Napp-IT -- drives are gone?

[T_Minus]

I've had this system running 15 months or so now, and took it offline 4 months ago and now I am bringing it back online to pull down my data and then re-build.

However upon starting Napp-IT I noticed my HBA and drives were not there... no problem, configure pass-through again in ESXI, reboot, assign, start Napp-IT and I can see the PCIE HBA but no drives. The drives were seen in esxi when I clicked on the SAS2008 HBA at the bottom it listed my 8x HDD attached, etc so I know it's the correct HBA. I then powered down host removed the other sas2 hba and nvme hba rebooted, re-passed through, rebooted, verified OmniOS has the PCIE HBA (pci bus 0x000b cardnum 0x00 function 0x00: vendor 0x1000 device 0x0072 LSI Logic / Symbios Logic SAS2008 PCI-Express Fusion-MPT SAS-2 [Falcon]).

No drives showing up in Napp-IT or OmniOS (format/ iostate -eE to view drives on CLI).

Whenever I've changed HBA in the past I just swap cables, re-pass through, reboot, re-import pool and it works, nothing needed. This time I can't even "See" HDD in OmniOS or Napp-IT

What did I break?

Software:
esxi 6.0 (vmug licensed)
napp-it 16.04 Pro


The entire point of me getting this up is to back it up then build new home AIO and import pool there. I prefer to have a 2nd, 3rd, etc, location of my data before attempting to migrate/import on a new system even though I know technically speaking that should be fine too. So going to baremetal to import and move on is last resort if I can't get this AIO working

 

[gea]

What makes me wonder is that you can see the disks in ESXi. This should not be the case if the HBA is in pass-through mode.

Have you rebooted after setting the HBA to pass-through?

 

[T_Minus]

Sorry if that wasn't clear.

I was just stating I could see them in ESXI before passing through, and rebooting.

Once rebooted I can't see them in ESXI anymore.

I also wasn't sure if these could have been cached in ESXI too.


I'm actually in the process of rebooting now after swapping to another HBA to see if it's a weird HBA issue.

 

[T_Minus]

I put in another HBA that was also in here (but removed) and working fine... it didn't (esxi) detect the HBA at all.
Removed all pass-through, old HBA, etc, rebooted again.. no HBA found.

Going to try a different PCIE slot this time.

Very strange.

 

[T_Minus]

Different PCIE slot -- 2nd HBA still not working.

Put in the previous one where HBA is detected/pass-through but no drives in Napp-IT (but in ESXI).
Booted up, drives seen in ESXI, re-configure pass-through, reboot.

0 disks in napp-it/omniOS

 

[T_Minus]

What do you think about doing a fresh install of AIO/NappIT ? See if anything shows there?

 

[T_Minus]

Updated Napp-IT versions, no difference but it did stop working in FireFox (web gui).

Drives were all cold, not sure if due to 5400rpm but none were warm/vibrating even slightly... I unplugged and plugged in a couple and they spun up and felt them vibrate and then they warmed up, but then they got cold again... not sure if they can be powering down when they're not supposed to?

Seems to be a rather big problem for 0 changes to the system

 

[gea]

ESXi + LSI HBA + pass-through is usually a no brainer.
Enable and it works.

You can try the installer of OmniOS 151024 - just to be sure, starting the installer should be enough to check if the disks are seen.

 

[T_Minus]

ESXi + LSI HBA + pass-through is usually a no brainer.
Enable and it works.

You can try the installer of OmniOS 151024 - just to be sure, starting the installer should be enough to check if the disks are seen.

Yeah haven't ever had this issue before, esp since I boot it up / restart it often and sometimes it's down for weeks, this time it was couple months and now problems, very strange indeed!

 

[T_Minus]

OmniOS and Napp-IT downloads are 150KB/s or less for me

Anyone happen to have napp-it-san024.ova in the USA?

4 hours left LOL!!

 

[gea]

Within Germany (different provider than the one that hosts the file) it is a few minutes from both sources.

You may need a traceroute to see where the delay is.
<irony>Maybe its London - Continent Cut Off - or they want to leave the EU without paying the bill</irony>

btw
what about the OmniOS iso/usb installer?
(Located in Europe as well)

 

[T_Minus]

I got same download performance from there as well so I went with your AIO OVA.

 

[pricklypunter]

About ~30mins 1st mirror, ~18mins 2nd mirror to Canada, for me

 

[T_Minus]

500mb done, and still 5hrs left... LOL

 

[T_Minus]

I am getting 1.2MB/s - 1.9 on remote server better than my 50KB/s...
I'll get it there then grab it, LOL, still faster!

 

[T_Minus]

Able to get up early and get some work done on this.

As soon as I booted up the new Napp-IT VM and went to web GUI ALL disks were there, and pool is ready to import. YAYE!

However, it states:


status: The pool was last accessed by another system.
action: The pool can be imported using its name or numeric identifier and
the '-f' flag.


Napp-IT GUI states:
Use the following only in case of a damaged pool, Last changes are lost!!
import former state of damaged pool option -F



My Question:
Do I use the -F or not? My pool wasn't damaged but it's suggestion I use it...

 

[gea]

You need the -F option only on a damaged pool when a normal import fails.
This is a repair option with a possible dataloss of last writes.

Just import the pool. The -f is used in napp-it on a simple import.

 

[T_Minus]

Thanks @gea I wasn't sure if it was telling me I had to do use it in this instance.

I'll report back