Could not find /dev/disk/by-id/ata

English Install/Boot/Login
1

Hi,

I turned on my machine this morning and got this:

could not find /dev/disk/by-id/ata-MAXTOR_STM (etc) -part 3
want me to fall back to /dev/disk/by-id/ata-MAXTOR_STM (etc)-part 3?
waiting for device /dev/disk/by-id/ata-MAXTOR_STM (etc)-part 3 to appear........................ not found--exiting to /bin/sh
$

I thought my hard drive had died. Later on today when I was more awake I tried to log into windows and that worked fine. I have just tried to log into Fedora and that works fine.

I have done fdisk and got this:
Disk /dev/sda: 160.0 GB, 160041885696 bytes
255 heads, 63 sectors/track, 19457 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x00013e16

Device Boot Start End Blocks Id System
/dev/sda1 1 1848 14844028+ 7 HPFS/NTFS
/dev/sda2 1849 1861 104422+ 83 Linux
/dev/sda3 * 1862 3901 16386300 83 Linux
/dev/sda4 3902 19457 124953570 f W95 Ext’d (LBA)
/dev/sda5 4030 7824 30483337+ 83 Linux
/dev/sda6 3902 4029 1028097 82 Linux swap / Solaris
/dev/sda7 7825 9732 15325978+ 83 Linux
/dev/sda8 9733 11094 10940233+ 83 Linux
/dev/sda9 11095 12365 10209276 83 Linux
/dev/sda10 12366 13642 10257471 83 Linux
/dev/sda11 13643 14910 10185178+ 83 Linux
/dev/sda12 14911 16311 11253501 83 Linux
/dev/sda13 16312 17481 9397993+ 83 Linux
/dev/sda14 17482 18473 7968208+ 83 Linux
/dev/sda15 18474 19457 7903948+ 83 Linux

Partition table entries are not in disk order

Disk /dev/sdb: 41.1 GB, 41110142976 bytes
255 heads, 63 sectors/track, 4998 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x00030242

Device Boot Start End Blocks Id System
/dev/sdb1 * 1 4998 40146403+ c W95 FAT32 (LBA)
Note: sector size is 2048 (not 512)

Disk /dev/sdh: 1015 MB, 1015808000 bytes
250 heads, 62 sectors/track, 32 cylinders
Units = cylinders of 15500 * 2048 = 31744000 bytes
Disk identifier: 0xba1b10d0

Device Boot Start End Blocks Id System
/dev/sdh1 * 1 32 991876 6 FAT16
Partition 1 has different physical/logical beginnings (non-Linux?):
phys=(0, 0, 62) logical=(0, 1, 1)

I am hoping someone will be able to explain what has happened.

Also I would really like to be able to retrieve my /home I have been very naughty and not backed up since June. I can see my home directory in Fedora, but I don’t have permissions so can’t access it.

Any help gratefully received.

2

We cannot guess from that long output (next time please use code tabs) which ones are your openSUSE partitions, but you should know. So - from Fedora - run fsck /dev/sdXn on those unmounted partitions.

3

I couldn’t get that to work. However I did df -h and got


Filesystem            Size  Used Avail Use% Mounted on
/dev/sda7              15G  3.6G   11G  26% /
tmpfs                 220M  696K  219M   1% /dev/shm
/dev/sda9             9.6G  151M  9.0G   2% /media/distro
/dev/sdb1              39G   33G  6.0G  85% /media/disk
/dev/sda14            7.5G  146M  7.0G   2% /media/distro_
/dev/sda13            8.9G  1.8G  6.7G  21% /media/_
/dev/sr0              101M  101M     0 100% /media/CDROM
/dev/sda3              16G   15G   35M 100% /media/disk-1
/dev/sda15            7.5G  146M  7.0G   3% /media/distro__
/dev/sda5              29G   25G  3.1G  89% /media/disk-2
/dev/sda8              11G  155M  9.6G   2% /media/distro___
/dev/sda12             11G  155M  9.9G   2% /media/distro____
/dev/sda2              99M   14M   81M  15% /media/boot
/dev/sda11            9.6G  150M  9.0G   2% /media/distro_____
/dev/sda1              15G  7.4G  6.9G  52% /media/disk-3
/dev/sda10            9.7G  151M  9.0G   2% /media/distro______

fdisk -l

Disk /dev/sda: 160.0 GB, 160041885696 bytes
255 heads, 63 sectors/track, 19457 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x00013e16

Device Boot Start End Blocks Id System
/dev/sda1 1 1848 14844028+ 7 HPFS/NTFS
/dev/sda2 1849 1861 104422+ 83 Linux
/dev/sda3 * 1862 3901 16386300 83 Linux
/dev/sda4 3902 19457 124953570 f W95 Ext’d (LBA)
/dev/sda5 4030 7824 30483337+ 83 Linux
/dev/sda6 3902 4029 1028097 82 Linux swap / Solaris
/dev/sda7 7825 9732 15325978+ 83 Linux
/dev/sda8 9733 11094 10940233+ 83 Linux
/dev/sda9 11095 12365 10209276 83 Linux
/dev/sda10 12366 13642 10257471 83 Linux
/dev/sda11 13643 14910 10185178+ 83 Linux
/dev/sda12 14911 16311 11253501 83 Linux
/dev/sda13 16312 17481 9397993+ 83 Linux
/dev/sda14 17482 18473 7968208+ 83 Linux
/dev/sda15 18474 19457 7903948+ 83 Linux

Partition table entries are not in disk order

Disk /dev/sdb: 41.1 GB, 41110142976 bytes
255 heads, 63 sectors/track, 4998 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x00030242

Device Boot Start End Blocks Id System
/dev/sdb1 * 1 4998 40146403+ c W95 FAT32 (LBA)

4

Thanks for your reply. I’m really sorry about that.

which ones are your openSUSE partitions, but you should know. So - from Fedora - run fsck /dev/sdXn on those unmounted partitions.

I couldn’t get that to work. However I did df -h and got


Filesystem            Size  Used Avail Use% Mounted on
/dev/sda7              15G  3.6G   11G  26% /
tmpfs                 220M  696K  219M   1% /dev/shm
/dev/sda9             9.6G  151M  9.0G   2% /media/distro
/dev/sdb1              39G   33G  6.0G  85% /media/disk
/dev/sda14            7.5G  146M  7.0G   2% /media/distro_
/dev/sda13            8.9G  1.8G  6.7G  21% /media/_
/dev/sr0              101M  101M     0 100% /media/CDROM
/dev/sda3              16G   15G   35M 100% /media/disk-1
/dev/sda15            7.5G  146M  7.0G   3% /media/distro__
/dev/sda5              29G   25G  3.1G  89% /media/disk-2
/dev/sda8              11G  155M  9.6G   2% /media/distro___
/dev/sda12             11G  155M  9.9G   2% /media/distro____
/dev/sda2              99M   14M   81M  15% /media/boot
/dev/sda11            9.6G  150M  9.0G   2% /media/distro_____
/dev/sda1              15G  7.4G  6.9G  52% /media/disk-3
/dev/sda10            9.7G  151M  9.0G   2% /media/distro______

I don’t really understand the above. I am used to seeing /dev/hda

fdisk -l

Disk /dev/sda: 160.0 GB, 160041885696 bytes
255 heads, 63 sectors/track, 19457 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x00013e16

   Device Boot      Start         End      Blocks   Id  System
/dev/sda1               1        1848    14844028+   7  HPFS/NTFS
/dev/sda2            1849        1861      104422+  83  Linux
/dev/sda3   *        1862        3901    16386300   83  Linux
/dev/sda4            3902       19457   124953570    f  W95 Ext'd (LBA)
/dev/sda5            4030        7824    30483337+  83  Linux
/dev/sda6            3902        4029     1028097   82  Linux swap / Solaris
/dev/sda7            7825        9732    15325978+  83  Linux
/dev/sda8            9733       11094    10940233+  83  Linux
/dev/sda9           11095       12365    10209276   83  Linux
/dev/sda10          12366       13642    10257471   83  Linux
/dev/sda11          13643       14910    10185178+  83  Linux
/dev/sda12          14911       16311    11253501   83  Linux
/dev/sda13          16312       17481     9397993+  83  Linux
/dev/sda14          17482       18473     7968208+  83  Linux
/dev/sda15          18474       19457     7903948+  83  Linux

Partition table entries are not in disk order

Disk /dev/sdb: 41.1 GB, 41110142976 bytes
255 heads, 63 sectors/track, 4998 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x00030242

   Device Boot      Start         End      Blocks   Id  System
/dev/sdb1   *           1        4998    40146403+   c  W95 FAT32 (LBA)

This one makes more sense to me now.
sda is my mine drive
sdb is my secondary drive, which is shared by linux and windows.
sda1 is my windows partition.
I believe sda2 is /boot
sda3 is root
sda5 is /home
sda6 is swap
I believe sda7 has all of Fedora.
From sda8 onwards is split into chunks. I was supposed to attempt to multiboot other distros, but they wouldn’t play nicely. I think it was Ubunta that changed all my partitions and I had to start all over again. I also believe that sda13 is linux XP desktop 2006.

I had to stop at that point as I needed to use the machine.

I have now found my original piece of paper from July 2009 with my partitions written down. I think your original request now makes sense.
If X is the drive letter and n is the partition?

I have tried fsck /dev/sda3
fsck /dev/sda5
Suse’s root and /home, but they are both mounted. Please can you explain again, I think I am missing something.

5

yes.

That’s because they get mounted automatically in Fedora’s /etc/fstab or somehow when you log in. It is not necessary to mount partitions you’re not using. IMO “automounting” is a bad thing. You should unmount these partitions before checking them with fsck. But if they are mounted … they should be all right.

Anyway, to check them :

su -l 
cd /
umount /dev/sda3
umount /dev/sda5
fsck /dev/sda3
fsck /dev/sda5
  • it is umount and not unmount
6

On 2010-12-13 22:06, nappy501 wrote:
>
> Hi,
>
> I turned on my machine this morning and got this:
>
> Code:
> --------------------
> could not find /dev/disk/by-id/ata-MAXTOR_STM (etc) -part 3
> want me to fall back to /dev/disk/by-id/ata-MAXTOR_STM (etc)-part 3?
> waiting for device /dev/disk/by-id/ata-MAXTOR_STM (etc)-part 3 to appear… not found–exiting to /bin/sh
> $
> --------------------

Boot a live CD of openSUSE. Do “ls -l /dev/disk/by-id/”, and check if the
one you need appears there.

Then edit the grub menu file, and the fstab file (of your real system, not
the live), and make sure the names are exactly correct in the appropriate
lines.

I can’t do this for you, but be warned: a single wrong letter and system
will not work.

–
Cheers / Saludos,

Carlos E. R.
(from 11.2 x86_64 “Emerald” at Telcontar)

7

Which means that the symlinks have been created successfully by the udev daemon of the Live CD, not that they will be available to the installed system. IMHO mounting by /dev/disk/by-id has only advantages while cloning hard disks. In most other cases, mounting by UUIDS (not /dev/disk/by-uuid) is safer (unless you’re going to duplicate hard disks). I have less problems with that method and always explicitely choose to mount partitions by UUIDS during openSUSE setup. It’s possible that it will work next time you boot openSUSE. Device symlinks might be sometimes missing. It doesn’t necessarely mean the the hard disk is damaged (although it can not be excluded either). More likely the partition table couldn’t be read properly or entirely for some reason.

8

On 2010-12-14 06:36, please try again wrote:
>
> robin_listas;2265698 Wrote:

> Which means that the symlinks have been created successfully by the
> udev daemon of the Live CD, not that they will be available to the
> installed system.

True enough. Any of id, uuid, labels, paths… depend on the symlinks being
created. And the sda nodes depend on the kernel finding them in the same
order, which in my case they don’t always.

I use labels.

–
Cheers / Saludos,

Carlos E. R.
(from 11.2 x86_64 “Emerald” at Telcontar)

9

Hi,
Thank you for your responses.
I downloaded and burned the live CD. It worked but was very slow and in fact seized up altogether when I was replying tonight. I was going to post my results, but it was all lost.

However I went back to Fedora, (ignoring the requests to install 63 updates, as I have never used it) and tried to see the partitions there. I got see them, but not the contents. After some searching on the internet I created a new user with the UID 1000 and logged into that. I was then able to see the contents of the partitions. So my /home is presently been copied onto an external hard drive. I am having trouble finding my Thunderbird profile, but shall copy that tomorrow.

I have to come clean and say when I set up the machine for some reason my root partition wasn’t big enough, see here:
Filling up root

I’m not sure why as it is not the first time I had installed SUSE or another distro, I did manage to clear a few MB’s, but it has been an ongoing problem and when I had time I was going to change the partitions and reinstall.

The day before yesterday I was told to install a security update on the kernel, but it failed because it needed 49 MB, when I only had 34 MB. So the next day I was going to do the backups and re install SUSE. However, I couldn’t get SUSE to boot up. I suspect this had something to do with it. So I suppose if you mistreat linux enough it will break.

I will post the results of your suggestions tomorrow.

10

I suspect your /tmp directory is full. Assuming your openSUSE root partition is sda3, under Fedora, try :

su -l
mount /dev/sda3 /mnt
ls -la /mnt/tmp

?

If it’s full of stuff, remove it:

rm -r /mnt/tmp/*
  • Before doing that, as you said openSUSE partitions were already mounted, please type:
df -hl
11

On 2010-12-15 04:06, nappy501 wrote:
>
> Hi,
> Thank you for your responses.
> I downloaded and burned the live CD. It worked but was very slow and
> in fact seized up altogether when I was replying tonight. I was going
> to post my results, but it was all lost.

Most of us have a live available, new or old. I could make do with the
installation dvd for this test.

> The day before yesterday I was told to install a security update on the
> kernel, but it failed because it needed 49 MB, when I only had 34 MB.
> So the next day I was going to do the backups and re install SUSE.
> However, I couldn’t get SUSE to boot up. I suspect this had something
> to do with it.

There is a good chance that yes, it could be the cause.

–
Cheers / Saludos,

Carlos E. R.
(from 11.2 x86_64 “Emerald” at Telcontar)

12

Got this here:

[root@localhost /]# fsck /dev/sda3
fsck 1.41.4 (27-Jan-2009)
e2fsck 1.41.4 (27-Jan-2009)
/dev/sda3: clean, 203608/1026000 files, 3882311/4096000 blocks
[root@localhost /]#

I couldn’t get sda5 to work as it said it was busy, but I couldn’t see that it was.

13

Well here it is:

[root@localhost ~]# df -hl
Filesystem            Size  Used Avail Use% Mounted on
/dev/sda7              15G  3.9G  9.8G  29% /
tmpfs                 220M  1.5M  218M   1% /dev/shm
/dev/sda15            7.5G  146M  7.0G   3% /media/distro
/dev/sda5              29G   25G  3.1G  89% /media/disk
/dev/sda13            8.9G  1.8G  6.7G  21% /media/_
/dev/sda1              15G  7.4G  6.9G  52% /media/disk-1
/dev/sda11            9.6G  150M  9.0G   2% /media/distro_
/dev/sda12             11G  155M  9.9G   2% /media/distro__
/dev/sda14            7.5G  146M  7.0G   2% /media/distro___
/dev/sda10            9.7G  151M  9.0G   2% /media/distro____
/dev/sda2              99M   14M   81M  15% /media/boot
/dev/sdb1              39G   33G  6.0G  85% /media/disk-3
/dev/sda9             9.6G  151M  9.0G   2% /media/distro_____
/dev/sda8              11G  155M  9.6G   2% /media/distro______
/dev/sdh1              30G  8.6G   22G  29% /media/disk-4
/dev/sdh2             118G   36G   77G  32% /media/disk-5
/dev/sda3              16G   15G   35M 100% /mnt
[root@localhost ~]# rm -r /mnt/tmp/*
rm: descend into directory `/mnt/tmp/hsperfdata_nappy'? y
rm: remove regular file `/mnt/tmp/hsperfdata_nappy10848'? y
rm: remove directory `/mnt/tmp/hsperfdata_nappy? y
rm: descend into directory `/mnt/tmp/kde-nappy? y
rm: remove symbolic link `/mnt/tmp/kde-nappy/ksycoca'? y
rm: remove directory `/mnt/tmp/kde-nappy? y
rm: remove directory `/mnt/tmp/kde-root'? y
rm: descend into directory `/mnt/tmp/ksocket-nappy'? y
rm: remove socket `/mnt/tmp/ksocket-nappy/klauncherMT5590.slave-socket'? y
rm: remove regular file `/mnt/tmp/ksocket-nappy/secret-cookie'? y
rm: remove directory `/mnt/tmp/ksocket-nappy/artsd-samples'? y
rm: remove directory `/mnt/tmp/ksocket-nappy'? y
rm: descend into directory `/mnt/tmp/orbit-nappy? y
rm: remove socket `/mnt/tmp/orbit-nappy/linc-2b98-0-1be0f5397e855'? y
rm: remove socket `/mnt/tmp/orbit-nappy/linc-1560-0-73d26a0aa06be'? y
rm: remove directory `/mnt/tmp/orbit-nappy'? y
rm: remove directory `/mnt/tmp/pulse-bkadverunGJA'? y
[root@localhost ~]# y
-bash: y: command not found
[root@localhost ~]# ls -la /mnt/tmp
total 116
drwxrwxrwt  5 root   root  98304 2010-12-15 16:15 .
drwxr-xr-x 21 root   root   4096 2010-12-12 20:54 ..
drwx------  2 nappy1 users  4096 2010-12-12 20:54 .esd-1000
drwxrwxrwt  2 root   root   4096 2010-12-12 20:54 .ICE-unix
drwxrwxrwt  2 root   root   4096 2010-12-12 20:54 .X11-unix

I am going to try rebooting now…

14

That’s the problem.

15

On 2010-12-15 17:36, nappy501 wrote:

> I couldn’t get sda5 to work as it said it was busy, but I couldn’t see
> that it was.

Please, write the exact commands you did and the results you got, not your
comments.

–
Cheers / Saludos,

Carlos E. R.
(from 11.2 x86_64 “Emerald” at Telcontar)

16

Sorry.

 [root@localhost ~]# cd /
[root@localhost /]# umount /dev/sda3
umount: /dev/sda3: not mounted
[root@localhost /]# umount /dev/sda5
umount: /dev/sda5: not mounted
[root@localhost /]# fsck /dev/sda3
fsck 1.41.4 (27-Jan-2009)
e2fsck 1.41.4 (27-Jan-2009)
/dev/sda3: clean, 203595/1026000 files, 3882295/4096000 blocks
[root@localhost /]# fsck /dev/sda5
fsck 1.41.4 (27-Jan-2009)
e2fsck 1.41.4 (27-Jan-2009)
/dev/sda5: clean, 268823/3817472 files, 6427366/7620834 blocks
[root@localhost /]#

Worked this time.

17

On 2010-12-15 23:06, nappy501 wrote:
> Worked this time.

Your problem is that “/dev/sda3” is 100% full, as “please try again” said.

If that happened while you where updating the system, your system is
probably missing important parts.

–
Cheers / Saludos,

Carlos E. R.
(from 11.2 x86_64 “Emerald” at Telcontar)

18

SORTED
It has taken some time, probably because I had to keep going away and coming back to this as I didn’t have a long stretch of time today.

I used Gparted live to delete a partition (I had 15) and move the other partitions around, my root is now 21.5 GB.

I used my original Suse 11.1 disk to try and rescue then repair the system, but they didn’t work. I then installed. The installation said it was only going to format the root partition.

For the first time since this all started, I am online on my machine. My root is only 4.3 GB now instead of 16 GB. I shall keep an eye on it to see what I manage to fill it up with.

My home drive survived the repartitioning and installation.

Can I say thank you very much for your help and advice. It all got a lot less stressful when I was able to copy my /home somewhere else, then it didn’t matter if the partition was lost.

Regards

19

On 2010-12-18 02:06, nappy501 wrote:

> I used my original Suse 11.1 disk to try and rescue then repair the
> system, but they didn’t work. I then installed. The installation said
> it was only going to format the root partition.

One possible rescue method is to install the same version on top without
reformatting. It does not delete existing things that don’t exist on a new
install - the rest is replaced or added.

I’m glad that it worked, anyhow :slight_smile:

–
Cheers / Saludos,

Carlos E. R.
(from 11.2 x86_64 “Emerald” at Telcontar)