I'm using old versions of Clonezilla/DRBL/kernel and I know that. I'm trying to understand what happens before updating my system.
I have a single-disk image that I'm trying to successfully restore on an HP ProLiant DL360 Gen10 Server. I'm getting the restore to work seamlessly on one hardware, but it is stopping on another hardware with the same configuration. In those cases, the restore stops at the black screen with a blinking cursor just after the DRBL wallpaper display. I can't switch TTY on the Client (nothing happens).
I ran a live Wireshark session on the station with Clonezilla Server edition to monitor what happens. The live capture screenshot I attached shows the moment the initrd-pxe.img transfer over TFTP completes during a successful restore (blue lines), at which point I'm guessing the kernel image deploys in Client RAM. Then we see the multicast packets sent from clonezilla server (pink lines). With the buggy configuration, the feed stops before those pink lines. Those just never appear.
I'm not familiar with the exact kernel deploy or even the drbl/clonezilla flows here, and without getting a terminal access on the Client I can't debug anymore. But I could not find differences in the BIOS configurations of both configurations.
Any idea what is happening or even how to investigate that situation?
Edit 09-Sep-2020: I corrected the problem statement mentioning the 530T as new tests revealed it has nothing to do with that hardware.
Edit 13-Sep-2020: corrected to sate the HW Cconfiga are the same.
"the restore stops at the black screen with a blinking cursor just after the DRBL wallpaper display." -> Maybe the kernel does not support your client machine? If there is no any kernel messsages shown on the screen, maybe it just failed to be loaded...
You can try the latest Clonezilla live first, and if it works well, try to upgrade your DRBL (Clonezilla SE) to the latest supported distribution, e.g., Ubuntu 20.04. Then install and configure DRBL again. Or you can use this mode: https://clonezilla.org/clonezilla-SE/use_clonezilla_live_in_drbl.php
Of course, it's easier to use Clonezilla lite server: https://clonezilla.org/show-live-doc-content.php?topic=clonezilla-live/doc/11_lite_server
Steven
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
My bad, I should have stated that the configurations are actually the same across the 2 Clients.
I'm stumped because it makes no sense. So I'm trying to pinpoint what could be different.
BIOS settings are the most obvious I can think of, but if you have other ideas I'm all ears.
Thanks
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Sorry for the delay in replying. I actually tested multiple other HW with the same config and it appears to be a HW fault in the PXE capabilities of said HW.
I still have no exact idea of what happens when a HW doesn't boot in PXE or how to debug it, but it is no longer my issue to solve (it is open for the OEM to investigate). I'll post back here if I get any useful info I can share on this to help this forum.
Thanks for the time and investment Steven.
👍
1
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi expert folks,
I'm using old versions of Clonezilla/DRBL/kernel and I know that. I'm trying to understand what happens before updating my system.
I have a single-disk image that I'm trying to successfully restore on an HP ProLiant DL360 Gen10 Server. I'm getting the restore to work seamlessly on one hardware, but it is stopping on another hardware with the same configuration. In those cases, the restore stops at the black screen with a blinking cursor just after the DRBL wallpaper display. I can't switch TTY on the Client (nothing happens).
I ran a live Wireshark session on the station with Clonezilla Server edition to monitor what happens. The live capture screenshot I attached shows the moment the initrd-pxe.img transfer over TFTP completes during a successful restore (blue lines), at which point I'm guessing the kernel image deploys in Client RAM. Then we see the multicast packets sent from clonezilla server (pink lines). With the buggy configuration, the feed stops before those pink lines. Those just never appear.
I'm not familiar with the exact kernel deploy or even the drbl/clonezilla flows here, and without getting a terminal access on the Client I can't debug anymore. But I could not find differences in the BIOS configurations of both configurations.
Any idea what is happening or even how to investigate that situation?
Thanks
Ubuntu 14.04.4 LTS with kernel 4.4.0-97
DRBL 2.25.11-drbl1
CLonezilla 3.27.20-drbl1
partimage 0.6.8-2.2ubuntu1
mkpxeinitrd-net 2.4.1-drbl1
tftpd-hpa 5.2-7ubuntu3.1
Edit 09-Sep-2020: I corrected the problem statement mentioning the 530T as new tests revealed it has nothing to do with that hardware.
Edit 13-Sep-2020: corrected to sate the HW Cconfiga are the same.
Last edit: EpicScope 2020-09-13
Updated problem statement.
Ideas anyone?
Anuone knows what happends under the hood?
Thanks!
Last edit: EpicScope 2020-09-09
"the restore stops at the black screen with a blinking cursor just after the DRBL wallpaper display." -> Maybe the kernel does not support your client machine? If there is no any kernel messsages shown on the screen, maybe it just failed to be loaded...
You can try the latest Clonezilla live first, and if it works well, try to upgrade your DRBL (Clonezilla SE) to the latest supported distribution, e.g., Ubuntu 20.04. Then install and configure DRBL again. Or you can use this mode:
https://clonezilla.org/clonezilla-SE/use_clonezilla_live_in_drbl.php
Of course, it's easier to use Clonezilla lite server:
https://clonezilla.org/show-live-doc-content.php?topic=clonezilla-live/doc/11_lite_server
Steven
Hi Steven,
Thanks for chiming in!
My bad, I should have stated that the configurations are actually the same across the 2 Clients.
I'm stumped because it makes no sense. So I'm trying to pinpoint what could be different.
BIOS settings are the most obvious I can think of, but if you have other ideas I'm all ears.
Thanks
Too many possibilities I can image.
BTW, have you tried the method I suggested?
Steven
Sorry for the delay in replying. I actually tested multiple other HW with the same config and it appears to be a HW fault in the PXE capabilities of said HW.
I still have no exact idea of what happens when a HW doesn't boot in PXE or how to debug it, but it is no longer my issue to solve (it is open for the OEM to investigate). I'll post back here if I get any useful info I can share on this to help this forum.
Thanks for the time and investment Steven.