HPE Proliant Servers

The disk is visible in

Storage/ Disks/

Iccan create it in

Strage/File systems/

but than its gone ....

its is sdf

while the raid disks sda, sdb, sdc.

but i dont which it is bay in the server. is it important somehow?

Zitat von geaves

We seem to have got crossed missed posts, dis you wipe the drive before creating a file system?

No! and thats it !
i wiped it in Disks and that formated it in Filesystem with Btrfs.
no i want to maky a daily (or 2 or more often) backup OR Rsync.

i think i will use rsync

Hello every one,

I installed OMV 3 back in 2016 on a Gen 8 proliant (bios J6) following the tips.

Here is what I did :

1. boot with debian live key in usb port and a fresh ssd in the front bay
2. installed debian (them omv) on that ssd
3. rebooted on the ssd
4. from debian, I flashed a usb key with supergrub2.
5. moved the ssd to the optic sata port
6. rebooted and everything worked fine.
7. installed 4 WD 3to in the front bay (software raid 5 etc.)

Tl;dr : I used to boot on an SSD (optical drive), launched via USB (supergrub).

Every thing worked flawlessly.

Since yesterday I can't boot on any hard drive which ever the port I place them.

I tried both disks on other machines (usb boot key and ssd) and I can mount them and see the partitions.

I tried to boot from super grub but I can't see any hard drive in super grub. When I list the partition, the hard drive is there, but I can't find any boot partition (altrough I booted from it the day before). I tried supergrub and another ssd (with debian 9 already installed) but it doesn't work either.

I can boot on a live usb image and reach the desktop where I can see my SSD partitions... But I still can't boot them.

Every time I try to boot on a hard drive, it's skipped and it tries to boot from network (NIC) which fails after the DHCP server gives it an IP.

What is wrong with my gen8 ? What can I do to solve this problem ?

Any help would be greatly appreciated.

thanks

Any error message ?

Maybe something like "error: 'grub_calloc' symbol not found."

Cheers,

Thomas

Zitat von ananas

Any error message ?

Maybe something like "error: 'grub_calloc' symbol not found."

Cheers,

Thomas

There's no error message...

But I tried something else :

I set up sata AHCI in bios and tried again.

I can boot on the first SATA port containing a hard drive, I can't boot from USB with grub

I'll try to create another key from recovery mode since it's now possible.

Thanks for your answer

What boot order do you have in your setup ?

Do you know about the power of the iLO ?

Cheers,

Thomas

Zitat von ananas

What boot order do you have in your setup ?

Do you know about the power of the iLO ?

Cheers,

Thomas

usb, cd, hd and the rest.

I can do ILO (I forgotted it existed since the nas hasn't failed in 4 years).

---

I created another bootable usb drive and it's not working.

As long as there's only one drive (ssd with system) in the SATA ports, I can boot from it.

As soon as there's another drive in the front bay it won't boot.

It doesn't seem to boot from the usb drive anymore...

Is there another way to boot from the 5th drive ?

In AHCI mode, it will always try to boot from the first HD.

I have 4 data drives, one SSD connected to the internal SATA Port and the bootloader on the internal SD-Card.

Was a bit tricky to get it running ...

You can use the "boot once" menu to boot from your installation/rescue media.

Maybe remove the 4 data drives, leaving the SSD and your USB stick connected and boot once

from (virtual) "debian-10.5.0-amd64-netinst.iso" and choose "repair grub"

Zitat von ananas

In AHCI mode, it will always try to boot from the first HD.

I have 4 data drives, one SSD connected to the internal SATA Port and the bootloader on the internal SD-Card.

Was a bit tricky to get it running ...

You can use the "boot once" menu to boot from your installation/rescue media.

Maybe remove the 4 data drives, leaving the SSD and your USB stick connected and boot once

from (virtual) "debian-10.5.0-amd64-netinst.iso" and choose "repair grub"

Alles anzeigen

Thanks for your answer.
I'll try that before patching the bios et setting up raid 1 for my ssd as explained here (french).

Hello there,

I've made some progress and here is what I found :

My Proliant Gen 8 won't boot anymore from SOME usb keys :

• linux distro (debian 10, 9, whatever) : fine, will boot and install a linux system on any hard drive
• grub USB key (the one you download online and flash from another computer) : fine also (but won't see the installed linux as it used to do)
• grub bootloader (the one you create after installing linux) : NO GOOD. WILL NEVER BOOT. It worked flawlessly for 4 years, but I can't make it work anymore

It will still boot from hard drive (only from the first one, as you know).

After many attempts I swaped the SSD in the first port of the front bay and the first Raid drive to the optical port.

It works...

As you can see on the picture, the SSD is in port 1 and the drive that used to be there is on top.

It's the only workaround I've found atm.

During the boot, mdadm seemed unhappy but managed to create the raid and OMV tells me it's "clean".

---

I guess I could create a raid 1 with only the SSD in it as decribed here (french link, sorry) which would allow me to boot directly from the SSD...

So, here's my question :

1. Does creating a raid 0 array for the SSD erases his data ? I guess so. (Dynamic Smart Array B120i raid)

---

I Had another question for which I've found on stackoverflow it's possible :

2. Is it possible to recover the MDADM raid 5 from another install ? There's still important data I would like to keep...

So...

To make it work I had to choose between something hazardous and something even more hazardous :

1. have the SSD in SATA 1 and a storage disk in SATA5

2. Activate RAID 1 on SATA 5 and reinstall everything.

I choose the fastest way and it seems to work (MDADM is still unhappy but it finally says the raid5 is "clean"). Here's what I did :

I unscrewed the backpanel with sata connectors, I removed the sata 1 and cut some tape to free enough lenght of cable. I pluged the SSD on sata 1.

I plugged the HD storage in the first port, with the cables that used to connect the SSD.

Now the Microserver boots directly on SSD, no usb/SD/raid1 required.

It is an ugly hack which I'm not proud of, but it works.

I guess I'll replace the whole server as soon as I'll have enough money...

Hi I could you help, I bought an HP ML30 Gen9 second hand, its shows a 270 - iLo cant communicate error at boot, I have tried to install the latest service pack but the firmware bios wont update, I have even tried installing the bios from the system menu and this fails also the intelligent provisioning fails to install, I have been trying to get an answer on the HP forums for 2 weeks with no success. Any help would be appreciated thanks

Just to let know, my problem is resolved.

RE: HP Proliant Mikroserver Gen 8 (celeron) & Omv 5

Hey Spy Alelo ,

thank you for helping all of us out here.

My question is kinda tech and general hardware related.

I am using a HP ProLiant ML10 v2 for over 6(?) years now (upgraded with a E3-1220). I have 7 HDDs and 2 SSDs (glued to the bottom of the case) and no chance to put in more drives. Performance is fine for OMV, Nextcloud and some containers.

Recently I thought about new network hardware and a 19" rack. Therefore I want to put my OMV-Server into the Rack too. Do you have any recommendations what to look for at eBay and so on? I have like no plan about rack servers, but I love iLO and would like to move on with HP/E. I am open for both viable options: 1. Move the ML10 parts into a rack case and 2. Buy a used server and have a second one.

I am kinda tending towards option 2 since I never reinstalled OMV after like Version 3 (just deleted php5 & 7.0 today ). I could buy a complete new SSD and 2 more drives, copy all data from the old server, reuse most of the drives from there and move on with the new server.

Thank you very much in advance!

Greetings

Andy