ReadyNAS OS6 issues with Ultra 4

[crond]

Hi All,

I'm having trouble with OS6 on latest OS6 firmware and wondering if somebody saw similar issues.

I have Ultra4 with 4x4TB WD Reds. I had no issues on OS4.

Upgrade to latest version went well, but tests showed that data got corrupted (I would copy 200k / 2TB files and than compare md5sums). I tried factory reset and rebuild volume but it didn't help. tried to turn x-RAID off / remove quote  experemt with different protocols to transfer (AFP, NFS, rsync) but all results are consistent

After 2 days of reboots/stress testing it finally gave up with kernel crash

[633273.945339] BUG: unable to handle kernel paging request at 000000100000001c
[633273.946019] IP: [<ffffffff880dd551>] find_get_entry+0x3a/0x69
[633273.946019] PGD 0 
[633273.946019] Oops: 0000 [#1] SMP 
[633273.946019] Modules linked in: vpd(PO)
[633273.946019] CPU: 0 PID: 2247 Comm: afpd Tainted: P           O    4.4.140.x86_64.1 #1
[633273.946019] Hardware name: NETGEAR  ReadyNAS/ReadyNAS , BIOS 080016  09/20/2010
[633273.946019] task: ffff88003d65d400 ti: ffff880005f24000 task.ti: ffff880005f24000
[633273.946019] RIP: 0010:[<ffffffff880dd551>]  [<ffffffff880dd551>] find_get_entry+0x3a/0x69
[633273.946019] RSP: 0018:ffff880005f27c78  EFLAGS: 00010246
[633273.946019] RAX: ffff88001d0af4b0 RBX: ffff88001d140360 RCX: ffff88001d0af4b0
[633273.946019] RDX: ffff88001d0af4b0 RSI: 0000000000073e02 RDI: 0000001000000000
[633273.946019] RBP: ffff880005f27c88 R08: ffff880005f27c60 R09: ffff88001d0af478
[633273.946019] R10: ffff88001d0af4b0 R11: ffffffff88b6449c R12: 0000000000073e02
[633273.946019] R13: ffff88001d140358 R14: 0000000000073e02 R15: 0000000073e00000
[633273.946019] FS:  00007fe51e89b880(0000) GS:ffff88003f000000(0000) knlGS:0000000000000000
[633273.946019] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[633273.946019] CR2: 000000100000001c CR3: 000000000bf92000 CR4: 0000000000000670
[633273.946019] Stack:
[633273.946019]  0000000000000007 000000000342004a ffff880005f27cc8 ffffffff880dd5fe
[633273.946019]  00000000000178a3 0000000000000002 0000000000000002 ffff88001d1401f0
[633273.946019]  ffff8800110d4c10 0000000073e00000 ffff880005f27d30 ffffffff882aa65e
[633273.946019] Call Trace:
[633273.946019]  [<ffffffff880dd5fe>] pagecache_get_page+0x25/0x176
[633273.946019]  [<ffffffff882aa65e>] prepare_pages+0x7b/0x14a
[633273.946019]  [<ffffffff882ab7dc>] __btrfs_buffered_write.isra.6+0x227/0x51b
[633273.946019]  [<ffffffff8809e963>] ? current_fs_time+0x20/0x25
[633273.946019]  [<ffffffff882aedec>] btrfs_file_write_iter+0x2ba/0x3c6
[633273.946019]  [<ffffffff8811c9bd>] __vfs_write+0x9a/0xc3
[633273.946019]  [<ffffffff8811d0ce>] vfs_write+0xb6/0x160
[633273.946019]  [<ffffffff8811dc09>] SyS_pwrite64+0x5c/0x7b
[633273.946019]  [<ffffffff888e540a>] entry_SYSCALL_64_fastpath+0x1e/0x8e
[633273.946019] Code: 89 e6 48 89 df e8 e3 8a 28 00 48 85 c0 48 89 c1 74 41 48 8b 38 48 85 ff 74 39 40 f6 c7 03 74 0b 40 f6 c7 01 75 d9 48 89 f8 eb 2a <8b> 57 1c 85 d2 74 cd 8d 72 01 89 d0 f0 0f b1 77 1c 39 c2 74 04 
[633273.946019] RIP  [<ffffffff880dd551>] find_get_entry+0x3a/0x69
[633273.946019]  RSP <ffff880005f27c78>
[633273.946019] CR2: 000000100000001c
[633274.291068] ---[ end trace cc1296c118ab43b8 ]---

So it looks like there's some compatibility issue. I ordered new memory module to see if it could be the problem but ti's highly unlikely.

I tried to downgrade to 6.9.3 and it's more stable, but after 1+day of stress testing it also gave up and netgear daemon stoped working. I was able to ssh into the box, there were not kernel crashed but the daemoung has crashed with segfault on restart. 

After reboot, system went through initialization and I got to the wizard.

So from my side I have 2 questions, does anyone run legacy model pro4/6 ultra4/6 with WD REDs on OS6 ? if yes what version.

second is how to downgrade back to OS4 ? I found couple link on this forum but all of them are dead

[StephenB]

---

@crond wrote:

how to downgrade back to OS4 ? I found couple link on this forum but all of them are dead

---

This post has a working link: https://community.netgear.com/t5/Community-Add-ons/Downgrade-to-RAIDiator-x86-from-ReadyNAS-OS-notsu...

---

@crond wrote:

So from my side I have 2 questions, does anyone run legacy model pro4/6 ultra4/6 with WD REDs on OS6 ?

---

I'm sure that a lot of people do.

On the memory - did you try running the diagnostic from the boot menu?

[crond]

<p>Thanks for the link. </p>
<p>I didn't run memtest, was planning to upgrade memory anyway.&nbsp;</p>
<p>Do you now where did http://netgear.nas-central.org/wiki/Convert_RAIDiator_to_ReadyNASOS go ?</p>
<p>&nbsp;</p>
<p>&nbsp;</p>

[StephenB]

---

@crond wrote:
I didn't run memtest, was planning to upgrade memory anyway.

---

Ok.  Of course bad memory would affect OS 4 also, but perhaps not in the same way.

---

@crond wrote:
Do you now where did http://netgear.nas-central.org/wiki/Convert_RAIDiator_to_ReadyNASOS go ?

---

I don't know who hosted nas-central.org, but the entire site disappeared in mid-August 2018.

The files themselves were maintained by @mdgm and were hosted on a Netgear server, but that folder was deleted when he left at the end of August.  Hopefully the new links will keep working.

[Rottenweiler]

I have three Pro6 (actually 2 RNP Pro6 and 1 RNP Business 6) and an Ultra 4 ALL running OS6 with WD Red drives, perfectly. Drives are varing sizes from 3 to 6TB in each of the different units, but identical within each unit) No issues. On OS 6.9.4.HF1. Mine have all been working well, in fact much better than expected given the age of the boxes and the fact that they were originally "unsupported" on OS6. So it IS certainly possible, very good luck with that combination here, at least in my case. Best wishes, I hope you are able to get it sorted.