Unable to mount hard drives after kernel update

[wedgetail]

No problem taking your time, there is more to life than pc's    Both your fdisk -l and the lshw file were helpful revealing bits of information, I think I am getting closer but I don't want to make a fool of myself yet, instead I would like you to do simple test or two.  I assume you have a LiveCD, I don't think it matters which one, can be MiniMe or PClinuxOS.  However it must not be one you have re-mastered or customised, must be one you have downloaded.

Start the LiveCD normally, you can limp along without much configurration, just need to get into konsole and then repeat what you have done before

Code: [Select]

[root@PCLOS2009 ~]# fdisk - l

Just for good measure add

Code: [Select]

[root@PCLOS2009 ~]# df

Post the results in code tags   

[mtonkinson]

Here's the output from a LiveCD i had of PCLinuxOS 2007. I cannot remember the video setting that I had to change to get into the GUI on the live CD so I sent the output of the commands to files (fdisk -l > fdisk07.txt and df > df07.txt) which are attached. In the 2007 LiveCD, I was able to mount both my secondary hard drive and external USB hard drive without problems.

[attachment deleted by admin]

[wedgetail]

A safe bet with livecd,  if you do not remember, would have been the VESA option. You could also easily have found your display data in the the two reports you have already done.   

Anyway quite ok the way you did it and from the text files you can copy  directly into code tags, as the topic may be of help to others as well I will insert, first is the result of df:

Code: [Select]

Filesystem            Size  Used Avail Use% Mounted on
/dev/root            1008M  968K 1007M   1% /
/dev/sdc1             597G  130G  467G  22% /mnt/temp
/dev/sdb2             187G   73G  114G  40% /mnt/temp2

And here is the result of the fisk -l

Code: [Select]

Disk /dev/sda: 640.1 GB, 640135028736 bytes
255 heads, 63 sectors/track, 77825 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

   Device Boot      Start         End      Blocks   Id  System
/dev/sda1               2       25497   204796588+   7  HPFS/NTFS
/dev/sda2               1           1        8001   df  BootIt
/dev/sda3   *       25498       32613    57159270    f  W95 Ext'd (LBA)
/dev/sda5           25498       26006     4088511   82  Linux swap / Solaris
/dev/sda6           26007       29296    26426893+  83  Linux
/dev/sda7   *       29297       32613    26643771   83  Linux

Partition table entries are not in disk order

Disk /dev/sdb: 200.0 GB, 200049647616 bytes
255 heads, 63 sectors/track, 24321 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

   Device Boot      Start         End      Blocks   Id  System
/dev/sdb2               1       24321   195358401    c  W95 FAT32 (LBA)

Disk /dev/sdc: 640.1 GB, 640135028736 bytes
255 heads, 63 sectors/track, 77825 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

   Device Boot      Start         End      Blocks   Id  System
/dev/sdc1               1       77825   625129281    c  W95 FAT32 (LBA)

From the fdisk -l result using the LiveCD I am concluding you have a 'normal' running system, pretty much like my own with internal hard drives and occasionally an external USB hard drive.  My system loads similar in the same manner as seen in above fdisk -l result.

I think you have a module dm_mod  that for some reason is mapping your external USB drive when you don't want it to do so, this is the part I simply can't get my head around.  I am assuming that in the past you have experimented with something which has left artifacts causing the odd behaviour.  I could be very wrong too

There is one more test I would like you to do.  You have a printer connected to USB port #1, would you please move your external drive to USB port #2 which I believe should be free.

Then do fdisk -l and df again and post the results, from your lshw profile information it seems you are sharing the USB port #4 between Iomega CD drive and the hard drive.  It is my vain hope that by moving the USB hard drive 'free' of this sharing to it's own USB port that the pesky mapping will disappear. 

If this fails we need some real experts to have a look.

[mtonkinson]

It does sound like dm_mod is mapping my drives when I don't want it to.

If I type:

Code: [Select]

[root@localhost /]# dmsetup remove_all
then attempt to mount by label and type df, I get:

Code: [Select]

[root@localhost /]# mount -L Elements /mnt/temp
[root@localhost /]# mount -L FAT32SHARE /mnt/temp2
[root@localhost /]# df
Filesystem            Size  Used Avail Use% Mounted on
/dev/sda7              26G  5.7G   19G  24% /
/dev/sda6              25G  754M   25G   3% /home
/dev/sdc1             597G  132G  465G  23% /mnt/temp
/dev/sdb2             187G   73G  114G  40% /mnt/temp2

the output of fdisk -l:

Code: [Select]

Disk /dev/sda: 640.1 GB, 640135028736 bytes
255 heads, 63 sectors/track, 77825 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x1bc30dc0

   Device Boot      Start         End      Blocks   Id  System
/dev/sda1               2       25497   204796588+   7  HPFS/NTFS
/dev/sda2               1           1        8001   df  BootIt
/dev/sda3   *       25498       32613    57159270    f  W95 Ext'd (LBA)
/dev/sda5           25498       26006     4088511   82  Linux swap / Solaris
/dev/sda6           26007       29296    26426893+  83  Linux
/dev/sda7   *       29297       32613    26643771   83  Linux

Partition table entries are not in disk order

Disk /dev/sdb: 200.0 GB, 200049647616 bytes
255 heads, 63 sectors/track, 24321 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x69205244

   Device Boot      Start         End      Blocks   Id  System
/dev/sdb2               1       24321   195358401    c  W95 FAT32 (LBA)

Disk /dev/sdc: 640.1 GB, 640135028736 bytes
255 heads, 63 sectors/track, 77825 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0xacdd9b22

   Device Boot      Start         End      Blocks   Id  System
/dev/sdc1               1       77825   625129281    c  W95 FAT32 (LBA)

After using dmsetup remove_all and mounted, the drives are mounted and accessible from KDE by their labels. Do you still want me to run the test with my external drive connected to USB port 2? Now the question is, how do I disable device-mapper or prevent it from attempting to map my secondary and external hard drive? It would be easier if device-manager never tried to mount these devices in the first place.

[wedgetail]

Yes I would like to see the test on USB2, I am uncomfortable with the string of devices where the hard drive is presently.

[mtonkinson]

here is the results of fdisk -l

Code: [Select]

[root@localhost /]# fdisk -l                                                   

Disk /dev/sda: 640.1 GB, 640135028736 bytes
255 heads, 63 sectors/track, 77825 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x1bc30dc0                     

   Device Boot      Start         End      Blocks   Id  System
/dev/sda1               2       25497   204796588+   7  HPFS/NTFS
/dev/sda2               1           1        8001   df  BootIt   
/dev/sda3   *       25498       32613    57159270    f  W95 Ext'd (LBA)
/dev/sda5           25498       26006     4088511   82  Linux swap / Solaris
/dev/sda6           26007       29296    26426893+  83  Linux               
/dev/sda7   *       29297       32613    26643771   83  Linux               

Partition table entries are not in disk order

Disk /dev/sdb: 200.0 GB, 200049647616 bytes
255 heads, 63 sectors/track, 24321 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x69205244                     

   Device Boot      Start         End      Blocks   Id  System
/dev/sdb2               1       24321   195358401    c  W95 FAT32 (LBA)

Disk /dev/sdc: 640.1 GB, 640135028736 bytes
255 heads, 63 sectors/track, 77825 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0xacdd9b22                     

   Device Boot      Start         End      Blocks   Id  System
/dev/sdc1               1       77825   625129281    c  W95 FAT32 (LBA)

Disk /dev/dm-0: 640.1 GB, 640135028736 bytes
255 heads, 63 sectors/track, 77825 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0xacdd9b22

     Device Boot      Start         End      Blocks   Id  System
/dev/dm-0p1               1       77825   625129281    c  W95 FAT32 (LBA)

Disk /dev/dm-1: 640.1 GB, 640132383744 bytes
255 heads, 63 sectors/track, 77824 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x73696420

This doesn't look like a partition table
Probably you selected the wrong device.

     Device Boot      Start         End      Blocks   Id  System
/dev/dm-1p1   ?      119512      153402   272218546+  20  Unknown
Partition 1 does not end on cylinder boundary.
/dev/dm-1p2   ?       82801      116350   269488144   6b  Unknown
Partition 2 does not end on cylinder boundary.
/dev/dm-1p3   ?       33551      120595   699181456   53  OnTrack DM6 Aux3
Partition 3 does not end on cylinder boundary.
/dev/dm-1p4   *       86812       86813       10668+  49  Unknown
Partition 4 does not end on cylinder boundary.

Partition table entries are not in disk order

Disk /dev/dm-2: 200.0 GB, 200049647616 bytes
255 heads, 63 sectors/track, 24321 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x69205244

     Device Boot      Start         End      Blocks   Id  System
/dev/dm-2p2               1       24321   195358401    c  W95 FAT32 (LBA)
[root@localhost /]#
I'm having troubles attaching the lshw report, I keep getting the error message the the upload file if full. How do I attach the report?

[wedgetail]

Why you can not attach a report can have two reasons as far as I understand:
One, wrong file type. Two, size is over limit and I don't think you would have either as messages would be different I expect. I don't know what is happening.  

The result from you having placed the external USB drive in another port.  The fdisk -l shows we are back to the beginning regarding the device mapper, but what is calling this into action?  Running RAID is one possibility. The dmsetup  is used by pclos, I have it installed but I don't know what it is used for.

I almost run a similar setup to you, a couple of internal hard drives and one external USB hard drive and I do not get the 'extra' drive mapping.  I think we have proved you do not need it either. You do have a 'better' mother board than I have.

Do you have anything in BIOS set that could be RAID related. Perhaps it has something to do with your BootITNG, can I assume you don't want to get rid of that one?  GRUB can easily handle the multibooting but you may be comfortable with what you have.

I have assumed you have not been into any scripts and changed something?

I am very curious about this, but I also have very little experience and understanding of what device mapper is used for, I have looked in the manual but I am not getting any wiser. I only see it as something belonging to RAID operation and I don't think my view is valid.

Perhaps I should have a look through your dmesg, do following and attach to next post:

Code: [Select]

[gert@PCLOS2009 ~]$ dmesg > 091229-mton_dmesg


[Bald Brick]

Why you can not attach a report can have two reasons as far as I understand:
One, wrong file type. Two, size is over limit and I don't think you would have either as messages would be different I expect. 

I just got the same error message as mtonkinson when trying to attach a very tiny picture to a post in another thread, so I suspect the upload file really is full.

[mtonkinson]

Here's another try to send the lshw report and the dmesg report. In these reports I didn't run any scripts. They were generated after a booted PCLinuxOS.

Marc


[attachment deleted by admin]

[wedgetail]

1.. The lshw showed that moving the ext hard drive to the "empty" USB port had no effect on the device mappers work.
2.. The dmesg gave me no ideas either, to me the file looks ok.

Since your setup works albeit not quite as it ought to I don't know what to suggest.

A.. Doing a quick new install is similar to starting up your system via LiveCD and which works normally ie I don't expect the device mapping to show up. This tells me there is something installed/leftover from earlier days and we have not been able to identify what may be the cause.

B.. I don't know how BootIT works and looking at the website, http://www.terabyteunlimited.com/bootit-next-generation.htm did not enlighten me either. I understand that the program is another program managing the selection of bootable systems. I would work towards getting rid of it.  PCLOS built in GRUB will handle boot management, so BootIT is not needed. Since you are using BootIT you are already familiar with the program and having to get involved with a GRUB exercise may not be appealing as well cosmetically BootIT may look better. I am guessing that the GRUB install could be more complicated for you than BootIT.

C.. If Linux is not your working system more experimentation then if I was in your situation I would start all over an do a complete re-install of PCLOS but this will take time because you may have much customisation to do (which I assume you have not diligently recorded, so you may not remember all the changes you have made). You would also need to make sure that all your data collected is saved so you can access this again. You should be starting with more knowledge about what you would want.

Having a working system then neither B or C really is attractive. 

What is your opinion?

[mtonkinson]

Thanks for your help wedgeling,
   I'm not sure what to think. I did take a look at the liveCD that I used to test and found that it doesn't appear to have device-mapper and I believe udev was mapping my drives on the distro disk. I also ran Synaptic last night to see if there were any updates for PCLinuxOS and it appears something that was part of the updates last night fixed the problem with my secondary hard drive since it now works so now only my external hard drive is the one that can't be mounted unless is do a dmsetup remove_all before mounting it.
   How does grub work with multiple operating systems? The reason I ask is I will often try other distro's of Linux and if I decide to upgrade to Windows 7 (Not very likely) or a future version of windows, I usually do a clean install so that I still have the previous working version of windows in the event the upgrade fails or the new version of windows is not stable and I decide to revert back to the previous version. This is one of the reasons I use BootIT NG. Another reason I use BootIT NG is so I have an operating system to boot into so I can do images of the OS partitions as a back-up in the event of a hard drive failure.
   I may try disabling device-manager during boot up to see if my drives are mapped with udev since the distro disk doesn't use device-mapper and it appears device-mapper may have something to do with these problems.

[mtonkinson]

After doing a clean install of PCLinuxOS 2010, this problem has disappeared! I want to thank all the developers who worked on making PCLinux2010 work so well. I didn't encounter any problems after installing PCLinuxOS 2010 and it is VERY stable. 

I also want to thank everyone who provided assistance to me when I first had this problem. I learned quite a bit about linux and troubleshooting problems.