RN 104 freezing/not responding/locking up with FW 6.4.0

[mdgm-ntgr]

Note there are some fixes in 6.4.2 beta that may also help if your lockups on your 104 started on 6.4.1. See https://community.netgear.com/t5/ReadyNAS-Beta-Release/ReadyNASOS-6-4-2-T72-Beta-2/td-p/999457/jump-...

Also note that there is currently a known performance issue with anti-virus enabled on 6.4.1+

Particularly if a volume is quite full, the larger node size from a factory default (wipes all data, settings, everything) on 6.1.5+ can be quite beneficial on the 104.

[ChristopheD]

RN 104000 is started freezing up 5 or 6 days ago. It started with once a day, it is now twice a day.Can't access it through any means. It does respond to the pressing the power button three times and shuts down. On restart, everything is fine.

Checked the logs : no indication of anything. Disk spin down and timer option were never selected, so that's not the issue.

The only flag in the logs is the fact that the back up job did not successfully execute (because I did not have an external storage unit attached).

Any idea what's happening?

I went back to 6.4.0. from 6.4.1 maybe a month ago, shortly after installing it, as it essentially paralized the NAS.

I would welcome any and all suggestions!

Thank you.

[StephenB]

Maybe try 6.4.2-T72.  There is a fix targetted specifically at RN104 raid stability.

[ChristopheD]

Thanks for replying. Yes, 6.4.1 did not work at all for us, as we had antivirus turned on.

I don't think I'll be allowed to go to a beta version.

Any leads as to why the NAS is freezing up regularly? Only 18% of the disk capacity is being used....

[StephenB]

---

@ChristopheD wrote:
 

Any leads as to why the NAS is freezing up regularly? Only 18% of the disk capacity is being used....

---

Hard to say, though Netgear might be able to sort it out if they had the logs.  Hopefully they'll ask 

There are a couple of fixes in 6.4.1 and 6.4.2 beta that might be related.

-Fixes potential network buffer overrun with heavy traffic on RN102/104/2120.
-Fix potential hang due to a race condition when using the CPU's RAID5 offload engine. (RN104)

[dRdoS7]

Hi,

I posted here, rather than start a new thread, as it's the same FW being used.

I have an RN102, it's been running fine since I bought it in Oct 2015. On Jan 11 2016 it started going off-line/freezing, only way to get back is power off by removing the plug.

I updated to 6.4.0 when I first bought it, have tried 6.4.1 (Jan 11) & 6.4.2 T72 (Jan 13) since problems started, no better.

I just disabled "disk spin-down" to see if that's a problem.

I have no apps installed, no anti-virus running, using SMB, NFS, HTTP, HTTPS, uPnP, & SSL.

It has 1 x 3Tb JBOD and 1x 2Tb JBOD.

Emailed the lastest logs to mdgm.

Thanks,

dRdoS7

[mdgm-ntgr]

Looks like one or more of the disks may be failing.

Can you power down, remove your disks, hook them up to your PC and test the SeaGate disk using SeaTools and the WD disk using WD Data Lifeguard Diagnostics?

[dRdoS7]

Hi,

Yes, I'll do that.

Why does a failing disk not send an error message?

Why doesn't the failing disk just go off-line, without causing the 102 to freeze?

Thanks,

dRdoS7

[ChristopheD]

Thank you for the great replies and pointing me in the right direction.... I will go buy a few disks and proceed with replacing the existing ones and verify that indeed they have started to fail, all while monitoring the release of 6.4.2 ....:-)

[dRdoS7]

Hi,

1 disk so far, only Seagate 3Tb, indeed has some problems. That was indicated after only a short test. The 2nd disk, a WD 2Tb, test is still running.

I've bought a WD Red 4Tb, currently transferrring from the dodgey one (seems OK ATM, hope it holds out) to the new one. I did a "select and drag" on folder via the RN web interface and it decided I wanted to move between drives, rather than just copy. It's taking a long time. Not the fastest transfer I've ever seen. I would have thought an internal file transfer would be quick given it's sata, not usb1.

I did a search about replacing one drive in a RN, but mostly they mention raid. The faulty drive is #1 slot and contains the "home" directories. I don't know what will happen once I remove it, and swap #2 (new drive) into #1, and then return the original #2. Will the "home" directories be created automatically on the new #1? I've copied the folders & files from the "home" directories across to a PC.

Thanks,

dRdoS7

[StephenB]

Start by backing up the configuration file.

Also, app data is probably stored on drive 1's volume (access \\nasip\volumename with the admin credentials if you want to confirm.  There should be a hidden .apps folder).

Assuming that is so, the app data will also be lost.  The safest thing to do is unstall the apps.

Then back up all data (including \home)

explicitly delete the volume

remove the bad disk.

hot-insert the new disk, create the new volume (maybe with the same name).

reinstall apps

restore the data.

FWIW, if something goes wrong, and you need to do a factory reset, it would be difficult to remount your other volumes (perhaps Netgear could, but the system won't do that automatically).  So it'd be a good idea to have a backup of your second volume. 

[dRdoS7]

Hi,

I'll remember all that if I have to do it again, which hopefully I won't for a few years.

I have backed up the config, several times, I can restore that if needed.

As I posted, I've put the new drive in, and I'm already transferring data from old to new.

I don't use any apps. The NAS is purely storage (& 2nd backup) for photos, music, movies & other data.

Strangely, since I've put the new drive in, and the old drive is still in too, it's been going for 14 hours without freezing. The old drive is the one with errors, although I have yet to run a full test.

The #2 drive which I was testing in my PC had no errors. I've since read I could have run a disk scan in the NAS.

Thanks,

dRdoS7.

[kroche]

Latest update:

I have now installed the very latest beta software, done the factory reset and I am now copying data back to the readyNAS. Seems to be stable so far with no drop outs. It might be a bit slow but at least it works. It will take a few days to complete the restore as I'm doing it a bit at a time and checking each step.

Kevin

[dRdoS7]

Hi,

An update.

After I copied all the files off the old to the new, I S/D the NAS, removed the old #1, swapped new drive > #1 slot, returned #2 drive and powered up.

Mmm... not so good. It all booted OK, I can see all the files via Netgear web interface, I just can't access via Win Explorer! It won't accept the old user name & PW. Also, a short while after, the NAS froze again! This was ths supposed "good" drive #2.

I've also done a full test of the "bad" drive #1, and guess what: it passed all tests! Plus it never froze whil copying all files.

Anyway, bit the bullet, removed #2, Factory Reset, set it all back up with the new #1, working OK.

Installed #2, had to reboot to get it seen by the NAS, mounted just fine. Files all accessible.

Copying all files to #1, again!

I'm also copying all files from #2 to #1, as there's enough room. In all that time, it hasn't frozen. It does take a loooooong time.

I'll first run for a week with a single (new) disk. See if it locks up again. If that's OK, I'll try each of the other disks in turn in #2 slot, see what happens. If it doesn't lock up, I'll run with both original disks. If it still doesn't lock up, the Factory Reset fixed it! Then it wasn't the disks after all. At least I have 2Tb more storage, which the Finance Minister would never have agreed too!.

Maybe it was the way I tested the first disk (I used an eSata powered cable - poor/loose connection??) caused initial test errors. I later did test the 2nd disk the same way, and a re-test of the 1st. But who knows.

My advice to anyone installing/replacing a #1 drive, when it's single JBOD per disk, not RAID:

1. Power off, remove all disks.

2. Insert new disk in #1, power up.

3. Set up as per new NAS.

4. Insert old #1 into #2, reboot if needed, copy files old #1 > new #1 (assuming old disk is OK).

5. Remove old #1 from #2 slot, insert #2 disk.

6. Reboot.

The thought of copying all files off old #1 to another drive, then copying them all back to the new #1, fills me with dread: It's double the work/time. Plus you need to have the empty disk space lying around unused.

Of course, in theory, as this is a backup, the files are available elsewhere, so only one copy operation should required.

It's a shame the old disks can't be used, as they are ex NAS, via the USB or SATA ports, it would speed thing up.

Replacing a #2 (single disk JBOD)? I haven't, yet. I'm sure that will be more fun.

dRdoS7.

[dRdoS7]

Hi,

An update:

After running fine for 2 days, I enabled Disk Spin-down (10 min).

Not long after, the NAS froze!

I was still running with 1 new drive & 1 old drive. Seemed silly to remove the old one since it had been fine for 48 hrs. It had a good workout copying files from old to new.

I've removed the old drive now, left DS-D enabled, see what happens next few hours/days.

dRdoS7

[NetworkMechanic]

Anyone have any success? 

Mine freezes when I attempt a time machine back up. 

Anyone have that problem? 

I'm running the latest firmware. 

[ChristopheD]

I certainly have this problem. Systematically, when I run a back up job (on the front USB), the NAS freezes.

I run 6.4.0 as 6.4.1 simply slowed down the unit to a virtual halt.

Yet all the other parameters seem to be ok....

In 6.2.5, my back up jobs were running like a charm.

I think the OS is simply not up to par and it does not seem that Netgear has an intent to fix. These problems have been dragging on for a long time now.

I am at a loss. With a non-negligible number of users, I think we'll have to move to a more robuts hardware/software combination. This level of performance is simply not acceptable.

[kroche]

My steps to cure the problem of the freeze is to do a factory reset which will lose all the current data so you need to backup first. My RN 104 has worked perfectly since I did that. I also stopped running the Virus scanner, but I'm not sure that makes any difference.

The current OS version is simply incompatible with the way the disk was formated by the earlier version.

Kevin