Topic: Upgrade Disk process - trouble mounting old drive

[pssl]

Hi.

Im upgrading from sme 6 to 7.  I'm on step 5 of the process (http://wiki.contribs.org/UpgradeDisk), trying to mount my old server drive.  The device is hdb.  When I enter

Code: [Select]

mount /dev/hdb2 /mnt/tmp
I get

Code: [Select]

/dev/hdb2 looks like swapspace - not mounted
mount: you must specify the filesystem type
fdisk shows that hdb2 is indeed linux swap. 

When I try to mount hdb1, (fdisk showing it as the boot partition, system linux), I get

Code: [Select]

mount: fs type ntfs not supported by kernel.
I'm not sure what to do now.  Any suggestions?

Update: I am able to mount hdb3, but will this partition work with the Upgrade Disk process?

[cactus]

Update: I am able to mount hdb3, but will this partition work with the Upgrade Disk process?

Perhaps you can show us the disk layout of the disk you are trying to mount, to me it seems hdb3 is perhaps the one you need, but it is hard to be sure with the limited amount of information provided.

[pssl]

Hi cactus.

Here's fdisk info for the drive.  Note: I noticed that I had the drive previously connected as a slave and have since moved it to the master position, so the device name is now hda.

Code: [Select]

Disk /dev/hda: 250.0 GB, 250059350016 bytes
/dev/hda1   *           1         288     2313328+  83  Linux
/dev/hda2             289         831     4361647+  82  Linux swap
/dev/hda3             832       30401   237521025   83  Linux


[cactus]

Judging from the size and type, hda3 is the one you need as that will contain the data you want to copy to your new server, but I am unsure why you made this disk master. Why not keep the new drives of your server as master?

[pssl]

OK.  Thanks very much.  I'll give it a go. 

The Upgrade Disk instructions are not clear as to which partition to mount.  Experienced techies would just know, but we who have a less than in-depth knowledge of SME are not sure.  When the instruction specifically mentions hdb2 then we the uninformed blindly follow that instruction, understanding of coarse that the device name might be difference (e.g., hda2 vs hdb2).

Thanks again.

[CharlieBrady]

The Upgrade Disk instructions are not clear as to which partition to mount.

Please go the Bugs link about, create a bugzilla account for yourself, and report the problem as a deficiency in Documentation.

[cactus]

Please go the Bugs link about, create a bugzilla account for yourself, and report the problem as a deficiency in Documentation.

Or get a member of the Documentation Team and start addressing the issues, more information can be found here: http://wiki.contribs.org/SME_Server:Volunteering#Wiki_Team

[raem]

pssl

I have updated the UpgradeDisk wiki article.
Please see if the new instruction is understandable and helps you determine which drive and partition to use. 
The command now refers to /dev/hdxn
which should force users to adequately investigate using the suggested fdisk -l /dev/hdx command.

[pssl]

Ray,

I checked out your changes.  That's great.  That tells the reader exactly which partition to mount.

But now I have another problem...not with the instructions.  I performed the procedure, which quite each to follow once I got the mount thing solved, but nothing got transferred.  After I entered signal-event restore-tape /mnt/tmp the command prompt came back immediately.  I was expecting it to take some time because I have 2.2 gigs of data.  When I checked out the server no users, groups, Ibays, etc. were transferred and no data.  I'm pretty sure I did all the steps.  I was very slow and careful entering the commands and following the steps.

Has anyone any idea what might have gone wrong?  I was so hoping not to have to rebuild the server from scratch. 

[janet]

pssl

It would be good if you posted the exact commands you issued, use the up arrow to review what you actually typed. Small errors can easily be made, and we are chasing red herrings then.

Assuming the commands are OK, and based on your earlier comments, when you issue the mount command I assume you did
mount /dev/hda3 /mnt/tmp

So if you change to /mnt/tmp do you see the contents of the old drive ?
Alternatively use midnight commander, type mc
cd /mnt/tmp
ls -al

You can also check the db setting is correct
config show backup

The golden rule of troubleshooting is to look at the log files eg
/var/log/messages
& other log files, or view them in server manager

[cactus]

It would be good if you posted the exact commands you issued, use the up arrow to review what you actually typed. Small errors can easily be made, and we are chasing red herrings then.

Or use the history command for a list of the last 1000 commands entered, which in turn can be trimmed by porting through head, tail or filtered by grep, e.g.:

Code: [Select]

history | head -10
history | tail -20
history | grep mount

[pssl]

Cactus, Mary,

Thanks for your help.  As you can tell I'm not a linux guy...barely know enough to be dangerous.  But it's been a good experience to plow through this.  The old server was built for me by a company that can no longer support it (lost their linux tech).  So, tag, I'm it.  I worked in Unix for a short time a long time ago but never at the admin level.  So thanks for you patience.

I'll get the info you asked for tonight.

[idp_qbn]

pssl,
Stop for a moment and remember what you set out to do: upgrade from SME6.x to SME7.x.
From what you have written in your two posts on this, I think you actually have two servers.

Forgive me if I have got the wrong end of the stick and I really do not mean to cause offence, but I think you have become very focussed on the resolution of the "upgrade disk" process and lost sight of what started all this.

If your SME6.x box still works and you do have a second server (the SME7.5.1) then the advice I gave in your other post (use the AFFA contrib) is still the best way for a non-linux person. See http://forums.contribs.org/index.php/topic,46708.0.html and http://wiki.contribs.org/Affa

Even I managed that and I am certainly a non-linux person.

Cheers, Ian

[pssl]

Hi.  Sorry it took so long to reply. 

First, idp_qbn: Thanks for your suggestion.  I chose not to go the AFFA route because I had spent time understanding the Upgrade Disk process and was reasonably familiar with it.  Once I understood it it seemed, and was, pretty straight forward...except it didn't seem to work for some reason.  That's what I'm trying to sort out here.

Mary, Cacutus
Here's info from my system.  I don't see anything that looks strange, but then I'm not that knowledgeable so maybe you folks'll pick something up.  I perused the messages file but nothing jumped out at me...then again...

Here's the history of the Upgrade Disk commands.  I left all commands in, even the goof ups just in case they are significant.

Code: [Select]

  383* mount /dev/hda3 /mnt/tmp/[B
  384  ls hda3
  385  ls /mnt/tmp
  386  fdisk -l | less
  387  mount -t vfat /dev/sdc /mnt/usb
  388  mount -t ntfs /dev/sdc /mnt/usb
  389  mount -t fat32 /dev/sdc /mnt/usb
  390  mount -t vfat /dev/sdc1 /mnt/usb
  391  cd /mnt/usb
  392  ls
  393  fdisk -l | grep hda > hdafdisk.txt
  394  less hdafdisk.txt
  395  umount usb
  396  umount sdc1
  397  umount /mnt/usb
  398  cd ..
  399  umount /mnt/usb
  400  db configuration setprop backup Progam CopyFromDisk status enabled
  401  signal-event pre-restore
  402  ls /mnt/tmp
  403  signal-event restore-tape /mnt/tmp
  404  ls /mnt
  405  fdisk -l | less
  406  df -h
  407  cd /etc/e-smith
  408  mv templates-custom templates-custom.old
  409  mkdir templates-custom
  410  mv templates-user-custom templates-user-custom.old
  411  mkdir templates-user-custom
  412  signal-event post-upgrade
  413  signal-event reboot

Here's the old server disk info:
ls /mnt/tmp

Code: [Select]

aquota.group
aquota.user
bin
boot
command
dev
etc
home
initrd
lib
lost+found
media
mnt
opt
package
proc
root
sbin
selinux
service
srv
sys
tmp
usb
usr
var

fdisk

Code: [Select]

Disk /dev/hda: 250.0 GB, 250059350016 bytes
/dev/hda1   *           1         288     2313328+  83  Linux
/dev/hda2             289         831     4361647+  82  Linux swap
/dev/hda3             832       30401   237521025   83 
Info about sda, drive one of new server.

Code: [Select]

Disk /dev/sda: 1000.2 GB, 1000204886016 bytes
/dev/sda1   *           1          13      104391   fd  Linux raid autodetect
/dev/sda2              14      121601   976655610   fd  Linux raid autodetect


Code: [Select]

Oct 18 14:15:14 cherubim kernel: SCSI device sda: 1953525168 512-byte hdwr sectors (1000205 MB)
Oct 18 14:15:14 cherubim kernel: SCSI device sda: drive cache: write back
Oct 18 14:15:14 cherubim kernel: SCSI device sda: 1953525168 512-byte hdwr sectors (1000205 MB)
Oct 18 14:15:14 cherubim kernel: SCSI device sda: drive cache: write back
Oct 18 14:15:14 cherubim kernel:  sda: sda1 sda2
Oct 18 14:15:14 cherubim kernel: Attached scsi disk sda at scsi0, channel 0, id 0, lun 0
Oct 18 14:15:14 cherubim kernel: md: bind<sda1>
Oct 18 14:15:14 cherubim kernel: md: bind<sda2>
Oct 18 14:15:14 cherubim kernel: md: could not bd_claim sda1.
Oct 18 14:15:14 cherubim kernel: md: could not bd_claim sda2.
Oct 18 14:20:55 cherubim kernel: SCSI device sda: 1953525168 512-byte hdwr sectors (1000205 MB)
Oct 18 14:20:55 cherubim kernel: SCSI device sda: drive cache: write back
Oct 18 14:20:55 cherubim kernel: SCSI device sda: 1953525168 512-byte hdwr sectors (1000205 MB)
Oct 18 14:20:55 cherubim kernel: SCSI device sda: drive cache: write back
Oct 18 14:20:55 cherubim kernel:  sda: sda1 sda2
Oct 18 14:20:55 cherubim kernel: Attached scsi disk sda at scsi0, channel 0, id 0, lun 0
Oct 18 14:20:55 cherubim kernel: md: bind<sda1>
Oct 18 14:20:55 cherubim kernel: md: bind<sda2>
Oct 18 14:20:55 cherubim kernel: md: could not bd_claim sda1.
Oct 18 14:20:55 cherubim kernel: md: could not bd_claim sda2.
Oct 19 19:53:30 cherubim kernel: SCSI device sda: 1953525168 512-byte hdwr sectors (1000205 MB)
Oct 19 19:53:30 cherubim kernel: SCSI device sda: drive cache: write back
Oct 19 19:53:30 cherubim kernel: SCSI device sda: 1953525168 512-byte hdwr sectors (1000205 MB)
Oct 19 19:53:30 cherubim kernel: SCSI device sda: drive cache: write back
Oct 19 19:53:30 cherubim kernel:  sda: sda1 sda2
Oct 19 19:53:30 cherubim kernel: Attached scsi disk sda at scsi0, channel 0, id 0, lun 0
Oct 19 19:53:30 cherubim kernel: md: bind<sda1>
Oct 19 19:53:30 cherubim kernel: md: bind<sda2>
Oct 19 19:53:31 cherubim kernel: md: could not bd_claim sda1.
Oct 19 19:53:31 cherubim kernel: md: could not bd_claim sda2.

Info about sdb, second drive of new server

Code: [Select]

Disk /dev/sdb: 1000.2 GB, 1000204886016 bytes
/dev/sdb1   *           1          13      104391   fd  Linux raid autodetect
/dev/sdb2              14      121601   976655610   fd  Linux raid autodetect


Code: [Select]

Oct 18 14:15:14 cherubim kernel: SCSI device sdb: 1953525168 512-byte hdwr sectors (1000205 MB)
Oct 18 14:15:14 cherubim kernel: SCSI device sdb: drive cache: write back
Oct 18 14:15:14 cherubim kernel: SCSI device sdb: 1953525168 512-byte hdwr sectors (1000205 MB)
Oct 18 14:15:14 cherubim kernel: SCSI device sdb: drive cache: write back
Oct 18 14:15:14 cherubim kernel:  sdb: sdb1 sdb2
Oct 18 14:15:14 cherubim kernel: Attached scsi disk sdb at scsi2, channel 0, id 0, lun 0
Oct 18 14:15:14 cherubim kernel: md: bind<sdb1>
Oct 18 14:15:14 cherubim kernel: md: bind<sdb2>
Oct 18 14:15:14 cherubim kernel: md: could not bd_claim sdb1.
Oct 18 14:15:14 cherubim kernel: md: could not bd_claim sdb2.
Oct 18 14:15:16 cherubim esmith::event[3699]: Running event handler: /etc/e-smith/events/bootstrap-console-save/S03conf-hostsdb
Oct 18 14:15:16 cherubim esmith::event[3699]: S03conf-hostsdb=action|Event|bootstrap-console-save|Action|S03conf-hostsdb|Start|1287425716 847628|End|1287425716 924002|Elapsed|0.076374
Oct 18 14:15:49 cherubim smbd[5315]: [2010/10/18 14:15:49, 0] passdb/pdb_smbpasswd.c:startsmbfilepwent(241)
Oct 18 14:20:55 cherubim kernel: SCSI device sdb: 1953525168 512-byte hdwr sectors (1000205 MB)
Oct 18 14:20:55 cherubim kernel: SCSI device sdb: drive cache: write back
Oct 18 14:20:55 cherubim kernel: SCSI device sdb: 1953525168 512-byte hdwr sectors (1000205 MB)
Oct 18 14:20:55 cherubim kernel: SCSI device sdb: drive cache: write back
Oct 18 14:20:55 cherubim kernel:  sdb: sdb1 sdb2
Oct 18 14:20:55 cherubim kernel: Attached scsi disk sdb at scsi2, channel 0, id 0, lun 0
Oct 18 14:20:55 cherubim kernel: md: bind<sdb1>
Oct 18 14:20:55 cherubim kernel: md: bind<sdb2>
Oct 18 14:20:55 cherubim kernel: md: could not bd_claim sdb1.
Oct 18 14:20:55 cherubim kernel: md: could not bd_claim sdb2.
Oct 18 14:20:57 cherubim esmith::event[3652]: Running event handler: /etc/e-smith/events/bootstrap-console-save/S03conf-hostsdb
Oct 18 14:20:57 cherubim esmith::event[3652]: S03conf-hostsdb=action|Event|bootstrap-console-save|Action|S03conf-hostsdb|Start|1287426057 561648|End|1287426057 642224|Elapsed|0.080576
Oct 19 19:53:30 cherubim kernel: SCSI device sdb: 1953525168 512-byte hdwr sectors (1000205 MB)
Oct 19 19:53:30 cherubim kernel: SCSI device sdb: drive cache: write back
Oct 19 19:53:30 cherubim kernel: SCSI device sdb: 1953525168 512-byte hdwr sectors (1000205 MB)
Oct 19 19:53:30 cherubim kernel: SCSI device sdb: drive cache: write back
Oct 19 19:53:30 cherubim kernel:  sdb: sdb1 sdb2
Oct 19 19:53:30 cherubim kernel: Attached scsi disk sdb at scsi2, channel 0, id 0, lun 0
Oct 19 19:53:30 cherubim kernel: md: bind<sdb1>
Oct 19 19:53:30 cherubim kernel: md: bind<sdb2>
Oct 19 19:53:31 cherubim kernel: md: could not bd_claim sdb1.
Oct 19 19:53:31 cherubim kernel: md: could not bd_claim sdb2.

grep from /var/log/messages

Code: [Select]

Disk /dev/hda: 250.0 GB, 250059350016 bytes
Disk /dev/sda: 1000.2 GB, 1000204886016 bytes
Disk /dev/sdb: 1000.2 GB, 1000204886016 bytes
Disk /dev/md1: 106 MB, 106823680 bytes
Disk /dev/md2: 1000.0 GB, 1000095219712 bytes
Disk /dev/dm-0: 994.5 GB, 994553364480 bytes
Disk /dev/dm-1: 5536 MB, 5536481280 bytes
Disk /dev/sdc: 1083 MB, 1083801600 bytes

If I want to redo the process I assume I'd have to do a fresh install.  Correct?

[janet]

pssl

All that looks OK.
Something appears amiss with the signal-event restore-tape /mnt/tmp stage of the process.
What is the output of
config show backup

Also you could mount the hda3 partition and see if you can manually copy some files from that drive to the new server hdd's, just to prove the drive can be copied from.

You have listed the old drive content folders, but is the data actually there on the old drive. Take a look at say
ls /mnt/tmp/home/e-smith/files/ibays/ibayname/
and so on.

If nothing was copied during the main restore process you did, which you say is the case in an earlier post, then it should be OK to redo the process.