Fail to boot proxmox on old dell r620

[ilia987]

i made a fresh install on raid1(2*200 ssds)
and after install stage i have the following hang screen:

Loading Linux 5.4.34-1-pve ...
Loading initial ramdisk ...

tried to upgdate-grub and received the output :

Generating grub configuration file ...
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
Found linux image: /boot/vmlinuz-5.4.34-1-pve
Found initrd image: /boot/initrd.img-5.4.34-1-pve
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
  /dev/sdb: open failed: No medium found
Found memtest86+ image: /boot/memtest86+.bin
Found memtest86+ multiboot image: /boot/memtest86+_multiboot.bin
done

There is no sdb (only two drives are physically connected and they are raided by hardware raid1)

 

[Stoiko Ivanov]

/dev/sdb: open failed: No medium found

This is quite often caused by the 'virtual drive' from the iDrac/IPMI (you should be able to see it in the dmesg output on a freshly booted system (`dmesg | grep sdb`) - usually this does not cause problems however.

and after install stage i have the following hang screen:

try removing the quiet from the boot commandline (in grub hit 'e' before the highlighted entry is booted)

that should show where the kernel hangs

I hope this helps!

 

[ilia987]

after disabling all virtual devices (ipmi) it no longer look for sdb, but again a fresh install still heaving issues.

i think there might be a bad ram module (running full diagnostic now )

 

[ilia987]

This is quite often caused by the 'virtual drive' from the iDrac/IPMI (you should be able to see it in the dmesg output on a freshly booted system (`dmesg | grep sdb`) - usually this does not cause problems however.


try removing the quiet from the boot commandline (in grub hit 'e' before the highlighted entry is booted)

that should show where the kernel hangs

I hope this helps!

i remved the "quiet" and press ctrl+x but i did not see any messages on the screen.
i am able to ssh to the machine (under root) and print out the syslog:
here is the end of if

Sep  7 14:06:22 pve-srv6 systemd[1]: Startup finished in 8.236s (kernel) + 7.310s (userspace) = 15.547s.
Sep  7 14:06:29 pve-srv6 kernel: [   22.654774] L1TF CPU bug present and SMT on, data leak possible. See CVE-2018-3646 and https://
www.kernel.org/doc/html/latest/admin-guide/hw-vuln/l1tf.html for details.
Sep  7 14:06:41 pve-srv6 systemd[1]: Created slice User Slice of UID 0.
Sep  7 14:06:41 pve-srv6 systemd[1]: Starting User Runtime Directory /run/user/0...
Sep  7 14:06:41 pve-srv6 systemd[1]: Started User Runtime Directory /run/user/0.
Sep  7 14:06:41 pve-srv6 systemd[1]: Starting User Manager for UID 0...
Sep  7 14:06:41 pve-srv6 systemd[1467]: Listening on GnuPG cryptographic agent and passphrase cache.
Sep  7 14:06:41 pve-srv6 systemd[1467]: Listening on GnuPG cryptographic agent (ssh-agent emulation).
Sep  7 14:06:41 pve-srv6 systemd[1467]: Reached target Paths.
Sep  7 14:06:41 pve-srv6 systemd[1467]: Listening on GnuPG cryptographic agent and passphrase cache (access for web browsers).
Sep  7 14:06:41 pve-srv6 systemd[1467]: Listening on GnuPG cryptographic agent and passphrase cache (restricted).
Sep  7 14:06:41 pve-srv6 systemd[1467]: Listening on GnuPG network certificate management daemon.
Sep  7 14:06:41 pve-srv6 systemd[1467]: Reached target Sockets.
Sep  7 14:06:41 pve-srv6 systemd[1467]: Reached target Timers.
Sep  7 14:06:41 pve-srv6 systemd[1467]: Reached target Basic System.
Sep  7 14:06:41 pve-srv6 systemd[1467]: Reached target Default.
Sep  7 14:06:41 pve-srv6 systemd[1467]: Startup finished in 46ms.
Sep  7 14:06:41 pve-srv6 systemd[1]: Started User Manager for UID 0.
Sep  7 14:06:41 pve-srv6 systemd[1]: Started Session 1 of user root.
Sep  7 14:06:45 pve-srv6 systemd[1]: systemd-fsckd.service: Succeeded.
Sep  7 14:06:47 pve-srv6 systemd-timesyncd[962]: Synchronized to time server for the first time 162.159.200.123:123 (2.debian.pool.
ntp.org).
Sep  7 14:07:00 pve-srv6 systemd[1]: Starting Proxmox VE replication runner...
Sep  7 14:07:00 pve-srv6 systemd[1]: pvesr.service: Succeeded.
Sep  7 14:07:00 pve-srv6 systemd[1]: Started Proxmox VE replication runner.
Sep  7 14:08:00 pve-srv6 systemd[1]: Starting Proxmox VE replication runner...
Sep  7 14:08:00 pve-srv6 systemd[1]: pvesr.service: Succeeded.
Sep  7 14:08:00 pve-srv6 systemd[1]: Started Proxmox VE replication runner.
Sep  7 14:09:00 pve-srv6 systemd[1]: Starting Proxmox VE replication runner...
Sep  7 14:09:00 pve-srv6 systemd[1]: pvesr.service: Succeeded.

 

[Stoiko Ivanov]

i am able to ssh to the machine (under root) and print out the syslog:

if you can ssh into the machine then it does boot up successfully - if you don't have a login-screen on the directly attached screen - I guess there is a problem with initializing the graphics driver - check for errors in the journal since the bootup: `journalctl -b`

I hope this helps

 

[ilia987]

if you can ssh into the machine then it does boot up successfully - if you don't have a login-screen on the directly attached screen - I guess there is a problem with initializing the graphics driver - check for errors in the journal since the bootup: `journalctl -b`

I hope this helps

i just tried to install an older version image (6.1 instead 6.2) and this is the erros i found on of `journalctl -b'


ACPI: SPCR: Unexpected SPCR Access Width. Defaulting to byte size


efifb: abort, cannot remap video memory 0x1e0000 @ 0x0
efi-framebuffer: probe of efi-framebuffer.0 failed with error -5



ACPI Error: No handler for Region [SYSI] ((____ptrval____)) [IPMI] (20190703/evregion-132)
ACPI Error: Region IPMI (ID=7) has no handler (20190703/exfldio-265)
No Local Variables are initialized for Method [_GHL]
No Arguments are initialized for method [_GHL]
ACPI Error: Aborting method \_SB.PMI0._GHL due to previous error (AE_NOT_EXIST) (20190703/psparse-
ACPI Error: Aborting method \_SB.PMI0._PMC due to previous error (AE_NOT_EXIST) (20190703/psparse-
ACPI Error: AE_NOT_EXIST, Evaluating _PMC (20190703/power_meter-743)
IPMI message handler: version 39.2
ipmi device interface
ipmi_si: IPMI System Interface driver
ipmi_si dmi-ipmi-si.0: ipmi_platform: probing via SMBIOS
ipmi_platform: ipmi_si: SMBIOS: io 0xca8 regsize 1 spacing 4 irq 10
ipmi_si: Adding SMBIOS-specified kcs state machine
ipmi_si: Trying SMBIOS-specified kcs state machine at i/o address 0xca8, slave address 0x20, irq 1
mei_me 0000:00:16.0: Device doesn't have valid ME Interface
input: PC Speaker as /devices/platform/pcspkr/input/input4
dcdbas dcdbas: Dell Systems Management Base Driver (version 5.6.0-3.3)
mgag200 0000:0a:00.0: remove_conflicting_pci_framebuffers: passed res_id (0) is not a memory bar
mgag200 0000:0a:00.0: vgaarb: deactivate vga console
[drm:mgag200_driver_load [mgag200]] *ERROR* can't reserve mmio registers
mgag200 0000:0a:00.0: Fatal error during GPU init: -12
mgag200: probe of 0000:0a:00.0 failed with error -12

 

[Stoiko Ivanov]

i just tried to install an older version image (6.1 instead 6.2) and this is the erros i found on of `journalctl -b'

does the system also boot with 6.2 - then I would strongly suggest to debug with that (you'll need to upgrade anyways)

else - try upgrading all firmwares (BIOS/UEFI/NICs - Dell is quite comfortable when it comes to upgrades AFAIR) in the system - this sometimes helps with issues like that.

 

[ilia987]

does the system also boot with 6.2 - then I would strongly suggest to debug with that (you'll need to upgrade anyways)

else - try upgrading all firmwares (BIOS/UEFI/NICs - Dell is quite comfortable when it comes to upgrades AFAIR) in the system - this sometimes helps with issues like that.

the first step i took is to upgrade all hardware, ipmi, bios ....
yes same error on 6.2.
running upgrade and dist-upgrade now

 

[apoc]

Reset BIOS to default settings and specific check memory timings.
I have had all sorts of issues with bad/non default BIOS settings already (not specific to proxmox)
Hth

 

[ilia987]

i gave up. ill find another use for this server. thanks everyone

 

[totalimpact]

This is pretty troubling, I have tried to install on a R720, and now R730 also fails with 7.1-1 iso image. I am booting the iso via the iDrac IPMI, click on Install Proxmox and on the next screen it fails:

waiting for /dev to be fully populated...| ACPI Error: No handler f0331/evregion-130) 
ACPI Error: Region IPMI (ID=7) as no handler...

Then it seems to get online and run the install from the internet successfully, because it took 20-30 minutes to finally install vs the normal 3 minutes from local media. The R720 would simply reboot, it would never continue the install.

I have deployed Proxmox dozens of times via IPMI, and the last 2 I even defaulted everything and made sure to update to all latest bios/firmware. It has been comforting to know I could easily recover from any issue without needing to be physically present, now not so much.