Restore Backup File

[Simon Rutherford]

Hi

I hope someone will be able to point me in the correct direction with an issue I'm having.

I rebooted an esos system yesterday and when it tried to load the usb stick it was no longer detected.  I moved it to a new server and the same response - looks like the usb stick was completely dead.

I've created a new esos usb stick, and it runs fine, however I'm obviously missing the config I previously had.

I had followed the instructions under https://github.com/parodyne/esos/wiki/25_Routine_Maintenance and had a cron job that backed up the config each day, so I have a copy of the etc and var folders, however, I don't seem to be able to find any information about using this information to restore the config if you need to recreate the boot disk.

I've mounted the new disk in linux and don't seem to see a way to copy any of that config back onto the disk.

I'm not sure if I'm missing something, but if someone could point me in the right direction that would be great :)

Thanks

Simon

[Marc Smith]

Hi Simon,

So you have the new USB flash drive mounted on a non-ESOS Linux system
right now? If so, not sure which file system(s) you have mounted
already, but do this:
- Mount the LABEL=esos_conf file system (eg, partition # 2) somewhere.
- Then you can change the current working directory to the root of
where you mounted the file system above.
- Then run this command: tar xvfzp /tmp/esos_conf_backup.tar.gz
- That should extract the tarball into your esos_conf file system
root, check that the etc/ and var/ directories are created there.
- Finally, unmount the file system and plug it into your ESOS server.

Let us know how it goes.


--Marc

> --
> You received this message because you are subscribed to the Google Groups
> "esos-users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to esos-users+...@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.

[Simon Rutherford]

Hi Marc

Thanks for the response.

I was afraid of that.  I'd managed to copy the scst.conf file across and load it, but scst then errored because it didn't appear to be loading the vdisk properly.  

I've followed your instructions and I'm getting the same thing now;  scst fails halfway through loading and dmesg shows the following:

[  186.768337] dev_vdisk: Registering virtual vdisk_fileio device BACKUP01 (ROTATIONAL)

[  186.768410] dev_vdisk: ***ERROR***: filp_open(/mnt/vdisks/draas01/backup01) failed: -2

[  186.768412] scst: ***ERROR***: New device handler's vdisk_fileio attach() failed: -2

After which, it just unloads all of the modules.

It looks as though the LVM volume that is used isn't mounting the lvm volume properly for /mnt/vdisks/draas01 listed above.

I've had a look at the /etc/fstab and the line that appears to be failing is:

LABEL=draas01 /mnt/vdisks/draas01 xfs defaults 1 1

I created the /mnt/vdisks/draas01 location at the command line and it once again failed with:

mount: mounting LABEL=draas01 on /mnt/vdisks/draas01 failed: No such file or directory

So I'm guessing something has gone missing with my LVM config that would be causing this?

Simon

[Marc Smith]

Yeah, run 'lvdisplay' and see if your LV is listed. If that doesn't work, check 'vgdisplay' and 'pvdisplay' output.

What storage does LVM run on top of?

Marc

[Simon Rutherford]

I've ran an lvdisplay and I can see the logical volume which i believe to be /dev/cl00/home - this is purely based on the size of it at 57.25Tib.

  --- Logical volume ---

  LV Path                /dev/cl00/home

  LV Name                home

  VG Name                cl00

  LV UUID                ckRvIZ-H87x-o1Hj-CD40-hEQJ-L4aO-L0I2OI

  LV Write Access        read/write

  LV Creation host, time localhost.localdomain, 2017-02-20 16:53:47 +0000

  LV Status              available

  # open                 0

  LV Size                57.25 TiB

  Current LE             15008615

  Segments               15

  Allocation             inherit

  Read ahead sectors     auto

  - currently set to     256

  Block device           250:4

If I do an lvscan, it is showing as Active as well, so it looks like it should be working.

[root@localhost vdisks]# lvscan

  ACTIVE            '/dev/cl00/swap' [4.00 GiB] inherit

  ACTIVE            '/dev/cl00/home' [57.25 TiB] inherit

  ACTIVE            '/dev/cl00/root' [50.00 GiB] inherit

  ACTIVE            '/dev/cl/swap' [4.00 GiB] inherit

  ACTIVE            '/dev/cl/home' [224.36 GiB] inherit

  ACTIVE            '/dev/cl/root' [50.00 GiB] inherit

The underlying storage is a number of D2700 HP shelves, each organised into 3.82TB logical volumes at the array level.

To unsubscribe from this group and stop receiving emails from it, send an email to esos-users+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to a topic in the Google Groups "esos-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/esos-users/R8qG8O4Y7aI/unsubscribe.
To unsubscribe from this group and all its topics, send an email to esos-users+unsubscribe@googlegroups.com.

[Marc Smith]

Those logical volumes look like they belong to an OS install... swap, home, etc?

Is that right?

If so, check the output of dmesg after trying to mount the XFS file system and see what it says.

Marc

--

You received this message because you are subscribed to a topic in the Google Groups "esos-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/esos-users/R8qG8O4Y7aI/unsubscribe.

To unsubscribe from this group and all its topics, send an email to esos-users+...@googlegroups.com.

[Simon Rutherford]

Hi Marc

I just wanted to come back and say thanks for you help.  In the end, as this was non-production data, I decided just to recreate it from scratch in the interests of getting the service running from the array back online.  

If I used a disk rescue tool, I could see the partition named correctly within the cl00 volume, but didn't manage to get it back to the point I could bring it online.

Thanks Again

Simon