Fresh install (TrueNAS-13.0-U5.3), missing disks in Storage-Disks

[idang]

Hi All,
I ran a fresh install on a new CWWK AIO-7 (J6413) with 6 new, never used before, Seagate IronWolf 8TB.
When attempting to create a new pool, only 4 of the disks are displayed (ada0, ada3, ada4,ada5).
In shell

Code:

geom disk list

shows all 6 disks (+the SSD with the OS).
Reboot of the system did not change the behavior.
What are the best ways to proceed? (I am stuck because I want the pool to have all 6 disks)
Thanks in advance.

 

[Patrick M. Hausen]

What's the output of gpart show ada1 and gpart show ada2, respectively?

 

[idang]

The response is
"gpart: no such geom: ada1"
"gpart: no such geom: ada1"
while geom disk list shows them:

and

 

[Patrick M. Hausen]

Code:

dd if=/dev/zero of=/dev/ada1 bs=1m count=1024
dd if=/dev/zero of=/dev/ada1 bs=1m oseek=7630000
dd if=/dev/zero of=/dev/ada2 bs=1m count=1024
dd if=/dev/zero of=/dev/ada2 bs=1m oseek=7630000

Please post text instead of screenshots so I can copy and paste your numbers for calculations. I hope I got it correct by retyping ...

A reboot might be necessary after that.

 

[idang]

Thank you so much for the assistance.

The output of these commands was:

root@truenas[~]# dd if=/dev/zero of=/dev/ada1 bs=1m count=1024
1024+0 records in
1024+0 records out
1073741824 bytes transferred in 4.774882 secs (224872934 bytes/sec)
root@truenas[~]# dd if=/dev/zero of=/dev/ada1 bs=1m oseek=7630000
dd: /dev/ada1: short write on character device
dd: /dev/ada1: end of device
886+0 records in
885+1 records out
928342016 bytes transferred in 7.617337 secs (121872252 bytes/sec)
root@truenas[~]# dd if=/dev/zero of=/dev/ada2 bs=1m count=1024
1024+0 records in
1024+0 records out
1073741824 bytes transferred in 4.739077 secs (226571907 bytes/sec)

Storage -> Disks in webUI still did not show ada1 and ada2.
A system restart (Dashboard- restart) did not change the situation, Storage -> Disks still doesn't show these 2 disks

 

[Patrick M. Hausen]

If you just go ahead and create a pool, do the disks show up? I am not quite sure if Storage > Disks is even supposed to show disks that are not part of a TrueNAS managed pool.

 

[winnielinnie]

If you just go ahead and create a pool, do the disks show up? I am not quite sure if Storage > Disks is even supposed to show disks that are not part of a TrueNAS managed pool.

Just tested this out.

A non-ZFS disk (obviously not part of any pool) does show up under Storage > Disks. (Under the "Pool" column it reads "N/A".)

It is also available when going to Storage > Pools > Add.

TrueNAS Core 13.0-U5.3

Makes me believe this could be a middleware/GUI issue for the OP? (Because the drives are indeed available via command-lines tools, according to the pasted output.)

 

[idang]

 

[winnielinnie]

Out of curiosity, what happens if you try a "dry-run" pool creation via the command-line? Using the -n flag nothing will happen, as it will only simulate a pool creation. (Still needs to be run with root privileges.)

For example,

Code:

 zpool create -n MyPool raidz2 /dev/ada0 /dev/ada1 /dev/ada2 /dev/ada3 /dev/ada4 /dev/ada5

Don't worry about any fancy options. Just a simple pool creation dry-run.

* Note: This isn't actually what the middleware will do, as TrueNAS first partitions your drives for swap + remainder, and it is the second partitions (p2) that are used as members for the pool's vdev.

 

[idang]

Out of curiosity, what happens if you try a "dry-run" pool creation via the command-line? Using the -n flag nothing will happen, as it will only simulate a pool creation. (Still needs to be run with root privileges.)

For example,

Code:

 zpool create -n MyPool raidz2 /dev/ada0 /dev/ada1 /dev/ada2 /dev/ada3 /dev/ada4 /dev/ada5

Don't worry about any fancy options. Just a simple pool creation dry-run.

* Note: This isn't actually what the middleware will do, as TrueNAS first partitions your drives for swap + remainder, and it is the second partitions (p2) that are used as members for the pool's vdev.

Any way I tried to copy the text resulted in partial text so this is the screenshot:

 

[winnielinnie]

Then perhaps this is indeed a middleware and/or GUI issue. Did you try what @Patrick M. Hausen suggested?

If you just go ahead and create a pool, do the disks show up?

 

[HoneyBadger]

Hello @idang

Can you provide some additional information about the hardware (specifically the motherboard) being used? Since the drives are visible in geom list it's not likely to be a missing or unsupported driver, but rather a middleware challenge with enumeration.

 

[idang]

Hello @idang

Can you provide some additional information about the hardware (specifically the motherboard) being used? Since the drives are visible in geom list it's not likely to be a missing or unsupported driver, but rather a middleware challenge with enumeration.

I only found an Aliexpress page that has the same product listed and the motherboard model is not specified. These are all the details specified and I found a forum that mentioned the motherboard of a previous version may have been a topton Nas board but I cannot verify this is the case.

 

[HoneyBadger]

Usually when we see an issue with drives not being visible in the UI, it's due to things such as a USB enclosure or SATA port multiplier - it might be possible that the motherboard is using the latter here, as the SoC of the Celeron J641x only supports two SATA ports, and this unit is claiming support for six. I suspect there might be a single port directly attached, and another going through a 1-to-5 JMicron or ASMedia chip. The output of lspci -nn and/or dmesg could be useful in determining that here.

You can also try the "Report a Bug" link at the top of the screen, and then create a Debug file (System -> Advanced -> Save Debug) and attach it to the ticket - that will keep any potentially sensitive information in the debug file limited to iXsystems visibility only.

 

[Ericloewe]

RTL8125BG? That's like begging for a Man in the Middle attack. The product naming person at Realtek clearly spends too much time alone, as they started a joke by going with two letters (which not even the RTL8111 and its alphabet soup of variants needed). What's with these obsessions around all the stupid variants? One variant every three years or so, times 3 footprints (optimistically) still allows for 20+ years until they run out of variants. Is it about creating a motherboard marketing paradise (and I pity those involved) where the vendors compete with each other by having "better" versions of the RTL8125? Or is it just a secret love for confusing people like me who try to make sense of it (sometimes still waters run deep and people express themselves in weird ways)?

The real tragedy of course is that Intel messed up the I215 big-time and left Realtek looking like the sane option, a situation unheard of since the turn of the century. I mean, even the whole "sorry, it's fixed in new silicon" seems to have been empty words, from all the ongoing hubbub. You win again Realtek, much to my despair. I wonder what lies down the road, as 2.5 GbE spreads...

 

[idang]

root@truenas[~]# lspci -nn
00:00.0 Host bridge [0600]: Intel Corporation Device [8086:4522] (rev 01)
00:02.0 VGA compatible controller [0300]: Intel Corporation Device [8086:4555] (rev 01)
00:08.0 System peripheral [0880]: Intel Corporation Device [8086:4511] (rev 01)
00:14.0 USB controller [0c03]: Intel Corporation Device [8086:4b7d] (rev 11)
00:14.2 RAM memory [0500]: Intel Corporation Device [8086:4b7f] (rev 11)
00:16.0 Communication controller [0780]: Intel Corporation Device [8086:4b70] (rev 11)
00:17.0 SATA controller [0106]: Intel Corporation Device [8086:4b63] (rev 11)
00:1a.0 SD Host controller [0805]: Intel Corporation Device [8086:4b47] (rev 11)
00:1c.0 PCI bridge [0604]: Intel Corporation Device [8086:4b38] (rev 11)
00:1c.1 PCI bridge [0604]: Intel Corporation Device [8086:4b39] (rev 11)
00:1c.2 PCI bridge [0604]: Intel Corporation Device [8086:4b3a] (rev 11)
00:1c.3 PCI bridge [0604]: Intel Corporation Device [8086:4b3b] (rev 11)
00:1c.4 PCI bridge [0604]: Intel Corporation Device [8086:4b3c] (rev 11)
00:1f.0 ISA bridge [0601]: Intel Corporation Device [8086:4b00] (rev 11)
00:1f.3 Audio device [0403]: Intel Corporation Device [8086:4b58] (rev 11)
00:1f.4 SMBus [0c05]: Intel Corporation Device [8086:4b23] (rev 11)
00:1f.5 Serial bus controller [0c80]: Intel Corporation Device [8086:4b24] (rev11)
01:00.0 SATA controller [0106]: JMicron Technology Corp. JMB58x AHCI SATA controller [197b:0585]
02:00.0 Ethernet controller [0200]: Intel Corporation Device [8086:125c] (rev 04)
03:00.0 Ethernet controller [0200]: Intel Corporation Device [8086:125c] (rev 04)
04:00.0 Ethernet controller [0200]: Realtek Semiconductor Co., Ltd. RTL8125 2.5GbE Controller [10ec:8125] (rev 05)
05:00.0 Non-Volatile memory controller [0108]: Silicon Motion, Inc. SM2263EN/SM2263XT SSD Controller [126f:2263] (rev 03)

 

[idang]

Usually when we see an issue with drives not being visible in the UI, it's due to things such as a USB enclosure or SATA port multiplier - it might be possible that the motherboard is using the latter here, as the SoC of the Celeron J641x only supports two SATA ports, and this unit is claiming support for six. I suspect there might be a single port directly attached, and another going through a 1-to-5 JMicron or ASMedia chip. The output of lspci -nn and/or dmesg could be useful in determining that here.

You can also try the "Report a Bug" link at the top of the screen, and then create a Debug file (System -> Advanced -> Save Debug) and attach it to the ticket - that will keep any potentially sensitive information in the debug file limited to iXsystems visibility only.

Thank you.
I sent the output of the lspci but the dmesg output was too long so I opened a bug as suggested. I hope there is still a way to solve this because this nice little box with 6 disks in it, sits idle atm.
The listing I sent specifies it can host uraid integrated as an option so I hope it means that at least unRaid is able to recognize and operate the 6 disks on this hardware so hopefully TrueNAS can as well.

 

[HoneyBadger]

01:00.0 SATA controller [0106]: JMicron Technology Corp. JMB58x AHCI SATA controller [197b:0585]

There's the JMB585 as expected, this might be an issue.

Popped open your debug; your ada1 and ada3 devices have the same serial number being returned which you can check yourself with camcontrol identify ada1 - similarly, ada2 and ada4 are sharing an S/N. But oddly, they all have different WWN (world-wide names) - where'd you buy the disks from?

I'd suggest shuffling the drives around (with the power off) and see if two different disks disappear.

 

[jgreco]

The JMB585 is known not to work well (or at all).

Multiply your problems with SATA Port Multipliers and cheap SATA controllers

In the last year or two, we've had a resurgence of users asking about SATA Port Multipliers and cheap SATA controllers. Please, do NOT use port multipliers, and use cheap SATA controllers only after extensive research. SATA controllers and SATA...

www.truenas.com

 

[idang]

There's the JMB585 as expected, this might be an issue.

Popped open your debug; your ada1 and ada3 devices have the same serial number being returned which you can check yourself with camcontrol identify ada1 - similarly, ada2 and ada4 are sharing an S/N. But oddly, they all have different WWN (world-wide names) - where'd you buy the disks from?

I'd suggest shuffling the drives around (with the power off) and see if two different disks disappear.

I added the actual SN of the 4 drives you mentioned to the bug.
I swapped ada1 and ada4, Storage -> Disks show ada0, ada2, ada4, ada5 (compared to the previous configuration ada2 appeared, ada3 vanished).
Disks are new, original, and verified by Seagate's website. All of them are recognized individually on an external sata bay