Hi JamesHaskell - I'm sorry about the experience that you had, we do work hard to try to provide excellent support. I really appreciate your detailed feedback showing line-items of the issues that you have had. I will take that feedback and try to incorporate it into improving our support process.
It does looks like the issue that you ran into is rather unique - I'm afraid it may have been compounded by some of our otherwise standard troubleshooting processes. I'll work on my side to get to the root of the issue itself, and then to refine the troubleshooting process to accomodate for such situations.
Things are better!! As of last Friday, it appears the system is back up and running properly. Also on Friday, a new tech support person (Josh Jones) started with working with me. He has been very helpful. Here's my last email exchange with him from this morning. -------------------------------------------------------------------------------------------
Josh:
Thank you so much for this report. I really appreciate it. You have been tremendous to work with.
I ran the disk test as we discussed yesterday, and I presume it ran OK... the system apparently rebooted itself as you said it would around 10:42 last night, and is back up and running. There is nothing in the FrontView status log about the disk test... there is just a single entry saying "System is up". So "no news is good news"? Everything is OK on the drives?
Thanks again Josh. Like I said, you have been fantastic - by far the best person I have spoken to and worked with at NetGear. You have been very helpful, you've followed up when you said you would, and your thorough research and explanation of this issue has greatly eased my mind about the reliability of my ReadyNAS. Could I please get contact information for your supervisor? I want to encourage them to not only recognize you, but to work to get your colleagues up to your level. All of the other tech support personnel I dealt with on this issue were very disappointing compared to you: 1. Most seemed to be off my time-zone (Phillipines?), so every little step took at least a day. 2. English is not their first language. There were communication problems. 3. I had no visibility to my case notes. This is a BIG issue... much of my frustration was due to a lack of proactive communication from NetGear. If I could have accessed the case notes on-line and seen progress, it would not have been so frustrating. As it was, it seemed like days would go by with no apparent progress. 4. You are by far the most knowledgeable person I spoke with. Your explanations and requests were clear, as demonstrated in your email below. 5. You were sympathetic and apologized to me. You empathized with me. Nobody else did. 6. You were prompt. You got back to me quickly with this full report.
Once again, great job. It has been a pleasure working with you.
James
On Mon, Jun 13, 2011 at 4:05 PM, Josh Jones <jjones@netgear.com> wrote:
Hi James,
In taking a look through the configuration on the NAS, everything does seem to be in-place. Even the more-tricky hard-linking of 64-bit FS tools seems to have been done correctly – which is a good sign.
Regarding the events that led-up to the issue, it looks like here’s what happened (leaving out some details that do not appear to have directly affected the issue, such as reboots etc):
1. Communication to/from disk 1 was lost. Disk 1 was kicked from the active RAID arrays (there are three arrays, the OS, swap and data array).
2. The chassis was replaced, disks 2-4 were put into the replacement unit and the unit was booted.
3. After the boot was successful, disk 1 was re-added and synced successfully, however, it appears that critical RAID superblock data was not written correctly to the OS and swap arrays on the disk. This may have been the “homehost” information which would link the RAID superblocks to the MAC address of the NAS.
4. About 12 hours later, communication to/from disk 4 was lost. It was kicked from the active RAID arrays.
5. A subsequent reboot caused the unit to boot with the impression that disk 1’s OS and swap arrays were not synced, and disk 4 was out of sync in general. This caused the both drives to appear as dead.
6. Throughout subsequent reboots, drive 4’s data array resync’d and updated correctly, but both drive 1 and 4’s OS and swap array information would not persist across reboots, causing them to appear as dead each time.
7. Support manually recreated theses arrays, erasing configuration. This ensured that the drives would be assembled into the array at boot.
8. During the subsequent manual rebuild of the OS, a necessary file (/etc/hostname) was not created/configured.
9. Upon creating that file, everything appears to be functioning.
If this is accurate, for the most part, then the following actions will need to be taken (I am working on doing these now):
1. The original issue of disk 1 showing as dead may have been caused by the auto-spindown setting. I’d recommend running a ‘Disk Test’ from the boot menu on the NAS as a precaution, but in the meantime I will try to get a hold of this drive model and experiment with the auto-spindown feature. While this is part of our standard qualification procedure, it’s possible that the combination of events that led-up to the failure were not experienced during testing.
2. I will try to reproduce the issue where the OS/swap arrays are not updated correctly, causing disks to appear as dead upon reboot.
3. I will notify support on the best practice for rebuilding the OS so as to preserve the configuration whenever possible.
As per our phone conversation, I’ll also put-in a feature request to create a Time Machine backup job option. Also, here is a link that discusses how to set-up ReadyNAS Remote so that the ReadyNAS can easily be accessed through a secure VPN tunnel by clients that are not on the same LAN:
Well, if the volume is healthy, you should not have any problems replacing the disk with a suitable replacement. We do suffer disk failures from time to time. If the volume is healthy and no other disks are about to fail, it is a matter of hot pulling the failing/failed disk and hot adding the replacement. So far, I have lost one disk over four years on my NV+ and three disks over two NVX units over the last year. I have RMA'ed each, and Seagate has promptly sent me a replacement on the first two. I have one that SeaTools for DOS has repaired, and one I have to RMA (it failed Saturday).
Just to cover the bases, it sure wouldn't hurt to make sure your backup is current. Then if a second drive fails while synching the replacement, you will be covered.
Given the history of chronic failures on this unit over the last 35 days, it is very unlikely it is a disk failure... it is probably more of the same stuff that I've been fighting with (fouled up system partition, damaged system data, etc).
I reported the issue yesterday to tech support... not a word yet in response...
I would certainly perform the vendor tests on the drive. But, again, I would make sure my backup was current.
If you hot pull the problem drive and perform the tests and then hot add it back it will be tested and then the system will go through a resync of the volume. You may have to remove the partitions while connected to the PC if it passes the SeaTools tests.
It is certainly possible even probably to lose a drive. Although after all the problems, I can understand why you would consider the unit being at fault, but the fact is that drive failure must be investigated and eliminated as the cause.
Things are still very wobbly. I'm going to start pressing NetGear and Amazon for a refund this week if things cannot be stabilized. The failures and instabilities continue. Here's the latest.
I have no way to test the drives outside of the NAS... all we have in our household are laptops, which I think is probably a very common situation these days. I think it is kind of "above and beyond" for NetGear to expect their customers to pull a drive out of the NAS, install it in a (working) PC (without messing up the existing OS), install and run the drive manufacturer's disk utilities, etc. And the bottom line is I would have to BUY a computer to do this...
As requested by Josh, I removed all the drives from the NAS, moved disk 2 into slot 1, and rebooted the NAS in "disk test" mode. Apparently the drive passed. I then did the same on disk 3 (as it had been a frequent "culprit" in the past), and it seemingly also passed. I then put all the drives back into the enclosure, and rebooted again. The system came up successfully, but said it was rebuilding drive 3. Which was weird, because drive 2 was the one that had "hardware failed". Drive 2 is now magically working fine. Drive 3 rebuilt, and everything seemed to be fine. Then the scheduled volume scan ran and said there was a problem. I rebooted with the volume check, and everything is now OK.
So why did disk 2 SEEMINGLY completely fail, and y'all thought it was a hardware failure on the drive, and now it's working fine???? That is NOT a typical hardware failure behavior...
And now I can't access my files for writing from our Macs... all of the AFP shares appear to be locked down as Read-Only. In FrontView, the default AFP permission is still Read/Write. I can't reset any of the file or folder permissions from the Mac, and disabling and re-enabling AFP on the shares hasn't helped either.
And still no contact from NetGear. I am so disappointed in this unit. I'll be asking Amazon for a refund. I've had the unit since March, and it's been broken longer than it has been operational.
If anyone is interested in purchasing this unit from me, please make me an offer. I'm not sure what's wrong with it... gut feeling is it's simply a very fouled up configuration, and nothing to do with hardware, since the issues and failures keep bouncing around so much. Time to move on... I need to find something else that will work for me.
It's an Ultra 4 Plus, model RNDP400U. It was purchased in March, but the original unit was replaced by NetGear in April.
No. I've offered that up to tech support at least twice, and they've never told me to do it.
Seriously, I'm done with NetGear - I'm not spending any more time on it. I just ordered a Synology. I'll try to sell this unit, as it doesn't look like Amazon will take it back, even though it's definitely been a lemon.
Over the past 10 months, I have had nearly the same issues as JamesHaskell has described but with a pro pioneer. I however have not sought technical support and instead have dealt with such issues through continuous software rebuilds.
