Thread Info
  • State Verified Answer
  • Locked Locked
  • Replies 4 replies
  • Subscribers 27 subscribers
  • Views 7857 views
  • Users 0 members are here
Options
Suggested
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Problem loading configuration file: Backup cannot be restored on current device

Emiel Zuidhof

Some background info
I've been a happy Sophos user for many years now, always used the software versions ( running on VMWare ESXi ). Unfortunately my raid setup failed me, so I'm in the process of recreating all my VM's including my virtual Firewall. I had set it up to e-mail the configuration on a weekly basis, so I thought it would be a walk in the park.... The last time I had to install Sophos, I started with Firmware: SFOS 17.0.6 MR-6

The annoying error
It keeps telling me: "Backup cannot be restored on current device" no matter what I do... The 'why' would be sooooo helpful so I could perhaps fix it! I briefly tried looking in log files ( /var/tslogs ) but could not find a relevant file and it’s a bit to much to open them all.

Things I already tried:

  • I downloaded the latest ISO file from the Sophos website ( SFOS 17.5.8 MR-8 ), installed it no problem, but could not recover the configuration. I noticed I wasn't up to date yet, my latest backup config said it was made by SFOS 17.5.6 MR-6, I searched: That ISO could not be found nor was there any way to downgrade.
  • Fortunately I found an old copy of my original installation ISO ( 17.0.6 MR-6 ), installed that, same error. Because I was already a little annoyed I did not payed attention, the checkbox to update to the latest version was checked… But this time I could choose to downgrade, I tried the first and last configuration backups with SFOS 17.0.6 MR-6, same error
  • Because I could not rule out something went wrong with the up- or down- grade, I started with another fresh installation of SFOS 17.0.6 MR-6, this time unchecked the checkbox. Unfortunately, same error

After some more searching on Google and the Sophos community I found https://community.sophos.com/kb/en-us/124585 which stated some possibilities why it could go wrong. This made me thinking, I was unsure of how many NICs I had configured, I remembered having played around with some setups. Because of my current setup I was sure I had at least 3 active NICs. So I retraced the above mentioned steps with 3, 4 and even 5 NICs… Same results.

Another thing I'm unsure about, is the RED interface, I've had it running in the past, but for the last year or so it's turned off, the KB doesn't clearly state if it counts a RED interface... I used the same serial number to register all installation as I had used within the backup configuration. Because I started with SFOS 17.0.6 MR-6, there was not yet the possibility to encrypt the configuration, so that should not be factor. In my last attempt I noticed that at step one ( entering a new Admin password ) there is an options to reload a config file, tried many backup files: same error

Anybody with suggestions, or things to try I haven thought of, would be much appreciated.

 

Thanks!



This thread was automatically locked due to age.
  • 0

    Emiel,

    are you using SW verion or Virtual appliance image?

    I have 17.5 MR7 gpg package, so I can try to send you the file and you can try to upload the gpg package.

    Inside the log, try to use tail -f /var/tslog *.log | grep "upload"

    Otherwise, during the upload of the configuration try ail -f /var/tslog *.log | grep "error"

  • +1 in reply to lferrara

    You should, if you do not know the exact number of interfaces, pump your VM until 8 Interfaces. 

    Actually the number of interfaces present in the Kernel matters in XG. 

    So get as many interfaces as you can get into the VM and retry the process. 

     

    You can check the applog.log in case of restore, if you see the reason. It should be logged there. 

  • 0 in reply to lferrara

    Luk,

    Thanks for the fast reply!

    I think it's the Virtual appliance image, because it's on my own hardware ( ESXi server ).

    If you could send it, that would be great, I'm open to try everything in order to prevent manually setting the whole thing up again.

    I tried your commands

    tail -f /var/tslog *.log | grep "upload"

    I'm far from a unix expert, but I figured you possibly made a typo?
    Because it kept saying:
    opcode upload_restorefile csc/1.2
    tail: read error: Is a directory

    I changed the command to: tail -f /var/tslog/*.log | grep "upload" and it got me the folling info during a restore attempt:
    opcode upload_restorefile csc/1.2
    Message    Oct 13 11:14:28 [upload_restorefile:3375]: Child exited with status 1
    Warning    Oct 13 11:14:28 [upload_restorefile:3375]: action with nofail failed
    opcode upload_restorefile csc/1.2

    tail -f /var/tslog/*.log | grep "error"

    Gave no results during the upload.

    I will play around with these commands a little more with different scenarios ( change # NICs, try different restore files, reload at step 1 or after initial setup is done )

  • 0 in reply to LuCar Toni
    Hi LuCar,
     
    Again thanks for the fast reply, call me amazed!
     
    Steps I took:
    • Powerd off VM
    • Threw in a total of 8 NICs
    • Reinstalled with 17.0.6 MR-6
    • At step one, tried loading config, this generated the follwing error:
      Message    Oct 13 11:46:18 [upload_restorefile:3375]: Child exited with status 1
      Warning    Oct 13 11:46:18 [upload_restorefile:3375]: action with nofail failed
      opcode uploade_restorefile csc/1.2
      Oct 13 11:46:22 upload_restorefile: Different timezone is found in backup hence changing timesone before reboot
    At least a reason! So I completed the initial setup, after reboot tried to upload my config again:
     
    SUCCESS!!
     
    I think the solution had to be the 8 NICs!
     
    Thanks for pointing it out!!!