKER

plus I think we have been down this road. The user's guides are written for Microsoft users and doesn't give any significant information in my recollection. I think they are around here somewhere...

ASKER

I am getting my screwdriver out to do swapon now.

ASKER

"swapon" gave USAGE: error, but "swapon -a" seemed to know what it was doing...

but the fsck -y -f /usr did the exact same thing.. maybe It thought for a solid minute or
two instead of 20 seconds. Maybe that was because I was staring at it.

ASKER

# cat /mnt/var/crash/info.0
Dump header from device /dev/ad0s2b
Architecture: i386
Architecture Version: 2
Dump Length: 156418048B (149 MB)
Blocksize: 512
Dumptime: Tue Sep 16 20:43:27 2008
Hostname: kv_bsd.sfsu.edu
Magic: FreeBSD Kernel Dump
Version String: FreeBSD 7.0-RELEASE #0: Sun Jun 22 10:42:15 PDT 2008
root@kv_bsd.sfsu.edu:/usr/obj/usr/src/sys/KV_KERN
Panic String: ufs_dirbad: /usr: bad dir ino 9750624 at offset 28160: mangled entry
Dump Parity: 2465829192
Bounds: 0
Dump Status: good
#

ASKER

# cat /mnt/var/crash/info.12
Dump header from device /dev/ad0s2b
Architecture: i386
Architecture Version: 2
Dump Length: 82092032B (78 MB)
Blocksize: 512
Dumptime: Thu Oct 9 18:11:22 2008
Hostname: kv_bsd.sfsu.edu
Magic: FreeBSD Kernel Dump
Version String: FreeBSD 7.0-RELEASE #0: Sun Jun 22 10:42:15 PDT 2008
root@kv_bsd.sfsu.edu:/usr/obj/usr/src/sys/KV_KERN
Panic String: ffs_blkfree: freeing free block
Dump Parity: 3928695930
Bounds: 12
Dump Status: good
# cat /mnt/var/crash/info.13
cat: /mnt/var/crash/info.13: No such file or directory
#

ASKER

is there a way to view an inode by number? that phaq page says I need to destroy that inode,
but I wanted to see what it was first. maybe back up whatever it is pointing to?

ASKER

this page says to use "dcat" but I don't have it. Is it something on FreeBSD7.0?

http://safari.oreilly.com/0321268172/ch17lev1sec8

221 Logout.
% dcat -f freebsd freebsd.dd 56 8 | dd bs=256 skip=9750624 count=1 | xxd
zsh: command not found: dcat
dd: skip reached end of input
zsh: command not found: xxd
%

ASKER

could it be that the hard disk is too big for my laptop? I don't know where the manual is right now, but might that be something it would tell me?

ASKER

I'm trying to run this dd command...

zsh: terminated dd bs=256 skip=9750624 count=1 |
zsh: exit 127 xxd
% dd bs=256 skip=9750624 count=1

It's been running for HOURS

I tried top

root 1738 0.0 0.1 1500 940 p2 I+ 9:35PM 0:00.00 man dd
# top

last pid: 1832; load averages: 0.04, 0.03, 0.00 up 0+06:04:55 22:21:29
69 processes: 2 running, 67 sleeping
CPU states: 6.6% user, 0.0% nice, 1.6% system, 0.4% interrupt, 91.4% idle
Mem: 208M Active, 194M Inact, 99M Wired, 644K Cache, 111M Buf, 494M Free
Swap: 2222M Total, 2222M Free

PID USERNAME THR PRI NICE SIZE RES STATE TIME WCPU COMMAND
810 root 1 96 0 107M 62692K select 4:58 0.29% Xorg
983 kayve 4 20 0 164M 151M kserel 30:27 0.00% firefox-bin
985 kayve 4 20 0 22484K 16848K kserel 0:40 0.00% gnome-terminal
931 kayve 1 96 0 3276K 2088K select 0:25 0.00% gam_server
939 kayve 1 96 0 13104K 9892K select 0:20 0.00% metacity
957 kayve 1 96 0 24284K 14324K select 0:11 0.00% wnck-applet
994 root 1 96 0 1336K 820K select 0:09 0.00% moused
969 kayve 1 96 0 11500K 8344K select 0:08 0.00% gnome-screensaver
942 kayve 1 96 0 26612K 17528K select 0:07 0.00% gnome-panel
928 kayve 4 20 0 18720K 13660K kserel 0:04 0.00% gnome-settings-daem
944 kayve 3 20 0 33064K 20196K kserel 0:03 0.00% nautilus
964 kayve 1 96 0 20768K 14448K select 0:03 0.00% clock-applet
921 kayve 1 96 0 5796K 4440K select 0:02 0.00% gconfd-2
952 kayve 1 96 0 7284K 4704K select 0:01 0.00% gnome-vfs-daemon
961 kayve 1 96 0 25148K 14908K select 0:01 0.00% mixer_applet2
901 kayve 1 96 0 18096K 12736K select 0:01 0.00% gnome-session
946 kayve 4 20 0 6660K 4164K kserel 0:01 0.00% bonobo-activation-s
819 root 1 96 0 3504K 3032K select 0:00 0.00% sendmail
1829 root 1 96 0 2408K 1660K RUN 0:00 0.00% top
966 kayve 1 96 0 16356K 11964K select 0:00 0.00% notification-area-a
1174 kayve 1 20 0 2880K 2716K pause 0:00 0.00% zsh
959 kayve 1 96 0 2948K 1612K select 0:00 0.00% mapping-daemon
991 root 1 20 0 2640K 2440K pause 0:00 0.00% zsh
919 kayve 1 96 0 3180K 2164K select 0:00 0.00% ssh-agent
829 root 1 8 0 1388K 1100K nanslp 0:00 0.00% cron
1347 root 1 20 0 2640K 2440K pause 0:00 0.00% zsh
986 kayve 1 4 0 3080K 1584K sbwait 0:00 0.00% gnome-pty-helper
647 root 1 96 0 1376K 1028K select 0:00 0.00% syslogd
987 kayve 1 8 0 1756K 1336K wait 0:00 0.00% sh
1345 kayve 1 8 0 1756K 1356K wait 0:00 0.00% sh
1173 kayve 1 8 0 1756K 1356K wait 0:00 0.00% sh
924 kayve 1 96 0 1916K 1572K select 0:00 0.00% dbus-daemon
581 root 1 96 0 528K 388K select 0:00 0.00% devd
803 root 1 96 0 10604K 4492K select 0:00 0.00% gdm-binary
974 kayve 1 8 0 1716K 1272K wait 0:00 0.00% sh
727 root 1 96 0 1288K 824K select 0:00 0.00% usbd
#
# top | grep dd

Should it be on top? Will it ever end?

ASKER

You can't tell from the above command line dump, but "top | grep dd" returned absolutely nothing.

gheist🇧🇪

Your filesystem is so tricky broken I'd suggest you fill bug report at FreeBSD to fix it, and for your computing experience install fresh OS on that computer.

It is actually a customized kernel, it may crash because of you. Why not GENERIC kernel?

gheist🇧🇪

You might be able to fix filesystem using OpenBSD or NetSBD CD-s. But otherwise reinstall.

ASKER

I think KV_KERN is because of acd0 only. It should be mostly the same as GENERIC

I want to look at the inode. What am I doing wrong with dd?

# dd bs=256 skip=9750624 count=1 > /root/bad.inode.txt
zsh: terminated dd bs=256 skip=9750624 count=1 > /root/bad.inode.txt
You have new mail.
# dd if=/mnt/usr bs=256 skip=9750624 count=1 of=/root/bad.inode.txt
0+0 records in
0+0 records out
0 bytes transferred in 0.000050 secs (0 bytes/sec)
# dd if=/dev/da0 bs=256 skip=9750624 count=1 of=/root/bad.inode.txt
dd: /dev/da0: Invalid argument
0+0 records in
0+0 records out
0 bytes transferred in 0.000222 secs (0 bytes/sec)
#

ASKER

# mount -t msdosfs /dev/da1s1 /mnt/pix
# dd bs=256 skip=9750624 count=1 > /root/bad.inode.txt
zsh: terminated dd bs=256 skip=9750624 count=1 > /root/bad.inode.txt
You have new mail.
# dd if=/mnt/usr bs=256 skip=9750624 count=1 of=/root/bad.inode.txt
0+0 records in
0+0 records out
0 bytes transferred in 0.000050 secs (0 bytes/sec)
# dd if=/dev/da0 bs=256 skip=9750624 count=1 of=/root/bad.inode.txt
dd: /dev/da0: Invalid argument
0+0 records in
0+0 records out
0 bytes transferred in 0.000222 secs (0 bytes/sec)
# dd if=/dev/da0 bs=256 skip=974 count=1 of=/root/inode.974.txt
dd: /dev/da0: Invalid argument
0+0 records in
0+0 records out
0 bytes transferred in 0.000244 secs (0 bytes/sec)
# dd if=/dev/da0s4 bs=256 skip=974 count=1 of=/root/inode.974.txt
dd: /dev/da0s4: Invalid argument
0+0 records in
0+0 records out
0 bytes transferred in 0.000222 secs (0 bytes/sec)
# dd if=/mnt/usr bs=256 skip=974 count=1 of=/root/inode.974.txt
0+0 records in
0+0 records out
0 bytes transferred in 0.000049 secs (0 bytes/sec)
# dd if=/dev/da0s4 bs=256 skip=974 count=1 of=/root/inode.974.txt
dd: /dev/da0s4: Invalid argument
0+0 records in
0+0 records out
0 bytes transferred in 0.000222 secs (0 bytes/sec)
# dd if=/dev/da0 bs=256 skip=974 count=1 of=/root/inode.974.txt
dd: /dev/da0: Invalid argument
0+0 records in
0+0 records out
0 bytes transferred in 0.000221 secs (0 bytes/sec)
# umount /mnt/var
# umount /mnt/tmp
# dd if=/dev/da0 bs=256 skip=974 count=1 of=/root/inode.974.txt
dd: /dev/da0: Invalid argument
0+0 records in
0+0 records out
0 bytes transferred in 0.000221 secs (0 bytes/sec)

I don't understand what is happening. You told me to do dd if=/dev/da0 and it was fine. Why
do my extra options mess things up?

ASKER

# ls /dev
acd0 ctty fido net1 stderr ttyv8
acpi cuad0 fw0 net2 stdin ttyv9
ad0 cuad0.init fw0.0 net3 stdout ttyva
ad0s1 cuad0.lock fwmem0 net4 sysmouse ttyvb
ad0s2 da0 fwmem0.0 network ttyd0 ttyvc
ad0s2a da0s1 geom.ctl nfs4 ttyd0.init ttyvd
ad0s2b da0s1a io nfslock ttyd0.lock ttyve
ad0s2c da0s1c kbd0 null ttyp0 ttyvf
ad0s2d da0s2 kbd1 pass0 ttyp1 ums0
ad0s2e da0s2b kbdmux0 pccard0.cis ttyp2 urandom
ad0s2f da0s2c klog pccard1.cis ttyv0 usb
agpgart da0s3 kmem pci ttyv1 usb0
apm da0s3c log ppi0 ttyv2 usb1
ata da0s3d lpt0 psm0 ttyv3 usb2
atkbd0 da0s4 lpt0.ctl ptyp0 ttyv4 usb3
bpsm0 devctl mdctl ptyp1 ttyv5 xpt0
console devstat mem ptyp2 ttyv6 zero
consolectl fd net random ttyv7
# dd if=/dev/da0 bs=256 of=/root/inode.974.txt
dd: /dev/da0: Invalid argument
0+0 records in
0+0 records out
0 bytes transferred in 0.000231 secs (0 bytes/sec)
#

ASKER

I guess I don't want to use this?

http://www.linux.com/feature/126525

You should not place the output directory on the same block
device you are trying to rescue files from. This might add the
same file to the block device ahead of the current reading posi-
tion, causing magicrescue to find the same file again later. In
the worst theoretical case, this could cause a loop where the
same file is extracted thousands of times until disk space is
exhausted. You are also likely to overwrite the deleted files
you were looking for in the first place.

-r recipe
Mandatory. Recipe name, file, or directory. Specify this as
either a plain name (e.g. "jpeg-jfif") or a path (e.g.
recipes/jpeg-jfif). If it doesn't find such a file in the cur-
rent directory, it will look in ./recipes and PREFIX/share/magi-
crescue/recipes, where PREFIX is the path you installed to, e.g.
/usr/local.

If recipe is a directory, all files in that directory will be
treated as recipes.

Browse the PREFIX/share/magicrescue/recipes directory to see

I'm not trying to rescue. I'm not sure what that was suggested on

http://osdir.com/ml/freebsd.devel.file-systems/2006-11/msg00069.html

ASKER

Okay I give up examining the inode. I am just going to do the stuff in single user mode fsdb:

http://phaq.phunsites.net/2007/07/01/ufs_dirbad-panic-with-mangled-entries-in-ufs/

gheist🇧🇪

If you do not need files on broken system then just reinstall with something useful. Otherwise boot off installation CD of either OpenBSD or NetBSD and run their fsck.

dd will not fix anything at least the way you use it.

Are you trying to look stupid or you just are like that?

ASKER

I can't dump my data unless I get another hard drive. Why is it stupid to try to learn better techniques??

http://lists.freebsd.org/pipermail/freebsd-current/2003-December/015992.html

I have negative GEN fields. I am posting pictures of single user mode

http://www.monkeyview.net/id/965/fsck/inodes/index.vhtml

Click on thumbnails for better view.

ASKER

I know now I need to learn better mirroring techniques. I know I could just reinstall. I want to reinstall but I guess my "problem" is that I am a pack rat. Right now the guy downstairs showed me a cool thing with kgbd and now I have discoved I can do what I wanted to with fsdb.

Is it safe to run fsdb not in single user mode?

I know I could burn dvds but I don't want to make that mess. Maybe I will try to get my stuff on hard disk
2 and reinstall that first.

gheist🇧🇪

I do not look into your pictures - they are on slow server and overly huge. Use ee-stuff to post them.

It is obvious bug in that fsck you are using it allocates 800MB of RAM. I beg you try OpenBSD's or NetBSD's fsck to see if bug is generic or specific.

OpenBSD has ffs recovery tool designed for you that lets you bring up even after overwriting beginning of disk.

gheist🇧🇪

1) since you were able to read whole disk odds are high you will install Linux till the state you are able to use smartctl or smartd to judge disk health.
2) for sake of bug reporting use G4U to transfer sick partition for safekeeping and bug reporting to some ftp server that has reasonable space
3) I am not in rat packing business - do you eat them over there in US?

ASKER

I don't know why you keep insiting on insulting me. I was using dd to see what was
in the inode to decide if I could just delete it or not.

ASKER

insiting = insisting

gheist🇧🇪

fsck -y includes saving inodes under lost+found, so you can examine their content after fsck. It is fsck bug it crashes allocating enormous amount of ram. so I suggest ignoring insults and getting OpenBSD, NetBSD, FreeBSD 6.3 or RatBSD if you choose so that you can check filesystem with working fsck.

It is actually more useful if you image your current failing disk and examine it later.

For reference: fsck counts blocks from beginning of filesystem aka BSD partition aka Disklabel partition like da5s3i

Are you able to burn ISO images into CD-RW so that you can diagnose your laptop without extra finance?

fsck -n prefers data-losing way.

ASKER

Why can't I mount my root directory for /dev/da1s2?

# df -m
Filesystem 1M-blocks Used Avail Capacity Mounted on
/dev/ad0s2a 290 74 192 28% /
devfs 0 0 0 100% /dev
/dev/ad0s2e 387 13 342 4% /tmp
/dev/ad0s2f 13252 9747 2445 80% /usr
/dev/ad0s2d 387 29 327 8% /var
/dev/da1s2d 496 221 235 48% /mnt/var
/dev/da1s2e 550 358 148 71% /mnt/tmp
/dev/da0s3 9675 1422 7478 16% /mnt/nexstar/var
/dev/da0s4 124822 49491 65345 43% /mnt/nexstar/usr
/dev/da1s2f 36333 23147 10279 69% /mnt/usr
# ls /dev
acd0 da0s1 fwmem0 pass0 ttyv2
acpi da0s1a fwmem0.0 pass1 ttyv3
ad0 da0s1c geom.ctl pccard0.cis ttyv4
ad0s1 da0s2 io pccard1.cis ttyv5
ad0s2 da0s2b kbd0 pci ttyv6
ad0s2a da0s2c kbd1 ppi0 ttyv7
ad0s2b da0s3 kbdmux0 psm0 ttyv8
ad0s2c da0s4 klog ptyp0 ttyv9
ad0s2d da1 kmem ptyp1 ttyva
ad0s2e da1s1 log ptyp2 ttyvb
ad0s2f da1s2 lpt0 random ttyvc
agpgart da1s2a lpt0.ctl stderr ttyvd
apm da1s2b mdctl stdin ttyve
ata da1s2c mem stdout ttyvf
atkbd0 da1s2d net sysmouse ums0
bpsm0 da1s2e net1 ttyd0 urandom
console da1s2f net2 ttyd0.init usb
consolectl devctl net3 ttyd0.lock usb0
ctty devstat net4 ttyp0 usb1
cuad0 fd network ttyp1 usb2
cuad0.init fido nfs4 ttyp2 usb3
cuad0.lock fw0 nfslock ttyv0 xpt0
da0 fw0.0 null ttyv1 zero
# ls /mnt
drive nexstar nexusr pix tmp usr var vista xcraft
# mkdir /mnt/root
# mount /dev/da1s2a /mnt/root
mount: /dev/da1s2a: Operation not permitted
# ls /dev/da1s2a
/dev/da1s2a
# pushd /dev/da1s2a
pushd: not a directory: /dev/da1s2a
# mount /dev/da1s2c /mnt/root
mount: /dev/da1s2c: Operation not permitted

ASKER

I did a clri 975064 in fsdb (not shown, but on monkeyview)
quit fsdb and fsck_ufs -y -f still didn't work (looked same as other runs)
went back into fsdb

fsck has always talked about INCORRECT BLOCK COUNT I=2779162
so I thought it would be good to go to that inode.

GEN=ffffffff.. very large unsigned or negative signed number. Doesn't seem right, does it?

I guess I didn't take pictures but inode 2779161 was okay and 2779163 and 2779164 also
had strange GEN.
pa180009.jpg
pa180010.jpg
pa180018.jpg
pa180020.jpg

ASKER

Now I have all three hardrives mounted and running. Doing cp -r of /usr (that was probably stupid I guess).

getting close to done maybe

# df -m
Filesystem 1M-blocks Used Avail Capacity Mounted on
/dev/ad0s2a 290 74 192 28% /
devfs 0 0 0 100% /dev
/dev/ad0s2e 387 13 342 4% /tmp
/dev/ad0s2f 13252 9754 2438 80% /usr
/dev/ad0s2d 387 29 327 8% /var
/dev/da1s2d 496 221 235 48% /mnt/var
/dev/da1s2e 550 358 148 71% /mnt/tmp
/dev/da0s3 9675 1422 7478 16% /mnt/nexstar/var
/dev/da0s4 124822 58535 56300 51% /mnt/nexstar/usr
/dev/da1s2f 36333 23147 10279 69% /mnt/usr
#

Lots and lots of space on stupid new hard drive.

frustrated by attempts to get / from da1s2
# mkdir /mnt/root
# mount /dev/da1s2a /mnt/root
mount: /dev/da1s2a: Operation not permitted
# ls /dev/da1s2a
/dev/da1s2a
# pushd /dev/da1s2a
pushd: not a directory: /dev/da1s2a
# mount /dev/da1s2c /mnt/root
mount: /dev/da1s2c: Operation not permitted

ASKER

Okay something is seeming funny now

Filesystem 1M-blocks Used Avail Capacity Mounted on
/dev/ad0s2a 290 74 192 28% /
devfs 0 0 0 100% /dev
/dev/ad0s2e 387 13 342 4% /tmp
/dev/ad0s2f 13252 9754 2438 80% /usr
/dev/ad0s2d 387 29 327 8% /var
/dev/da1s2d 496 221 235 48% /mnt/var
/dev/da1s2e 550 358 148 71% /mnt/tmp
/dev/da0s3 9675 1422 7478 16% /mnt/nexstar/var
/dev/da0s4 124822 63482 51353 55% /mnt/nexstar/usr
/dev/da1s2f 36333 23147 10279 69% /mnt/usr

ASKER

oh dear. {:( rm -rf will take almost as long.

ASKER

0 0 0 100% /dev
/dev/ad0s2e 387 13 342 4% /tmp
/dev/ad0s2f 13252 9754 2438 80% /usr
/dev/ad0s2d 387 29 327 8% /var
/dev/da1s2d 496 221 235 48% /mnt/var
/dev/da1s2e 550 358 148 71% /mnt/tmp
/dev/da0s3 9675 1422 7478 16% /mnt/nexstar/var
/dev/da0s4 124822 65278 49557 57% /mnt/nexstar/usr
/dev/da1s2f 36333 23147 10279 69% /mnt/usr
# ps -aux | grep mnt
root 1133 19.5 0.7 7704 6964 p0 D+ 4:24PM 22:26.27 cp -r /mnt/usr .
root 1436 0.0 0.0 404 252 p2 D+ 6:37PM 0:00.00 grep mnt
# kill 1133
# df
Filesystem 1K-blocks Used Avail Capacity Mounted on
/dev/ad0s2a 297326 76288 197252 28% /
devfs 1 1 0 100% /dev
/dev/ad0s2e 396526 13582 351222 4% /tmp
/dev/ad0s2f 13570958 9988396 2496886 80% /usr
/dev/ad0s2d 396526 29806 334998 8% /var
/dev/da1s2d 508654 226762 241200 48% /mnt/var
/dev/da1s2e 564206 366782 152288 71% /mnt/tmp
/dev/da0s3 9907722 1456756 7658350 16% /mnt/nexstar/var
/dev/da0s4 127818042 64065320 53527280 54% /mnt/nexstar/usr
/dev/da1s2f 37205502 23703162 10525900 69% /mnt/usr
# df -m
Filesystem 1M-blocks Used Avail Capacity Mounted on
/dev/ad0s2a 290 74 192 28% /
devfs 0 0 0 100% /dev
/dev/ad0s2e 387 13 342 4% /tmp
/dev/ad0s2f 13252 9754 2438 80% /usr
/dev/ad0s2d 387 29 327 8% /var
/dev/da1s2d 496 221 235 48% /mnt/var
/dev/da1s2e 550 358 148 71% /mnt/tmp
/dev/da0s3 9675 1422 7478 16% /mnt/nexstar/var
/dev/da0s4 124822 62542 52293 54% /mnt/nexstar/usr
/dev/da1s2f 36333 23147 10279 69% /mnt/usr
# df -m
Filesystem 1M-blocks Used Avail Capacity Mounted on
/dev/ad0s2a 290 74 192 28% /
devfs 0 0 0 100% /dev
/dev/ad0s2e 387 13 342 4% /tmp
/dev/ad0s2f 13252 9754 2438 80% /usr
/dev/ad0s2d 387 29 327 8% /var
/dev/da1s2d 496 221 235 48% /mnt/var
/dev/da1s2e 550 358 148 71% /mnt/tmp
/dev/da0s3 9675 1422 7478 16% /mnt/nexstar/var
/dev/da0s4 124822 62466 52370 54% /mnt/nexstar/usr
/dev/da1s2f 36333 23147 10279 69% /mnt/usr
# df -m
Filesystem 1M-blocks Used Avail Capacity Mounted on
/dev/ad0s2a 290 74 192 28% /
devfs 0 0 0 100% /dev
/dev/ad0s2e 387 13 342 4% /tmp
/dev/ad0s2f 13252 9754 2438 80% /usr
/dev/ad0s2d 387 29 327 8% /var
/dev/da1s2d 496 221 235 48% /mnt/var
/dev/da1s2e 550 358 148 71% /mnt/tmp
/dev/da0s3 9675 1422 7478 16% /mnt/nexstar/var
/dev/da0s4 124822 62251 52585 54% /mnt/nexstar/usr
/dev/da1s2f 36333 23147 10279 69% /mnt/usr
#

ASKER

# pushd /usr/home/kayve
/usr/home/kayve ~
# ls
.ICEauthority .subversion
.Trash .thumbnails
.Xauthority .xsession-errors
.bash_history .zcompdump
.classpath .zshrc
.config 7.0-RELEASE-i386-bootonly.iso
.cshrc Desktop
.dmrc DoD_DNS.pdf
.eclipse GSECpart.pdf
.evolution NIST_SEC_DNS.pdf
.gconf OUSD Calendar 2007-2008.pdf
.gconfd SG20partitioning.pdf
.gnome bankruptcy
.gnome2 bankruptcy.iso
.gnome2_private bills
.gstreamer-0.10 chess
.gtkrc-1.2-gnome2 copy_part.pdf
.lesshst cs162
.login freeBSD70
.login_conf install43.iso
.mail_aliases memtest.iso
.mailrc memtest86-3.4.iso.zip
.metacity monkeyview
.mozilla national.pdf
.nautilus piss.jpg
.profile pp01.mpg
.project pp02.mpg
.recently-used public_html
.recently-used.xbel sniff.jpg
.rhosts thesis.pdf
.shrc train_project
.ssh ubcd411.iso
# burncd -f /dev/acd0 data install43.iso fixate
next writeable LBA 0
writing from file install43.iso size 207428 KB
written this track 38400 KB (18%) total 38400 KB

ASKER

I don't know how much I care about using this OpenBSD 4.3 install43.iso, but what the hey.

I think I will just say oh well on the / partition and use this openSolaris that is in a nice corporate
looking jacket

ASKER

This is going to be moot I guess. Maybe I should just give you points for the not really helping advice of blowing everything away, or maybe that was good advice. Professor Stan says same thing and I know you need backups as a sysadmin. Here are a couple of pictures of what the SECOND hard drive looks like. This is the first hard drive I am running now.

# uname -a
FreeBSD 6.2-RELEASE FreeBSD 6.2-RELEASE #0: Fri Jan 12 11:05:30 UTC 2007 root@dessler.cse.buffalo.edu:/usr/obj/usr/src/sys/SMP i386
#

The broken 160GB hard drive (149 Gibi bytes or whatever) is the THIRD hard drive. Above I
have mounted all three hard drives together, back a long time ago you were trying to give me
advice about the external driving enclosures and it turns out I didn't know to push it hard
so all the pins stuck in. I know that sounds real stupid, but I didn't want to force things it seemed
so delicate to me. A classmate showed me that I have to push it right so the pins go in. So
the Xcraft, and the NexStar a
pa180001.jpg
pa180002.jpg
nexStar.jpg
Xcraft.jpg

gheist🇧🇪

You did not run fsck from another system. Then you edited mounted filesystem. I think you are wasting my time. Your problem has been solved up there 10 times at least. Have a nice sleep, Sir!

ASKER

No I didn't try to figure out a bunch of new stuff.. I took your OTHER advice and finally wiped one of
my partitions clean. I will give you points when I get around to it.

ASKER

I have FIVE partitions on THREE hardrives. Okay I've been an idiot for not cleaning one.

# ls /dev
ad0 <----- old clunky 60GB HD
ad0s1 <---- broken Vista partition ONE (but with data)
ad0s2 <---|
ad0s2a <---|
ad0s2b <---|
ad0s2c <---|------- FreeBSD 6.2 running again right now (TWO)
ad0s2d <---|
ad0s2e <---|
ad0s2f <--- |

da0 <--- Hard disk THREE 160GB topic of this discussion
da0s1 <--- |
da0s1a <---|------ root (/) partion for FreeBSD 7.0 (THREE partitions)
da0s1c <---|
da0s2 <---|
da0s2b <---|-----swap for FreeBSD 7.0
da0s2c <---|
da0s3 <------ /var for FreeBSD 7.0
da0s4 <------------/usr for FreeBSD 7.0
da1 <--- Hard disk THREE in black Xcraft enclosure
da1s1 <--- Working MS Vista partition (FOUR partitions)
da1s2 <--- Old broken FreeBSD 6.3(?) now NEW openSolaris (in progress)
da1s2a <--- (FIVE partions)
da1s2b <--- Solaris being irritating.
da1s2c <--- Downloading iso for solar now despite the
da1s2d <--- stupid disk I already head.
da1s2e <--- stupid solaris wants me to use wireless
da1s2f <--- gave me a scare had to ping 4.2.2.2. to get dang wire

I am stupid. I guess I will try solaris fsck when I get around to it.

ASKER

Solaris bites my wire. I'll give you points

ASKER

You gave enough advice to certainly deserve points.

gheist🇧🇪

If you happen to go with OpenSOLARIS be so kind to update GRUB bootloader after updating system. It is great system in right hands. Have a nice day!

ASKER

It seemed they were just giving a 'boot from CD' situation. I don't like all those awful sounds coming from my CD drive! I decided to just put a FreeBSD 7.0 on that free Partion. It felt easier like something I knew better. Sorry about that. Maybe I will try the boot from CD to fsck at some point. For now, maybe I am happy to have a
nonbootable but mountable hard drive for future backups.

gheist🇧🇪

Download Mandriva One or Kubuntu - they boot from CD to make use of your Clunked OMPUTER (tm). Owing to technological differences they boot and install without that much CD head movements.