Question

FATAL Imaging thread 'svm' failed with reason [none]

  • 16 November 2020
  • 20 replies
  • 2705 views

Hi all,

I am trying to install Nutanix Community edition on a Hpe DL360 Gen 9 with the latest firmware.

This is the configuration:

256GB RAM
HPE Ethernet 1Gb 4-port 331i Adapter
Smart HBA H240 with:

  • 3x  1TB RAID 0 SSD for DATA DISK
  • 1x 1TB RAID 0 SSD for CVM Boot Disk
  • 1x USB Sandisk Ultra 3.0 for Hypervisor Boot

After selecting the discs after a few minutes I get this error:

Any ideas?

 

Thanks


20 replies

Badge

Hi Carlos,

 

I’m getting exactly the same error. Did you manage to find a fix/solution


Thanks​​​​​​​

Hi Mohideenyassarvm,

I have tried several installations but I have not found the solution. It seems that the installation is not so easy 

 

Badge

I’m trying to install this in a 4 Node server. 3 Nodes i was able to install Nutanix CE successfully. However only one node is giving this error. Difference between this node and other node is that the not working node has 4x1TB HDD, 1x400GB SSD and working nodes are having 3x1TB HDD , 1x400 GB HDD and 1x 400GB SSD

Thanks
Mohideen

 

 

Userlevel 2
Badge +1

I have the same error on my Intel NUC7

 

Intel NUC 7 i5 NUC7i5BNH, Intel Core i5-7260U with storage shown below (BIOS is current)

  • SanDisk - SDCZ430-016G-G46 16GB Ultra Fit USB 3.1 Flash Drive
  • 120 GB SSD Drive
  • Crucial P2 1TB 3D NAND NVMe
  • 40 GB Memory

 

Userlevel 2
Badge +1

To clarify, the “120 GB SSD Drive” is my SATA drive.

 

Badge +1

I am having exactly the same issue. Any ideas how to fix this?

Badge +1

For me it turns out it was an issue witht the RAID controller. I used the default PERC H710, which did not work. Swapping it for an LSI 2008 / 9211-8i made the difference.

Have this same issues with Dell R230 

4x 8TB on raid 5 

2x SSD 250GB 

in 60/2430 have always this error 

Any idea how to fix this ? IN other server install no have any problem 

Hello. I have the same issue. 

Any ideas to fix it?

I had this same problem on one of the 4 nodes in a Dell C6220II. The installer selected /dev/sdc as the CVM disk and /dev/sdd as the AHV disk. I wanted the AHV disk to be /dev/sdc and the CVM disk to be /dev/sdd, but every time I tried to modify during the install process, it would result in this particular error at around 420 or 450 out of 2430. When I accepted the default selection from the installer, it completed w/o error. I tested both SSDs successfully - both were perfectly operational, so not a disk problem. Very weird...

I had the same problem with installation ce-2020.09.16 on the Dell R730.

Before I was using the BIOS. But when I switched to EFI - 3 nodes were installed without any problems.

I have the same problem.

Any solution?

Hi Guys,

I found a quick and dirty workaround for this issue.

It looks like the partition marker for the CVM partition is missing at the runtime of the imaging.

 

I’ve changed the line 603 of the svm.py and assign a value of my CVM boot partition ‘sdb1’ to the self.boot_part variable to avoid the call of find_nutanix_partition.

Please try this one and report your success of fail.

Regards,

Roman

Hi guys.

 

Today, I got faced into same problem.

 

devices I used as following below:

  • HPE DL360 Gen8
  • HDD 450GB  × 2 
  • SSD  320GB  ×    1

and one USB that formatted for Nutanix Comity Edition.

 

Then at first, the install display comes up like this;

 

then installation ended up with failure reason with as everyone mention “FATAL Imaging thread 'svm' failed with reason [none]” 

 

2nd trial

I added one more HDD and then tried installing again.

Display was like below; letter “I” newly showed up, which you can’t type in with your keyboard.  You can do so with “h”, “c” and “d” tho.

All letters were automatically set and I did nothing with those.

 

finally installation got done.

please try this tip and hope this helps you solve the problem!!

 

 

 

 

 

I added one more HDD and then tried installing again.

Display was like below; letter “I” newly showed up, which you can’t type in with your keyboard.  You can do so with “h”, “c” and “d” tho.

All letters were automatically set and I did nothing with those.

 

finally installation got done.

please try this tip and hope this helps you solve the problem!!

 

 

 

 

Cheers Ayakyta, after failing several times I added another hard drive and it worked a treat!  Thanks for sharing.

 

I followed this blog, same server similar hardware and regardless of how I setup the drives, I still get a failed install.  https://kenmoini.com/post/2021/10/deploying-nutanix-ahv-ce/.  I’ve updated the BIOS and Firmware on the server to current.  I’ve tried different ISO media for ce-2020.09.16, booted from CD and USB same error.  

 PERC H710 Mini (Embedded)

1 x RAID 0 - SAS 232Gb - Hypervisor

1 x RAID 5 - SAS 1.3TB - Data 

Intel C600/X79 series chipset 6-Port SATA AHCI Controller

1 X SATA 200Gb SSD - CVM

Any recommendations, anything I haven’t tried?

 

Badge

Hi Guys,

I found a quick and dirty workaround for this issue.

It looks like the partition marker for the CVM partition is missing at the runtime of the imaging.

 

I’ve changed the line 603 of the svm.py and assign a value of my CVM boot partition ‘sdb1’ to the self.boot_part variable to avoid the call of find_nutanix_partition.

Please try this one and report your success of fail.

Regards,

Roman

This workaround fixed for me.. Thanks 

Hi Guys,

I found a quick and dirty workaround for this issue.

It looks like the partition marker for the CVM partition is missing at the runtime of the imaging.

 

I’ve changed the line 603 of the svm.py and assign a value of my CVM boot partition ‘sdb1’ to the self.boot_part variable to avoid the call of find_nutanix_partition.

Please try this one and report your success of fail.

Regards,

Roman

Long shot you will see this after all this time - but gotta try.

I’m finding this file in /root/phoenix/svm.py - and I can change everything no problem, however a reboot causes the file to go back to original state.

Where/when do you change it - or how do you resume the installation after it has failed?

 

Thanks in advance

Badge +1

Hi Guys,

I found a quick and dirty workaround for this issue.

It looks like the partition marker for the CVM partition is missing at the runtime of the imaging.

 

I’ve changed the line 603 of the svm.py and assign a value of my CVM boot partition ‘sdb1’ to the self.boot_part variable to avoid the call of find_nutanix_partition.

Please try this one and report your success of fail.

Regards,

Roman

Long shot you will see this after all this time - but gotta try.

I’m finding this file in /root/phoenix/svm.py - and I can change everything no problem, however a reboot causes the file to go back to original state.

Where/when do you change it - or how do you resume the installation after it has failed?

 

Thanks in advance

Hi, did you find an answer to your question? When are you planning to edit the file?

Hi Guys,

I found a quick and dirty workaround for this issue.

It looks like the partition marker for the CVM partition is missing at the runtime of the imaging.

 

I’ve changed the line 603 of the svm.py and assign a value of my CVM boot partition ‘sdb1’ to the self.boot_part variable to avoid the call of find_nutanix_partition.

Please try this one and report your success of fail.

Regards,

Roman

Long shot you will see this after all this time - but gotta try.

I’m finding this file in /root/phoenix/svm.py - and I can change everything no problem, however a reboot causes the file to go back to original state.

Where/when do you change it - or how do you resume the installation after it has failed?

 

Thanks in advance

Hi, did you find an answer to your question? When are you planning to edit the file?

Don’t reboot. Once you change the file, run “./ce_installer”, once it is done printing text, type “logout” and it will launch back into the installer, and the change to the file should persist.

Reply