ReadyNas 6.2 (NAS locks up entirely)

I posted a (long) while back that my RN104 NAS freeze problem was solved as of Beta 3... however, I've been lurking behind the scenes on this form ever since then because my unit froze shortly after I posted that last note.

I've tried all the betas, each one freezes. Sometimes its a matter of hours, sometimes it's a matter of days, but it still freezes. It's been 4 months since I've had a useable, reliable NAS!

I've decided that I can't stand it anymore. Concerns of data integrity, a useless NAS, and a total loss of trust in Netgear's products and their ability to stand behind them have caused me to purchase a new NAS from a different vendor.

Netgear says they are working hard on a solution and that they are monitoring a few units remotely to determine why they are freezing. That was a month ago now. Where's the status? We've seen no apparent progress that Netgear is any closer to figuring this out.

Reading how Netgear states that other people's units are working great with 6.2.x doesn't help me. My unit isn't working, along with a significant number of others that seem to have the same issue. No replacements units are offered because Netgear is claiming they would not be able to reproduce the issue on their end. Did they even try to verify that claim? I refuse to pay for Netgear support because I know that they don't have a solution. If they did, they would fix it with a new firmware release.

I used to big a big Netgear fan; I'd easily choose Netgear products over another vendor's simply because I trusted the name. That is no longer the case; the pendulum has completely swung the other way. I'm never buying another Netgear product again. I can't convey adequately how disappointed I am.

Good luck to all that are still hanging on here hoping for a solution. I'm bailing out.

Another day another lock up. This time after starting backup job. It's ridiculous. I know you working hard and I know it's hard find so nasty bug (bugs), but i think you should provide patch to downgrade software to last stable relase (yes, yes no security patch no new feature but at last is working like a charm).

To all who want to go back to 6.1.x, please PM mdgm for method to downgrade.
Please note you WILL LOOSE all data on the NAS as it requires Factory Default.

Like @sauronek i waiting for a patch to downgrade since 3months now... I really don't care about security if the nas can work again on my closed local network... And i really don't understand how work Netgear dev... More than 3 month to find the issue, why not restart from last stable release and reapply one by one each new feature to find the issue ? Code Versionning exist for this...

We have already tried in some 6.2.3 beta releases reverting as much as possible back to how it was in 6.1.9 and on those beta releases we have still got reports of lockups from some users in this thread.

Rify, we do keep many internal development builds. What you suggest would make perfect sense if we could reliably reproduce the lockups that have been experienced by users in this thread in house. That's one of the things that has been so frustrating for us as the first step in resolving an issue typically is to reproduce it.

We would like to have a few systems shipped to us to be examined. If you are in the US and are still experiencing lockups please let me know and I will provide more details via PM.

I've been getting lockups regularly since going to the latest beta. Maybe once every couple of days - not a huge problem for me just inconvenient.
But, if it is at all helpful, I just exported my logs immediately after restarting following a lockup and am happy to send them to you.

Thanks,
Kevin

I have a 104 running the latest beta that continually freezes, sometimes after a day, then sometimes after an hour. It has been doing this for quite a while and i keep loading the latest beta version of the 6.2.x firmware to try and solve it, all to no avail. The only way to make the system usable again is to unplug the power cord and then power it back up, do you guys have any idea why it is doing this ?? This sort of unreliability was never around under 6.1 .. can i downgrade ? If i do, do i loose all my data ?? not ideal but i suppose better than the pita issue i have now? thoughts ??

mdgm wrote:

We have already tried in some 6.2.3 beta releases reverting as much as possible back to how it was in 6.1.9 and on those beta releases we have still got reports of lockups from some users in this thread. Rify, we do keep many internal development builds. What you suggest would make perfect sense if we could reliably reproduce the lockups that have been experienced by users in this thread in house. That's one of the things that has been so frustrating for us as the first step in resolving an issue typically is to reproduce it. We would like to have a few systems shipped to us to be examined. If you are in the US and are still experiencing lockups please let me know and I will provide more details via PM.

And if the first step in resolving an issue does not work, the next step then should be to systematically compare for any similarities of the systems with issues. One such thing is that all are RN10x. Are all specifically RN104's maybe? With 4 disks (all slots actually in use) perhaps? Could we perhaps get a beta with an option to get rid of the btrfs and format/use EXT3/4 instead to see if the problem is related to some internals of the 'fantasic' btr filesystem? I don't know, lots of 'things' can be the origin, but at least you know it has something to do with what's new or changed from 6.1.9 to 6.2.0.'
Have you done any kind of real thorough and systematic comparing of systems/configurations to narrow down the origin of this lockup-problem? (Other than trying to, qoute: reverting as much as possible)

I have a feeling that you are running around somewhat blindfolded and playing some kind of qualified guessing game doing some tweaking and tuning here and there which may or not may make the system more stable for SOME users. We got a range of various beta's where some seems to fix the problem for some users, while the same betas will make the system even worse for others. For instance like: A beta fixes the issues, everything looks good, and then one of the next betas f'**ks the very same system up again (or perhaps introdues a new issue??) I've been rock solid on T1721 since that one was put out, and now I do not DARE to even TRY anything newer or older, because of this 'guessing game'...I will stick with T1721 forever unless you guys can confirm 100% that the issue indeed has been located and dealth with. (or if a rollback to 6.1.9 is to become Netgear's offical policy to get a reliable RN10x-system)

Responses in this tread to all your betaefforts can be summed up like this. None of them addresses any actual or specific issue, the origin of the problem. They just makes some of the problematic systems a bit more or less reliable. In fact, this means (at least it looks like that to us the users) that you are no closer to any confirmed solution or fix than you where 3 or 4 months ago. You actually haven't got a clue on what there is to be fixed, have you? (Please tell me I'm wrong:) )

ArneG wrote:
And if the first step in resolving an issue does not work, the next step then should be to systematically compare for any similarities of the systems with issues. One such thing is that all are RN10x. Are all specifically RN104's maybe? With 4 disks (all slots actually in use) perhaps?

We have been looking into similarities and have fixed some lockup issues that some users have encountered but there is still more to go.

ArneG wrote:
Could we perhaps get a beta with an option to get rid of the btrfs and format/use EXT3/4 instead to see if the problem is related to some internals of the 'fantasic' btr filesystem?

That's not something that's going to happen. If the choice filesystem was inherently the problem then users would not have been claiming that things worked fine before the upgrade to 6.2. Furthermore EXT3 and EXT4 are inferior filesystems with their own issues.

ArneG wrote:
I don't know, lots of 'things' can be the origin, but at least you know it has something to do with what's new or changed from 6.1.9 to 6.2.0.

That appears to most likely be the case and our troubleshooting to date has proceeded with that assumption, but we do have to keep an open mind.

ArneG wrote:
Have you done any kind of real thorough and systematic comparing of systems/configurations to narrow down the origin of this lockup-problem? (Other than trying to, qoute: reverting as much as possible)

Logs, crash dumps, user comments on what they do with their NAS, configuration have all been looked into.

ArneG wrote:
where some seems to fix the problem for some users, while the same betas will make the system even worse for others

Which makes threads like these difficult to manage because issues that are different may appear to be the same to users so it can be very difficult/impossible to keep track of where different users are at.

We are in communication with users and 3rd parties to assist with diagnosing this. We are doing everything we can to get to the bottom of what's going on.

mdgm wrote:
Which makes threads like these difficult to manage because issues that are different may appear to be the same to users so it can be very difficult/impossible to keep track of where different users are at. We are in communication with users and 3rd parties to assist with diagnosing this. We are doing everything we can to get to the bottom of what's going on.

Ok, so you are absolutely sure that there is not ONE mainproblem at the bottom of this?

And your hypothesis is that the upgrade from 6.1.x to 6.2.x introduced a number of DIFFERENT issues for 'a random number of RN10x-boxes', which all results in the SAME scenario; the RN10x freezes without anything useful logged and the only way to 'fix'' the crash is to pull the powercord?? I find that hard to believe. Many different issues and bugs introduced by a new major version, yes, ok, that could be understandable (and should be fixable). But so many DIFFERENT and separated issues which all of them literally kills the box WITHOUT any warning or traces in the log?

Stick with the open mind theory, I wouldn't be supprised it it turns out that you couldn't see the forest because of all the trees blocking your view...