Hi all,After some playing with the box and having fitted two brand new 4GB RED drives in it, today I started my first serious data moving. I have a failing HDD on another RAID I would like to replace and I was moving about 3TB of data on the NAS.All good, 98MB/s average (when moving big files), and at some point the NAS disappeared. No ping, no nothing. Tried pushing the power button once, no messages on the LCD about pushing it again to initiate a shutdown. LAN LEDs were active.Tried sticking in a USB drive - to see if the LCD was reporting it, nothing. No activity on the USB, no backup LED, no messages.I had to shut it down holding the button, then I restarted it. No errors, logs (on frontview) show no errors. Skipped the current file and the transfer restarted.I run some memory tests (2 or 3) weeks ago and all was good. Temperature are good as well.Any ideas? The NAS was possibly transferring a 1.5GB file when it disappeared. Any help appreciated.Thanks

Uhm... I have initiated two cat /dev/urandom > /dev/null and started a transfer at the same time. Also covered the CPU air intake till the temperature reached 50° and the fan speeded up. No signs of issues.The only thing, which I have seen before, the speed changes every now and then and this seems unrelated to the CPU stress test (it's not running now). I know the speed can depend on a number of variables but I'm sort of puzzled on the sudden change in speed, like something happened.

Tough subject? I've opened a ticket with Netgear, but I trust you more guys! :)

Are you just sending one large file, or is there a mix of files?Are this step changes happening with reads or writes (or both)?Also, are you past your disk issues?

ReadyNAS Business 6 crashed | NETGEAR Communities

13 Replies

tony359

Apprentice

May 05, 2014

Found this on the kernel.log. The time should be the one, considering when I posted my post above.
Does it mean anything? There are more lines

May  5 15:52:58 nas-EA-15-03 kernel: ------------[ cut here ]------------
May  5 15:52:58 nas-EA-15-03 kernel: ------------[ cut here ]------------
May  5 15:52:58 nas-EA-15-03 kernel: kernel BUG at fs/ext4/inode.c:2495!
May  5 15:52:58 nas-EA-15-03 kernel: invalid opcode: 0000 [#1] SMP
May  5 15:52:58 nas-EA-15-03 kernel: last sysfs file: /sys/devices/system/rn_lcd/rn_lcd0/lcd_line2
May  5 15:52:58 nas-EA-15-03 kernel: CPU 1
May  5 15:52:58 nas-EA-15-03 kernel: Modules linked in: nv6gpio nv6lcd nv6vpd(P)
May  5 15:52:58 nas-EA-15-03 kernel:
May  5 15:52:58 nas-EA-15-03 kernel: Pid: 2648, comm: flush-253:0 Tainted: P            2.6.37.6.RNx86_64.2.4 #1 To Be Filled By O.E.M. To Be Filled By O.E.M./To be filled by O.E.M.
May  5 15:52:58 nas-EA-15-03 kernel: RIP: 0010:[<ffffffff8813e9f4>]  [<ffffffff8813e9f4>] ext4_da_writepages+0x834/0xa80
May  5 15:52:58 nas-EA-15-03 kernel: RSP: 0000:ffff88003834baa0  EFLAGS: 00010202
May  5 15:52:58 nas-EA-15-03 kernel: RAX: ffffffffffffffff RBX: ffff88003800f3a8 RCX: 000000000000000c
May  5 15:52:58 nas-EA-15-03 kernel: RDX: 0000000000000000 RSI: 0000000000001021 RDI: ffff88003834bc00
May  5 15:52:58 nas-EA-15-03 kernel: RBP: ffff88003834bc80 R08: ffff88003847b2b8 R09: 00000000000000d5
May  5 15:52:58 nas-EA-15-03 kernel: R10: 000000000003e8d5 R11: 00000000000d5000 R12: ffff88003834bd70
May  5 15:52:58 nas-EA-15-03 kernel: R13: ffff88003847b3e0 R14: 000000000003e8d6 R15: ffff88003847b2b8
May  5 15:52:58 nas-EA-15-03 kernel: FS:  0000000000000000(0000) GS:ffff88003fa80000(0000) knlGS:0000000000000000
May  5 15:52:58 nas-EA-15-03 kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
May  5 15:52:58 nas-EA-15-03 kernel: CR2: 00000000f73ff0a0 CR3: 000000003820c000 CR4: 00000000000006e0
May  5 15:52:58 nas-EA-15-03 kernel: DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
May  5 15:52:58 nas-EA-15-03 kernel: DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
May  5 15:52:58 nas-EA-15-03 kernel: Process flush-253:0 (pid: 2648, threadinfo ffff88003834a000, task ffff88003e0a27d0)
May  5 15:52:58 nas-EA-15-03 kernel: Stack:
May  5 15:52:58 nas-EA-15-03 kernel:  0000000000000400 ffff88003fa0d680 ffff88003834bd00 ffff88003847b1f8
May  5 15:52:58 nas-EA-15-03 kernel:  0000000000008000 ffff88003847b1f8 000000000003b6b8 00000000000114c0
May  5 15:52:58 nas-EA-15-03 kernel:  ffff88003dc2ee70 0000000000000000 0000000000000001 0000000000000001
May  5 15:52:58 nas-EA-15-03 kernel: Call Trace:
May  5 15:52:58 nas-EA-15-03 kernel:  [<ffffffff880839ec>] do_writepages+0x1c/0x30
May  5 15:52:58 nas-EA-15-03 kernel:  [<ffffffff880ccb2b>] writeback_single_inode+0x8b/0x220
May  5 15:52:58 nas-EA-15-03 kernel:  [<ffffffff880cd03c>] writeback_sb_inodes+0xac/0x120
May  5 15:52:58 nas-EA-15-03 kernel:  [<ffffffff880cd66a>] writeback_inodes_wb+0xfa/0x130
May  5 15:52:58 nas-EA-15-03 kernel:  [<ffffffff880cd8cf>] wb_writeback+0x22f/0x2a0
May  5 15:52:58 nas-EA-15-03 kernel:  [<ffffffff88083bb5>] ? determine_dirtyable_memory+0x15/0x30
May  5 15:52:58 nas-EA-15-03 kernel:  [<ffffffff880cda72>] wb_do_writeback+0x132/0x1c0
May  5 15:52:58 nas-EA-15-03 kernel:  [<ffffffff880cdbd4>] bdi_writeback_thread+0x64/0x140
May  5 15:52:58 nas-EA-15-03 kernel:  [<ffffffff880cdb70>] ? bdi_writeback_thread+0x0/0x140
May  5 15:52:58 nas-EA-15-03 kernel:  [<ffffffff8804d056>] kthread+0x96/0xa0
May  5 15:52:58 nas-EA-15-03 kernel:  [<ffffffff88003014>] kernel_thread_helper+0x4/0x10
May  5 15:52:58 nas-EA-15-03 kernel:  [<ffffffff8804cfc0>] ? kthread+0x0/0xa0
May  5 15:52:58 nas-EA-15-03 kernel:  [<ffffffff88003010>] ? kernel_thread_helper+0x0/0x10
May  5 15:52:58 nas-EA-15-03 kernel: Code: 48 89 de e8 af 91 ff ff 85 c0 74 ba 48 8b 0b 48 8b 53 20 48 8d 7d 80 4c 89 f6 e8 e8 ca ff ff 8b 45 b8 85 c0 74 c1 e9 4a ff ff ff <0f> 0b eb fe 0f 0b eb fe 8b 85 dc fe ff ff 85 c0 0f 94 c3 e9 c6
May  5 15:52:58 nas-EA-15-03 kernel: RIP  [<ffffffff8813e9f4>] ext4_da_writepages+0x834/0xa80
May  5 15:52:58 nas-EA-15-03 kernel:  RSP <ffff88003834baa0>
May  5 15:52:58 nas-EA-15-03 kernel: ---[ end trace b6a87a470d02d980 ]---

tony359
Apprentice
May 06, 2014
Any ideas?
tony359
Apprentice
May 06, 2014
Uhm... I have initiated two cat /dev/urandom > /dev/null and started a transfer at the same time. Also covered the CPU air intake till the temperature reached 50° and the fan speeded up. No signs of issues.

The only thing, which I have seen before, the speed changes every now and then and this seems unrelated to the CPU stress test (it's not running now). I know the speed can depend on a number of variables but I'm sort of puzzled on the sudden change in speed, like something happened.
tony359
Apprentice
May 08, 2014
Tough subject? I've opened a ticket with Netgear, but I trust you more guys! :)
StephenB
Guru
May 08, 2014
Are you just sending one large file, or is there a mix of files?

Are this step changes happening with reads or writes (or both)?

Also, are you past your disk issues?
tony359
Apprentice
May 08, 2014
Thanks for your reply Stephen,

the largest file would be something like 30GB. It's a mixture of small documents and big videos.
The step happens while copying files - cannot remember whether it was writing or reading, I think the latter, but wasn't changing. And I believe it was happening WHILE it was copying a large file.
Disk issues? I've got two brand new RED drives now. The same behaviour was seen before though.

I could understand if the network was slowing down at times, but these massive steps...

I shall try it again bypassing the main switch - which happens to be a Netgear/Virgin Media router, maybe not the best on the world.
StephenB
Guru
May 09, 2014
Just thinking out loud here...

-The disk throughput varies, since the platters spin fastest at the outer edge (linear speed, the RPM is constant of course). So a large fragmented file could have a speed-shift like that, if the fragments were saved in different parts the disk. Fragmentation could be happening on either the NAS or the PC btw.

-Disk caching could also play a role here, possibly the cache was working more effectively at the end of the test.

-Reading a large file is much faster than reading a lot of small files. That could create a speed shift.

A network bottleneck between the NAS and the PC could also potentially explain it, so bypassing the router and retesting makes sense.
tony359
Apprentice
May 09, 2014
I will do more testing, thanks for your input, appreciated.

Back to the original topic, I have left the NAS on memory test mode tonight. 8 hours, 18 passed tests. None failed.
tony359
Apprentice
May 09, 2014
Netgear got back to me.

They say the 4TB RED are not on the compatible list and "have been known to cause issue in similar NASes". I'm concerned now. If it was just a matter of not being on the compatibility list fine, but if they are known to cause trouble...
Is that true, are 4TB RED causing trouble??
StephenB
Guru
May 09, 2014
I don't think the REDs are the cause. I think Netgear support was giving you the normal "not on the HCL" story, and the support contact may have tossed in the "known to cause issue" for good measure.

The HCL story is very tiresome. If they can't keep the list of good drives up to date, at least they could create a list of drives that they know for sure won't work.