Bricked NSA320S during 2017.07 flash

Cool!

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

Hi

I have sourced another 2x 3.5inch wd reds and I am trying to get them working with this NSA320s and something interesting happens:

u-boot sees 3 USB devices (1x USB disk with rootfs + 2xWD reds)

arting USB...
USB0:   USB EHCI 1.00
scanning bus 0 for devices... 3 USB Device(s) found
       scanning usb for storage devices... 
Use USB retry period from the environment: 15 second(s)
1 Storage Device(s) found

Reset IDE: Bus 0: OK Bus 1: not available  
  Device 0: Model: WDC WD20EFRX-68AX9N0 Firm: 80.00A80 Ser#:  WD-WMC301425233
            Type: Hard Disk
            Supports 48-bit addressing
            Capacity: 1907729.0 MB = 1863.0 GB (3907029168 x 512)
Unknown command 'mmc' - try 'help'

but kernel sees only one 3.5inch WD RED [sda] and USB disk [sdb] at dmesg

 12.429139] sd 0:0:0:0: [sda] 3907029168 512-byte logical blocks: (2.00 TB/1.82 TiB)
[   12.453444] sd 0:0:0:0: [sda] 4096-byte physical blocks
[   12.459674] sd 0:0:0:0: [sda] Write Protect is off
[   12.469877] sd 0:0:0:0: [sda] Mode Sense: 00 3a 00 00
[   12.470060] sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[   12.504646]  sda: sda1
[   12.515667] sd 0:0:0:0: [sda] Attached SCSI disk

and cli with fdisk -l

Disk /dev/sda: 1.8 TiB, 2000398934016 bytes, 3907029168 sectors
Disk model: WDC WD20EFRX-68A
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disklabel type: dos
Disk identifier: 0x987d4fe9

Device     Boot Start        End    Sectors  Size Id Type
/dev/sda1        2048 3907029167 3907027120  1.8T 83 Linux

BUT NOT the second WD RED.

Much obliged,

enki,

No , u-boot did not see the 2nd HDD. The 3 USB devices are the hubs and the USB rootfs.

> USB0: USB EHCI 1.00
> scanning bus 0 for devices... 3 USB Device(s)
> found
> scanning usb for storage devices...
> Use USB retry period from the environment: 15
> second(s)
> 1 Storage Device(s) found
>

There is only 1 HDD showed below:

> Reset IDE: Bus 0: OK Bus 1: not available
> Device 0: Model: WDC WD20EFRX-68AX9N0 Firm:
> 80.00A80 Ser#: WD-WMC301425233
> Type: Hard Disk
> Supports 48-bit addressing
> Capacity: 1907729.0 MB = 1863.0 GB
> (3907029168 x 512)


So I would say this is hardware related. Try removing the 2nd HDD and insert it back in. And if it still not appear, remove both HDDs and insert the 2nd HDD into the 1st slot, and then insert the 1st HDD in the 2nd slot....etc. Swapping them around like that would tell you where the problem is.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)



Edited 1 time(s). Last edit at 06/03/2020 04:35PM by bodhi.

Hi Bodhi,

I have already did that and both WD reds are technically ok. Booted exactly what you have suggested and it seems that it always recognize the disk placed on '1st hdd' slot on the nas but not both.

When I had 2x 2.5inch wd reds they were both recognized but they were only 1TB.

I cannot imagine is a power problem as 5A is the standard psu for this NSA320s.

I have formated both of the 3.5inch as Linux partition (type 83). They were type 'fd' Linux raid autodetect...

I am thinking to downgrade u-boot to earlier versions as I never had any issues with the GFNs since 2009...

Can I do downgrade of the uboot safely ?

enki,

> When I had 2x 2.5inch wd reds they were both
> recognized but they were only 1TB.

Quote

Reset IDE: Bus 0: OK Bus 1: OK
Device 0: Model: WDC WD10JFCX-68N6GN0 Firm: 82.00A82 Ser#: WD-WX91A157DLAR
Type: Hard Disk
Supports 48-bit addressing
Capacity: 953869.7 MB = 931.5 GB (1953525168 x 512)
Device 1: Model: WDC WD10JFCX-68N6GN0 Firm: 82.00A82 Ser#: WD-WXJ1E6496Y59
Type: Hard Disk
Supports 48-bit addressing
Capacity: 953869.7 MB = 931.5 GB (1953525168 x 512)

> I cannot imagine is a power problem as 5A is the
> standard psu for this NSA320s.
>

Yes, because it was OK previously as you tried with the 2.5 HDDs, I don't think it is a power problem.

> I have formated both of the 3.5inch as Linux
> partition (type 83). They were type 'fd' Linux
> raid autodetect...

Try reformat them both Ext3, and use MBR partition. After you got them recognized in Linux, then do the RAID thing,

> I am thinking to downgrade u-boot to earlier
> versions as I never had any issues with the GFNs
> since 2009...

No, don't downgrade. The latest u-boot (2017.07) has been installed a lot of times.

> Can I do downgrade of the uboot safely ?

You could downgrade to uboot.2016.05-tld-1. But going much earlier, you need to also reflash the envs image for that version.


=====

I will double check this box kernel DTS to see if I can find any thing strange.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

Hi Bodhi,

This is what you have noted in here (https://forum.doozan.com/read.php?3,12381)

-----------------------------------
Updated 30 Aug 2015

I've uploaded the Zyxel NSA310S/320S u-boot-2014.07-tld-4 image.

The installation instruction is the same for the NSA310S/320S as for other boxes that have a different stock env location: iConnect, Pogo V4/Mobile, and Netgear Stora. In the instruction below (First Updated 21 Dec 2014), I’ve modified it to include this new release.

Caveat: the NSA320S currently has a problem with using 3.5" HDD in the 2nd HDD slot. This seems to be a power-related problem, since using a 2.5" HDD in the 2nd slot is working fine.

Ah :) thanks for finding that! As you can tell, I have done too many devices I could not remember the details after a while.

I thought we solved it, but I could be wrong. I will look at the DTS to confirm.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

Bodhi,

Each of the boxes they have their own quirks. A solution for this NSA320s 3.5inch two disks would be great. Looking forward.

enki,

Refreshed my memory with notes, looked like that I've forgotten all about this issue. But I am optimistic. Will upload a new DTB so you can try!

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

enki,

I've attached here the updated kirkwood-nsa320s.dtb.

Download and copy to the /boot/dts:

cp -a /boot/dts/kirkwood-nsa320s.dtb /boot/dts/kirkwood-nsa320s.dtb.bak
cp -a kirkwood-nsa320s.dtb /boot/dts/
sync

And reboot.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

Bodhi

Thank you. I shall try it in couple of hiurs and shall get back to you...

E

Bodhi,

I have tried the updated dtb on the nas with both sata disks in and it is the same behaviour as in only the first sata disk is recognized. Maybe I am missing something ...

Bodhi

If I leave the boot sequence unintrerupted I have errors and it breaks...

EHCI timed out on TD - token=0x80008d80

now I need to break the autoboot sequence and do

setenv devices usb
boot

to skip check on sata disks and boot using the rootfs usb in debian.

Question: how I make sure the DTB I've just update its loading because I am not sure the boot is taking the new DTB in. Do I miss anything when I do just boot after setenv or do I need to tell it to load DTB? Thank you.

Bodhi,

It does load the new DTB as prints when booting from usb

Found bootable drive on usb 0
loading uImage ...
5409256 bytes read in 763 ms (6.8 MiB/s)
loading uInitrd ...
9963913 bytes read in 1137 ms (8.4 MiB/s)
loading DTB /boot/dts/kirkwood-nsa320s.dtb ...
13021 bytes read in 1646 ms (6.8 KiB/s)
## Booting kernel from Legacy Image at 00800000 ...
   Image Name:   Linux-5.6.5-kirkwood-tld-1
   Created:      2020-06-03  19:25:21 UTC
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    5409192 Bytes = 5.2 MiB
   Load Address: 00008000
   Entry Point:  00008000
   Verifying Checksum ... OK
## Loading init Ramdisk from Legacy Image at 01100000 ...
   Image Name:   initramfs-5.6.5-kirkwood-tld-1
   Created:      2020-06-03  19:25:35 UTC
   Image Type:   ARM Linux RAMDisk Image (gzip compressed)
   Data Size:    9963849 Bytes = 9.5 MiB
   Load Address: 00000000
   Entry Point:  00000000
   Verifying Checksum ... OK
## Flattened Device Tree blob at 01c00000
   Booting using the fdt blob at 0x1c00000
   Loading Kernel Image ... OK
   Loading Ramdisk to 0f18b000, end 0fb0b949 ... OK
   Loading Device Tree to 0f184000, end 0f18a2dc ... OK

Starting kernel ...

Uncompressing Linux... done, booting the kernel.

What do we need to look into next ?

enki,

Power up, interrupt serial console and

setenv devices usb
boot

Let it boot until finish. Login to Debian

and set the devices permanently.

fw_setenv devices usb

And post the entire serial console log here.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

Hi Bodhi

Here they are fw_printenv variables and the serial console of u-boot (screen.log) snapshots into files attached.

Enki



Edited 2 time(s). Last edit at 06/06/2020 05:04AM by enki.

Bodhi

In previous log attached there is 'ata2: SATA link down (SStatus 0 SControl F300)'

ata1: SATA link up 3.0 Gbps (SStatus 123 SControl F300)
[   12.353512][  T134] ata1.00: ATA-9: WDC WD20EFRX-68AX9N0, 80.00A80, max UDMA/133
[   12.361003][  T134] ata1.00: 3907029168 sectors, multi 0: LBA48 NCQ (depth 31/32)
[   12.386790][  T134] ata1.00: configured for UDMA/133
[   12.402412][    T7] scsi 0:0:0:0: Direct-Access     ATA      WDC WD20EFRX-68A 0A80 PQ: 0 ANSI: 5
[   12.418715][    T7] sd 0:0:0:0: [sda] 3907029168 512-byte logical blocks: (2.00 TB/1.82 TiB)
[   12.443044][    T7] sd 0:0:0:0: [sda] 4096-byte physical blocks
[   12.449323][    T7] sd 0:0:0:0: [sda] Write Protect is off
[   12.459716][    T7] sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[   12.493656][    T7]  sda: sda1
[   12.505912][    T7] sd 0:0:0:0: [sda] Attached SCSI disk
[   12.530701][    T1] Key type big_key registered
[   12.544084][  T145] usb 1-1: new high-speed USB device number 2 using orion-ehci
[   12.586378][  T145] usb 1-1: New USB device found, idVendor=05e3, idProduct=0608, bcdDevice=85.32
[   12.604102][  T145] usb 1-1: New USB device strings: Mfr=0, Product=1, SerialNumber=0
[   12.611911][  T145] usb 1-1: Product: USB2.0 Hub
[   12.617250][    T1] Key type encrypted registered
[   12.624972][  T145] hub 1-1:1.0: USB hub found
[   12.630067][  T145] hub 1-1:1.0: 4 ports detected
[   12.635754][    T1] ht1382 0-0068: setting system clock to 2020-06-06T08:16:38 UTC (1591431398)
[   12.756149][  T136] ata2: SATA link down (SStatus 0 SControl F300)
[   12.775607][    T1] Freeing unused kernel memory: 1024K

enki,

Yes that was only part of the info. I put in the regulator for 12V but it did not seem to have any effect (SATA1 is the 2nd port)

[   14.139008][  T187] v5.0-sata1: supplied by pwr_en_sata1
[   14.167847][  T187] v12.0-sata1: supplied by pwr_en_sata1

If you do this you can filter out the SATA activities:

dmesg | grep -i ata

I need to look at my notes when I built this u-boot to recall why I stopped investigating 2nd SATA 12V power (whether I ran out of time, or was it a deadend).

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

Hi Bodhi,

Ofcourse please take your time as it needs a bit of attention. I wonder if we can have a look at how was original uboot/kernel from Zyxel to recognize two 3.5 disks. Any ideas please give a shout and I shall try (as much as I can) to look into it as well. This guy on madadmin.com initiates sata disks with '

setenv init_sata 'mw.l f1010104 c70e0ffd;ide reset'

which I don't understand fully but anything you can think of give a shout. We need to fail this quick as we don't want to spend too much on it. 2.5inch disks work so we have a way out.

Anytime next week and Thank you in advance,
Kindest Regards
E

enki ,

> This guy on madadmin.com
> initiates sata disks with
> setenv init_sata 'mw.l f1010104 c70e0ffd;ide
> reset'

That's our user chrlee:

https://forum.doozan.com/read.php?2,19720

Basically that command set the GPIOs in u-boot 2nd GPIO bank to activate the SATA powers. I'm not quite sure that has all neccessary GPIOs.

I'm sure somewhere in the stock GPL (which I have, and used to understand this NSA310s/320s box) there might be some information that I have not come across back then.

You could try that in u-boot (with 2x 3.5" HDD attached)

Power up, interrupt serial console and

setenv init_sata 'mw.l f1010104 c70e0ffd; ide reset'
run init_sata

See if both are recognized OK.

But let me upload another DTB to clean it up before you try. If that poke commands work, then there is no need for regulator in DTS.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)



Edited 2 time(s). Last edit at 06/06/2020 07:50PM by bodhi.

enki,

If the above setenv commands work in u-boot, then go ahead and boot into Debian. Use this updated DTB.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)



Edited 2 time(s). Last edit at 06/06/2020 07:52PM by bodhi.

Bodhi,

Almost gave up on this one, however we have good news as below!

The second SATA is UP and SPINNING.

[    8.841425] libata version 3.00 loaded.
[   12.078680] sata_mv f1080000.sata: version 1.28
[   12.078963] sata_mv f1080000.sata: slots 32 ports 2
[   12.087715] scsi host0: sata_mv
[   12.092177] scsi host1: sata_mv
[   12.096388] ata1: SATA max UDMA/133 irq 32
[   12.101161] ata2: SATA max UDMA/133 irq 32
[   12.166411] 0x000002500000-0x000008000000 : "data"
[   12.613736] ata1: SATA link up 3.0 Gbps (SStatus 123 SControl F300)
[   12.623339] ata1.00: ATA-9: WDC WD20EFRX-68AX9N0, 80.00A80, max UDMA/133
[   12.643730] ata1.00: 3907029168 sectors, multi 0: LBA48 NCQ (depth 31/32)
[   12.654076] ata1.00: configured for UDMA/133
[   13.193626] ata2: SATA link up 3.0 Gbps (SStatus 123 SControl F300)
[   13.203062] ata2.00: ATA-9: WDC WD20EFRX-68AX9N0, 80.00A80, max UDMA/133
[   13.210466] ata2.00: 3907029168 sectors, multi 0: LBA48 NCQ (depth 31/32)
[   13.222605] ata2.00: configured for UDMA/133

I have set

setenv init_sata 'mw.l f1010104 c70e0ffd; ide reset'
run init_sata

and used your latest DTB.

Like we all do here, I trust this shall help anyone else who may need to run raid on NSA320s with 3.5inch drives.

Many thanks Bodhi,

Enjoy the weekend and good rest to all.E

enki,

> Almost gave up on this one, however we have
> good news as below!
>
> The second SATA is UP and SPINNING.

Cool! I'll add this note to the u-boot release and update the DTS in next kernel release.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

Hi Bodhi,

Yes please. There is a small thing, on how we include these instructions onto u-boot env.

Let me explain and ask for a robust solution. The instructions to include init_sata into u-boot envs must be constructed in the way users don't need to stop uboot and execute them manually. I am not expert in u-boot instructions so asking the community to construct the u-boot env command to execute at booting time

setenv init_sata 'mw.l f1010104 c70e0ffd; ide reset'
run init_sata

after 'run init_sata' we need to wait (most probably a delay around 7-10secs) and let to complete BUS1 'waking-up'. This is how long it takes BUS1 to be marked as OK.

How u-boot envs would look like? So please include these also in the instructions for NSA320s, if possible.

Kindest Regards, E

enki,

OK. Let me write the new envs and you can try it.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

enki,

Let's do 2 more test to make sure we are not messing up the current configuration.

With the latest DTB, set u-boot env to boot only with USB like you did. So devices should only show usb.

fw_printenv devices
devices=usb

Don't run init_sata. We only boot with USB and want to see if the SATA drives are still OK like it was before with the latest DTB.


Test 1. Insert the 3.5" HDD in the 1st slot, 2.5" HDD in the second slot. Boot and see if both SATA are spun up during kernel booting.

Test 2. Insert the 2x 2.5" HDD in the both slots. Boot and see if both SATA are spun up during kernel booting.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

Bodhi,

Quote

Test 1. Insert the 3.5" HDD in the 1st slot, 2.5" HDD in the second slot. Boot and see if both SATA are spun up during kernel booting.

result Test 1: Both 3.5'' (1st) and 2.5'' (2nd) HDDs spin after kernel boots; all OK.

There is another small things I've noticed: regardless is 2.5'' or 3.5'' the disk on slot 2 when you write on it it makes 'a bit of noise almost like clicks'. When you write data on disk on slot1 nothing is sounding/clicking... I don't mind but just saying maybe you may have any ideas why...

I shall do test 2 next.

E/

Bodhi,

Quote

Test 2. Insert the 2x 2.5" HDD in the both slots. Boot and see if both SATA are spun up during kernel booting.

result Test 2: Both 2.5'' HDDs spin after kernel boots; all OK. same 'click sounds' when you write on disk on slot no.2

Please note: with 3.5'' disk on slot 2 wihtout init_sata at u-boot to innitialize Bus 1... disk slot 2 doesn't spin and it's not recognized by kernel so we need to execute init_sata at uboot level to get 3.5'' disk slot 2 to spin. Saying this 2.5'' disk slot2 spins with and without init_sata.

When Bus 1 activates at u-boot after init_sata:

U-Boot 2017.07-tld-1 (Sep 05 2017 - 00:50:06 -0700)
ZyXEL NSA310S/320S 1/2-Bay Power Media Server

SoC:   Kirkwood 88F6702_A1
DRAM:  256 MiB
WARNING: Caches not enabled
NAND:  128 MiB
In:    serial
Out:   serial
Err:   serial
Net:   egiga0
MV88E1318 PHY initialized on egiga0
Hit any key to stop autoboot: 10  9  0 
NSA3x0S> setenv init_sata 'mw.l f1010104 c70e0ffd; ide reset'
NSA3x0S> run init_sata

Reset IDE: Bus 0: OK Bus 1: .......OK (completes almost after 10seconds)
  Device 0: Model: WDC WD20EFRX-68AX9N0 Firm: 80.00A80 Ser#:  WD-WMC300321381
            Type: Hard Disk
            Supports 48-bit addressing
            Capacity: 1907729.0 MB = 1863.0 GB (3907029168 x 512)
  Device 1: Model: WDC WD20EFRX-68AX9N0 Firm: 80.00A80 Ser#:  WD-WMC301425233
            Type: Hard Disk
            Supports 48-bit addressing
            Capacity: 1907729.0 MB = 1863.0 GB (3907029168 x 512)

Please note:
If I do execute just

NSA3x0S> run init_sata
without 
NSA3x0S> setenv init_sata 'mw.l f1010104 c70e0ffd; ide reset'
 set before then uboot gets hooked into sata bus innitializtion over and over and it is powered down/off.





Reset IDE: Bus 0: OK Bus 1: OK 
  Device 0: Model: WDC WD10JFCX-68N6GN0 Firm: 82.00A82 Ser#:  WD-WXJ1E6496Y59
            Type: Hard Disk
            Supports 48-bit addressing
            Capacity: 953869.7 MB = 931.5 GB (1953525168 x 512)
  Device 1: Model: WDC WD20EFRX-68AX9N0 Firm: 80.00A80 Ser#:  WD-WMC300321381
            Type: Hard Disk
            Supports 48-bit addressing
            Capacity: 1907729.0 MB = 1863.0 GB (3907029168 x 512)

Reset IDE: Bus 0: OK Bus 1: OK 
  Device 0: Model: WDC WD10JFCX-68N6GN0 Firm: 82.00A82 Ser#:  WD-WXJ1E6496Y59
            Type: Hard Disk
            Supports 48-bit addressing
            Capacity: 953869.7 MB = 931.5 GB (1953525168 x 512)
  Device 1: Model: WDC WD20EFRX-68AX9N0 Firm: 80.00A80 Ser#:  WD-WMC300321381
            Type: Hard Disk
            Supports 48-bit addressing
            Capacity: 1907729.0 MB = 1863.0 GB (3907029168 x 512)

Reset IDE: Bus 0: OK Bus 1: OK 
  Device 0: Model: WDC WD10JFCX-68N6GN0 Firm: 82.00A82 Ser#:  WD-WXJ1E6496Y59
            Type: Hard Disk
            Supports 48-bit addressing
            Capacity: 953869.7 MB = 931.5 GB (1953525168 x 512)
  Device 1: Model: WDC WD20EFRX-68AX9N0 Firm: 80.00A80 Ser#:  WD-WMC300321381
            Type: Hard Disk
            Supports 48-bit addressing
            Capacity: 1907729.0 MB = 1863.0 GB (3907029168 x 512)

Reset IDE: Bus 0:






E/







Edited 2 time(s). Last edit at 06/07/2020 06:46PM by enki.

enki,

Did the clicks occur before we started all this? i.e. with the two 2.5" HDDs in slots, and the released DTB (straight from the kernel 5.6.5 tarball).

If it clicks once, then it is OK. But if the clicking happens at every new disk write, then it is not OK.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

Bodhi,

It happened before all of latest DTS/init_sata also with the 2.5'' disk on slot 2 indeed. It clicks when BUS1 activates and then after whenever it writes on the disk.

Would it may be a diffrent bit from “Functional Specifications for Kirkwood's GPIO Data Out Register” to activate BUS1 in a mode which is appropiate and silent? I know it is not quite good to hear 'clicks' when writing on disks... but what is the appropiate solution?

Kindest Regards,
E