Save data behind a hardware RAID controller, DELL T7910

[Anonymous]

Hi SupportForum,

I am working on a system that is setup behind a hardware RAID controller, AVAGO MegaRAID 9361-8i. There are two virtual drives, one is RAID5 with all the data, and one is RAID0 with the OS (windows server 2012).
The The RAID5 consists of 3 HDDs (each 3.8TB) and has a size of 7.6 TB and the RAID0 consists of 1 SSD.

The SSD failed and now I cannot startup my system to save the data. As I had good experience with Rescuezilla on my private Laptop with rescuezilla I made a USB stick and booted into it using UEFI. The rest of the system uses Legacy boot.

Right now I cannot see the virtual drives in the file manager. Is it possible to access files behind a hardware RAID controller?

As I am not that familiar with maintaining servers and only got into because I had to, any help is appreciated. That said, I wo not know why the system is set up like that.

Thank you for any help!

Best,
Nils

[Anonymous]

PS. I started a backup with rescuezilla on a 4TB drive. But since it is substantially smaller, I fear that will not work as I hope.

[Rescuezilla]

In my experience, hardware RAID is often configured in the BIOS and provides a transparent mechanism from the operating system (Linux in the case of Rescuezilla), because the hardware RAID controller exposes the underlying disk as a single disk in the operating system.

In that situation you can use Rescuezilla to backup,restore and clone without issue, because Rescuezilla doesn't require any special handling of the situation.

In your situation the RAID0-configured single SSD has failed. This is a bit confusing to me, because typically RAID0 is mirroring the same data across 2 or more disks. Because that redundancy would make a single SSD failing easier to recover from.

I would have expected you to be able to see the hardware RAID5 disks from Rescuezilla, because they're in a healthy state according to your post. I would have expected your RAID0 SSD to potentially be visible but maybe not due to the hardware failure you mentioned.

If a disk in your RAID5 environment has failed, you typically need to swap out the bad disk and rebuild the RAID using the provided RAID utilities (in this case with hardware RAID card it's likely from Windows Server 2012 not the BIOS).

The real question to identify from Rescuezilla is why the RAID0 single SSDs failed. Is it viewable from the BIOS. Can Rescuezilla see it (eg, from GParted). There are internal Linux kernel logs and ways to probe hardware from Rescuezilla that can more deeply identify where your SSD failed.

[Rescuezilla]

PS. I started a backup with rescuezilla on a 4TB drive. But since it is substantially smaller, I fear that will not work as I hope.

Since you say your local hardware RAID5 environment produces a 7.6TB virtual disk, this will not work with Rescuezilla v2.4.2, because Rescuezilla cannot to restore to smaller disks except under some select situations.

What I suspect you may have done is started to create a backup of a single underlying disk constituting the RAID5 setup. If so, this is not a good idea and should not be done.

Cloning your storage data from the 7.6TB virtual disk that's exposed as a single virtual disk to an 8TB external disk is an OK idea though.

I'd focus on identifying the issue with main SSD and trying to get back into Windows Server 2012 though.

[Anonymous]

Thank you for your advice!
The RAID0 also confused me. But as I did not set up the system, I cannot explain it. I can see all the virtual disks in gparted but not in the file manager to actually access the data.
As I can see the SSD I will try to look into that. Do you have any specific advice on that?
My plan is to :
1. order a 8 TB disk to backup the data with rescuezilla as an image. Because rescuezilla can see the RAID controller. Just to be on the safe side.
2. Make an image of the SSD.
3. Find out the reason why the SSD failed.
4. Try to mirror the SSD on a new one and insert the new disk into the system.

If anything I try is not so smart, please let me know.

[Anonymous]

I tried to clone the SSD but got this error message:

Fehler beim Wiederherstellen der Sicherungskopie: Traceback (most recent call last):
  File "/usr/lib/python3/dist-packages/rescuezilla/clone_manager.py", line 111, in do_clone_wrapper
    self.do_clone()
  File "/usr/lib/python3/dist-packages/rescuezilla/clone_manager.py", line 146, in do_clone
    is_success, message = self.backup_manager.start_backup(selected_drive_key=self.image.long_device_node,
  File "/usr/lib/python3/dist-packages/rescuezilla/backup_manager.py", line 104, in start_backup
    return self.do_backup()
  File "/usr/lib/python3/dist-packages/rescuezilla/backup_manager.py", line 200, in do_backup
    process, flat_command_string, failed_message = Utility.run("Saving " + raid_device_query_filepath,
  File "/usr/lib/python3/dist-packages/rescuezilla/utility.py", line 502, in run
    process = subprocess.run(cmd_list, encoding='utf-8', capture_output=True, env=env)
  File "/usr/lib/python3.10/subprocess.py", line 503, in run
    stdout, stderr = process.communicate(input, timeout=timeout)
  File "/usr/lib/python3.10/subprocess.py", line 1152, in communicate
    stdout, stderr = self._communicate(input, endtime, timeout)
  File "/usr/lib/python3.10/subprocess.py", line 2041, in _communicate
    stdout = self._translate_newlines(stdout,
  File "/usr/lib/python3.10/subprocess.py", line 1029, in _translate_newlines
    data = data.decode(encoding, errors)
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xdc in position 42: invalid continuation byte

[Anonymous]

I do not understand what the error means. Can someone explain this to me?

[Rescuezilla]

Sorry the "UnicodeDecodeError: 'utf-8' codec can't decode byte 0xdc in position 42" is a bug that impacts a small number of users with Rescuezilla v2.4.2 because Rescuezilla is not correctly handling something (in this case reading some RAID information, and failing to process it). The bug will be resolved in a future update.

Until then, to continue your particular plan you may prefer to use Clonezilla either from its live USB environment or within Rescuezilla itself from a terminal. Warning though, Clonezilla has much more confusing menus than Rescuezilla.