Time Machine verification failure (again)

Did you open a support ticket / report a bug at support.netgear.com?

I tried but after over a week trying to explain and being told that I would have to pay for a technical support contract even to get the problem analysed to see if it was a software ot hardware fault, they sent me back here.

I just took my RN102 back to the local branch of the supplier I bought it from. They were quite perplexed and appalled at Netgear's attitude to their customers, especially the bizarre 3 month technical support vs 3 yesr hardware warranty policy which does not give a customer the possibility of demonstrating hardware failure.

However, they agreed with my suspiscions that this is an error in the OS which denies Time Machine to delete old incremental backups. They are replacing the RN102 and, like me, expect it to fail in about three months time as the last one did first time around. They said that as this will be new hardware, I should be entitled to three months technical support on Netgear's terms; so hopefully, if it fails again once the sparsebundle is full, it will be before the three months is up. If it does happen again and Netgear fail to respond appropriately again, another approach may be necessary.

If I had requested support when the first instance occurred, instead of simply doing as Time Machine 'helpfully' suggested, I might be on a better footing. In future, I won't hesitate in requesting customer support for every little issue because otherwise it is difficult to demonstrate when a problem started.

Very good, keep us posted.

As for myself, I have been contacted by one Netgear staff helper and supplied him the device Logs as requested.

So far so good. You can check my progress, in case, on the other thread I started.

While I was on the logs, I had a close look at errors and warning messages and made an interesting discovery.

I found several entries like this one:

Volume: Bit rot has detected an error within /data/.timemachine/Sputnik.sparsebundle/bands/40fb and cannot correct the error.

More or less once every start of the month and I suspect that that message matches with the Time Machine failures.

This might be a lead to for the issues I am suffering however such messages appear in the log until the beginning of the month April whereas my last backup failure was around mid september.

It really makes you wonder.

To be continued..

Yes, I am subscribed to your thread.

I remember reading a discussion on here some time ago about bitrot protection - I had searched for it because I got an alert about it but I'm struggling to remember what it said. It certainly doesn't sound a healthy thing for a backup to have.

I'm having the exact same issue, and getting very little support from Netgear. If you find a solution to the problem, please let us know what to do.

Thanks

Michael

Users of Apple's own Time Capsule run into this problem which would strongly indicate the problem is at Apple's end. Their own community shows that Time Capsule users run into this problem.

---

mdgm wrote:

Users of Apple's own Time Capsule run into this problem which would strongly indicate the problem is at Apple's end. Their own community shows that Time Capsule users run into this problem.

---

Blame game ping pong.

As an outsider who abandoned ReadyNAS due to this issue (but is still subscribed to this thread), I feel liberated to say exactly what I feel. The Netgear support was so abysmal that I took the RN102 back to the supplier who sold it to me and got a full refund a year after purchase because they could not fathom the problem.

Yes, there are a few cases of this error being displayed from Time Capsule users but having spent three years with this issue on two different ReadyNAS products I can say that the overwhelming majority of problems are using third party equipment and only a handful, if that, are Time Capsule users.

The reported error is generic (I may not be a developer but I was a prolific ResEdit explorer in the 80s and 90s), it does not specify the problem, so could apply to any number of causes so what Time Capsule users and ReadyNAS users are experiencing could be entirely different issues.

What I discovered before I gave up on ReadyNAS is that most third party NAS manufacturers advertise their products as Time Machine 'compatible' rather than Time Machine 'compliant' and that is the dirty little secret which nobody wants to talk about. There are legal discussions which establish the distinction between compatibility and compliance but suffice to say, compatibility does not necessarily mean that it will work - hence the perennial refutation that it's all Apple's fault.

The criteria for Time Machine compliance are laid out in this document:

https://developer.apple.com/library/mac/documentation/NetworkingInternetWeb/Conceptual/TimeMachineNetworkInterfaceSpecification/TimeMachineRequirements/TimeMachineRequirements.html

I am still treading cautiously with my Synology DS213 but what I have noticed is that it seems to have a slightly different approach to creating the backup and when the first backup was made, it took less than half the 36+ hours which my RN102 took.

I might finish by adding that the support on this forum is extremely selective and, I'd venture to suggest, targeted at low hanging fruit. I am not the only person who has played fetch with admins; i.e. asked for more information only to provide it and then be ignored. And I might add that you, mdgm, are the worst offender. And that you, mdgm, gave me advice which bricked my ReadyNAS Duo and then ignored me when I reported the fact back to the thread.

Good luck to anybody who has this problem; you will need it because NOBODY is going help you fix it. NOBODY.

Oh dear.

I am having the same issue backing up to a timemachine share on an RN102, and I have been having the same verification failure.

Given everyone else's experience, i do not have the time to waste on products that do not work; nor am I prepared to not have a backup that I can rely on.

Until the Ready NAS, I have always run 2 alternate stand alone USB disks to ensure I had a level of protection.

Looks like I will need to revert back to this again unless something has advanced over the lat 6 months since the last post.

Backups are not negotiable.

Today I updated my RN 312 from 6.5.1 to 6.5.2.

Some minutes after reboot, my Mac (OS X 10.11.6) reported "Time Machine completed a verification of your backups ... must create a new backup for you.".

I had this problem on my Mac for ages and ages too with the ReadyNAS.  I was looking around for another solution, then realised, that I hadn't had this issue for more tahn 2 months now - everything was begin backed up properl over that time.

I suspect that it has something to do with a Mac OS update, not a Netgear firmware update.  For your information, My current OS X version is 10.10.5.

Would be interested if it went away for anyone else....

Tommy.

I believe the verification issue is an Apple issue, not a Netgear or other NAS issue.

I purchased a WD Mycloud alternative and it did exactly the same thing.

To solve the issue I turned off the Automated timemachine backup process, and now run all of my timemachine backups manually, and have not had an issue since.

I believe Timemachine cannot cope with a loss of connection to the Netgear, or other NAS system, and basically corrupts itself.

I believe I have caused the issue in the past when I have disconnected my Macbook Pro from the network while it is in the middle of a backup process, not realising the backup process was running.  (I probably closed the lid and put it in sleep mode, and when I "woke" my MacBook at a customer site, it could not locate the backup disk, and corrupted itself.)

Obviously this is unacceptable in this day and age, but apple do not seem to have acknowledged the need to fix it.

Running Time machine backups manually has resolved the issue for me as I am in control of when they occur and ensure the NAS is available throughout the backup process.

In my view Time machine is not sophisticated enough to recover from a loss of backup disc connection.

I hope this provides you some assistance.

Message for Apple.   You really should engage someone with the necessary expertise to update TimeMachine to incorporate full roll-back recovery when there is a loss connection with the backup system

Actually, I run Time Machine with Apple's Time Capsule for my laptop. The 'problem' has only occurred twice in over two years. Whereas, in running Time Machine with ReadyNAS for my iMac, the 'problem' occures every week or so. This suggests the issue is with the Time Machine - ReadyNAS interface, and not inherently an Apple issue.

Actually, I had this happen SEVERAL times recently, also under heavy load conditions. In my case it was not a scrub, but instead it was running several rsyncs for backup/restore/migrating data -- though resyncs while adding/swapping disks can case it too. In reality, I don't think it's CPU that's the biggest issue in this case. In my usage, I found that increasing RAM had a bigger impact. It would seem that the way that AFP and TM thrash about, accessing the sparseimage bundle, requires a fair amount of data to be accessed. The NAS works a lot harder if it doesn't have sufficient memory to keep enough cached.

So until I knew I was out of the woods (more RAM), I completely disabled TM from doing automatic backups. Or it'd just complain over and over (after 1+ hour delays) that the backup was corrupt, even if it was empty.

So, YMMV, but that's what I've found. If the NAS is quiet, or has enough memory to avoid having to re-cache data it needs from the sparsebundle very frequently, then all is well. I will say that, regardless of NAS activity, I completely stop having this problem after increasing RAM. I will also point out that with my add ons, my background real usage is around 2-3GB. So use cases and actual workload are a factor to be considered.

I have experienced this problem several times now. About a year ago, Netgear support suggested that I turn on the Scrub feature to run 2x/year. Earlier this month, the problem recurred.

After reading a lot of support blogs, the consensus appears to tag this is primarily as a Time Machine issue. The solution for me has been to archive the failed sparesbundle (it can still be used to restore files), and let TM start again. I also use Carbon Copy Cloner as a (hopefully) failsafe backup to a USB disk. 

You may be correct.  The blame may sit with both apple and Netgear NAS interface solutions, but I have seen this issue reported with Apple's own Time capsule so it an issue even within the apple ecosystem.

At the end of the day, the apple timemachine software should be robust enough to protect itself from any backup unit that drops the connect for whatever reason.  Basic roll-back technology has been around a long time (ie. decades) and should be standard technology in backup solutions in 2017.

I expect that time-machine should roll-back, notify that the backup had failed and provide a reason for the failure for trouble-shooting.

That to me is an apple deficiency irrespective of NAS used as the target.

Anyway.  Just my opinion.  I expect better from Apple.

I am having the same problem.  Verification issue.  Worked great for 6 months until i upgraded to Sierra.  Now this problem.  Dont know whether Sierra is the cause or just coincidence.

Nope it's working fine for me on Sierra. But note that I was experiencing this issue until I added more RAM to my NAS. It's the one repetive symtpom I've observed in every one of my installations. So much so that I completely disabled use of TM until I'd added more RAM. This is controversial because doing so will void your warranty. But in practice I find it makes my NAS adventures far more pleasurable and error free.

I still think it is an apple time machine and file system issue.

I will be interested to see if anyone suffers this issue now that Apple is moving to a more robust (supposedly) file system under the newly announced file system.

Until then, I continue to disable the auto scheduled time machine backup and only run it when I have a controlled environment.

I've not seen anything definitive about how APFS will change Time Machine, but it's pretty safe to say it will have some pretty significant impacts. Some of the HFS+ tricks it uses to track files won't exacdtly work the same, and there are better ways to manage it with APFS. Only time will tell, of course. I've heard some bits to suggest AFS (netatalk) will no longer be used... but not sure if CIFS is the way they'll go. Also something else to see...

I will be somewhat cautious about upgrading to High Sierra though, until the TIme Machine impact and any downstream issues with NAS'es are resolved. This will not be a quiet change for some users....

---

btaroli wrote:

 

I will be somewhat cautious about upgrading to High Sierra though, until the TIme Machine impact and any downstream issues with NAS'es are resolved. This will not be a quiet change for some users....

---

History says you likely are right.  Usually a major OSX release from Apple drives all the other vendors crazy for a while.