Dependancy failed for local file system. After July update

Hi Tom,

In no way have I suggested that, but I have suggested that the issue you experienced was most likely not related to the one described in this thread.

Yes, because you said that you saw the words ‘Dependency failed for local file system’ on your unit. Unfortunately the device you returned was missing files on /boot (you say it was formatted). I understand you were concerned about using the installer, as you felt that it could cause harm to your computer.

It is standard policy to return items without diagnosed fault at the expense of the buyer.

P.S. I noticed that the power supply is an older one however, and I will replace it with a newer one.

It will be with you tomorrow. In the interim, I have received an email from @fxschaeffer and @AdR and will follow up with them

Best,

Sam

Hi,

Unfortunately, I am still without an affected unit but I have opened up a dialog with @fxschaeffer so hopefully something good will come of that. In the mean time, some more information will help narrow down this problem significantly.

These steps should be followed:

Installing latest OSMC

  • Download and image the OSMC June update from Download - OSMC
  • With the OSMC SD card (official), boot and complete the install
  • SSH in, edit /etc/apt/sources.list
  • Change
deb http://apt.osmc.tv jessie main

to

deb http://apt.osmc.tv jessie-devel main

And run the following command:

sudo apt-get update 
sudo apt-get -y dist-upgrade
sudo reboot

If the boot fails

If the boot now fails, plug the SD card in to the computer. You will see a file called uEnv.txt. Open it, and add the following to the end of the line:

rootflags=nofail rd.fstab=0

If this still fails, please add loglevel=7 and try and video the boot process.

If this still fails, please remove the HDMI cable and try and boot again. Can you ping the device?

If this still fails, please try another non-Sandisk SD card and follow the above instructions again. Note that some SD cards may not be SanDisk branded, but still use a SanDisk based controller.

Sam

1 Like

Dear All,

Just wanted to keep you all up to date:

Sam,

I am wondering why I still haven’t heard from you!

Anyway, I have given this whole thing a couple of thoughts and have come a decision:
When you announced OSMC and the Vero last year, it seemed to me as a good idea and perfect solution for my multimedia demands at home. So, I decided to support you in this endeavour not just by donating a couple of bucks for OSMC and get myself a Pi2, but instead I bought a Vero and therefore became one of your early sponsors and biggest donors. For that, I expected you receive a proper working device with premium support on the road ahead. I had at that time no doubt, you would run a proper business and trusted you with a lot of money for this completely unknown device.

In the last couple of weeks, I followed the discussions between yourself and other of your customers who are dealing with the same dependancies problems as I do with my Vero. Like Tom, I find myself as amazed and confused. We are not your beta testers, but paying customers to whom you have sold a product for which it is your responsibility to guarantee a working condition, even after updates. It is neither our fault, nor our responsibility to fix it, but yours!

I have given you more than three tries to solve those constant and reproducible dependancy problems which my Vero encounters, whenever I trie to run a post June revision of OSMC. You clearly have failed to solve the problem and openly admit, that you are not able to. Therefore, I herby declare a failed guarantee case.

I have decided to return the product to you and hereby demand a full refund! (Attached you will find the details of my order.)
Please send me a link to an already paid for, trackable and pre labeled return sticker so I can ship the Vero ASAP.

In the hope to settle this very unpleasant business once and for all.

Regards,
Adrian

Hi Adrian,

I am currently working with another forum member who is affected by the issue to determine the cause of the problem and resolve it in a timely manner.

We are not your beta testers, but paying customers to whom you have sold a product for which it is your responsibility to >guarantee a working condition, even after updates. It is neither our fault, nor our responsibility to fix it, but yours!

Absolutely understood – but you must understand that without a device at hand to replicate the issue nor anyone’s log file, we cannot resolve, or even confirm what the issue is. Tom’s Vero was simply imaged incorrectly, and a re-image brought it back to life. A ‘proper business’ as you say cannot incur costs for user error, so we simply charged him the shipping cost of the device back. This is standard policy with most companies.

I have decided to return the product to you and hereby demand a full refund! (Attached you will find the details of my >order.)

I am afraid that is not possible. You can return your unit and we can replace it (if we can find an issue with the device). We cannot refund you for the item. It is still working as sold and the software that shipped with the device runs on it fine. Therefore the guarantee has not failed. We will fix the issue, but for now, if you’re still afflicted by it, either send it back so we can work with an affected unit hands on, follow the troubleshooting steps I continue to post in the forum to try and get some insight, or revert to an older version until it’s fixed.

I will keep you informed when the dependency issue is resolved

Sam

Thank you Adrian,

I also kept much of my comunications with Sam to email in order to give him the benefit of the doubt. I find his continued dismissal of my problem as user error to be rather offensive.

  • I bought the device and it worked
  • I upgraded it through his GUI in July and it failed as per this thread
  • I reverted to the Jun image using a fairly recognised technique and it worked perfectly
  • It always failed with the same symptoms when I upgraded, whether through the GUI or installing a fresh image

Good luck with your situation; with hindsight I would have probably gone for a RPi

Yours, Tom

Hi Tom,

I did not solely dismiss your issue as user error. I found in my extensive testing that the card was formatted incorrectly, and when re-imaged the unit worked fine. I therefore consider the issue to either be specific to some environments, or just the consequence of some bad luck when imaging the card.

The correct installation procedure is to use the OSMC installer. I would also like to take this opportunity to remind you that is clean (as per your concerns in the email).

You should not be offended that I assumed an alternative imaging method has the potential to yield an incorrect installation.

I am waiting for @fxschaeffer to do some testing of the above instructions and hopefully put this issue to bed once and for all.

Best

Sam

Hi,
I’m currently speaking with Sam via email to check that i am really affected by this issue before sending my Vero back.
You will here from Sam and me soon.
Regards,
Xavier

Dear Tom,

yes, indeed! A RPi2 at half the price of a Vero would certainly the better choise!

Obviously, Sam doesn’t get that we are not some bunch of geeks collaborating on his project but paying customers to whom he owes since without us, his little startup would not exist in the first place.

Let’s face it!
Stuck in the current state, our Veros are just not worth the price we paid for them and they never will be!
So, our investment in the Vero is gone and we will never get a working device nor our money back!

Why? Well consider the facts!

More than twice the announced fix to our problem proofed to be a complete failure which did in fact not solve our problem anyway. Although claiming to have a device of each batch available, Sam is incapable to reproduce the problem on any of his devices and admits to be clueless. So, without a faulty device, there’s no fix in sight.

Like you, I also have offered to send in my device to help him out on this. Although promising a solution, to this day Sam never came back to me with a concrete exchange procedure. When I read your reports I finally decided, that the Vero was just not for me and under claim of guarantee demanded my money back.

Sam simply refused this and hilariously told me I had no claim, since the device worked properly with the firmware it originally was shipped with. Yeah, that beta piece of extra buggy fun!

If the cyberattacks of 2013 and 2014 (like Heartbleed etc.) have taught us anything, then that any piece of hardware depending on online connections - computers, smartphones, tablets or even multimedia devices like a Vero - can only be used securely as long as they are being updated regularly.
Otherwise, they are open gates to your local network and are therefore a clear and present danger to your private data and your identities/passwords!

So, a safe hardware product relies on CONSTANT WORKING FIRMWARE UPDATES!

If by now you consider to get a RPi2 and sell your Vero on eBay for a reasonable price, well good that with that!
Who will be willing to pay that when you have to tell potential buyers that they have to live with the fact that they can only run an outdated version with no chance to update to the current firmware (and might never will be able to)?

Quite a stuck situation we are in!
So consider this for a moment: If this was a device from Samsung, Apple or Amazon, you would definitely and rightfully demand an exchange or your money back! And you know what? Any of these companies would have no problem to comply to your demands! Probably they would offer a general recall to all customers with faulty devices!

The manner in which Sam is dealing with this problem and us is totally unprofessional and unacceptable!

Sad but true: Sam is just another small startup runner who can’t get to base!
Like Ponzi and Madoff he keeps on promising things he just cannot deliver (like OSMC for ATV1 for example!) to those willing to believe!

Good luck to you as well.

Yours, Adrian

In fairness to Sam, my Vero is now working and accepting updates (though I’m likely to hold my breath on each occasion in the future). My main concern was over his attitude and therfore the service provided. If it was just a software issue it really shouldn’t require a unit return to resolve. If you want it fixed then you may want to consider sending it him, albeit accepting the fact that he will probably expect you to cover postage in both directions.
Tom

Return post only if the Vero turns out to be defective.

Mr. Madison, what you have just said is one of the most insanely idiotic things I have ever heard. At no point in your rambling, incoherent response were you even close to anything that could be considered a rational thought. Everyone in this room is now dumber for having listened to it. I award you no points, and may God have mercy on your soul.

~ from Billy Madison (1995)

Sounds like Sam is actually willing to send you a new device. But you refuse to send your current one back?

Adrian,

I understand your concern here. Allow me to clarify some things:

Indeed – but informing me that things ‘don’t work’ isn’t enough for me to work out the problem. It is much akin to telling a car mechanic that your car no longer starts. There are a multitude of reasons why this could be the case.

Judging by recent comments, you seem very concerned with reselling the Vero. This is perhaps why you have refused (in email) to have your unit repaired, or if this is not possible, replaced free of charge. To reiterate: I am happy to accept your unit back and replace it with a working model. Unfortunately you have turned this down and have repeatedly asked for a full refund, plus reimbursement, which seems excessive.

We don’t recommend port-forwarding in OSMC, but OSMC tracks all security updates aggressively. These updates are not tied to your kernel and can be applied without issue.

I said it before, and I’ll say it again, I will fix this, if it is a software issue. @fxschaeffer sent me a log today (following the above instructions) and I was able to quickly identify the issue. It’s power sequencing related. He had his unit plugged in to an extension cord. Since changing, he’s able to get further in the boot. I think it could be an issue with his power supply. I will be sending him a replacement one shortly and we will see how he gets on

OSMC builds for AppleTV are already available, but they are experimental. I’m happy to send you a link, or you can grab one from http://download.osmc.tv directly (just look for ATV). Use at your own risk until we announce proper support in a week or so.

Tom,

In my email, I said it is our policy to charge shipment if an issue is not found. I replaced your power supply as it had an odd noise to it when I tested your unit. Upon further inspection, with my oscilloscope, I found it had ripple above what I’d consider tolerable. I consider this a fault, and likely, a cause of your failed boot. You will see a refund of shipping charges on your card within 7 days. I don’t see why you’d lie about such an issue. There is no point. I worked out that your card being incorrectly formatted was because you formatted under Windows, which does not expose all partitions, but I believe this was independent of your issue.

Our power supply manufacturer, Stontronics, manufacture the Raspberry Pi official power supply. I chose them based on this reputation. Unfortunately, it seems we may have been victim of a less than stellar batch. We had to send 1000 units back in May and use a different supplier. Here are my recommendations for those still affected:

  • Remove any extension cords and try directly at the wall
  • If you have another 5V charger between 2A and 3A, with a 2.5mmx2.5mm charger (2.1mm will work if you hold it in), then you can test with this.

You’re obviously wondering why the current kernel is problematic, when the former worked fine. Power sequencing has changed, and thus the Vero is more vulnerable to voltage stabilisation at the initial boot. This can potentially be fixed in software, but until we can do so, it’s indicative of a PSU problem. With that said – the installer which uses the same kernel is fine, which makes the issue certainly odd. I just wish I had an affected unit infront of me, then I could work it out for myself. I think Tom is the only affected user on 230v.

Adrian – afford me some respect. If you send your Vero back, I’ll send you a working one back. I can promise that.

Sam

Hmm… attitude and therefore the service provided.

Wait, THAT is the attitude you are complaining about?

That’s genuinely insane. You do realise that, right?

Sam, thank you for the refund of return shipping as a result of your diagnosis that the root cause was hardware, though it does not cover the c£10 for packaging and shipping recorded delivery to you at the outset.

I am rather confused by your suggestion that I lied. I have made efforts to reread all communications (particularly forum posts) before sending them to ensure they are factual and not overly emotional. As per my earlier post, I only formatted the sd card once when I sent it to you as I had no desire to send anything that might have had personal information about my network or other devices to you; the formatting could therefore not be part of the problem. You (and other members of your team) seem to be taking a poor customer relations situation and making it worse; this is even more difficult to understand as my earlier response to Adrian was, I thought, in general supportive of you.

Tom

I think you are misreading what I am saying. I am saying that there would be no point in you lying, which prompted me to give you the benefit of the doubt.

Please re-read my above post

Sam!
Now wait a second and let’s be honest here!
Neither in my posts here, nor in my e-mails to you did I ever refuse to send in my device!

In fact I did accept your offer multiple times, first in my post way back on Sep 15!

Sad but true: You ignored me and never came back with a concrete exchange procedure!
See my posts on Sep 22:

and on Sep 29:

After that I patiently kept waiting for your response, which never came. Instead you published more updates, promising they would solve the problem, yet they did not.

It was then, that I lost faith in your customer support completely.
That’s why, I decided to no longer bother myself with this whole Vero mess and opted for return and a full refund.
That was the content of my e-mail to you.


Anyway, I do want this over and done with! So I will grand you this last chance to exchange my not updatable Vero for a properly working one:

Please send me a brand new, boxed and thoroughly checked Vero (including a flawlessly working power supply) in advance. In addition, please include a paid for return sticker and I I will be ship my original Vero back to you, ASPAP.
If needed, I will provide both of my SD-Cards as well as log files or whatever you need.

Do afford me and my family some respect too!
After all my troubles and precious time I spent on this, I am simply not willing to tell my family that they will have to live without our central media player for a couple of weeks.
The costs for my second SD-Card should be roughly equivalent to the shipping costs you would have to pay for the return way. I think that’s a fair offer.

Please comply.

Adrian

Adrian,

Your first correspondence with OSMC’s email channel was on 20th October. There has only been one update since then. It’s a shame this has not resolved your issue.

As previously offered we can repair your unit or offer a replacement upon its receipt.

Let me know when you wish to proceed with this,

Best

Sam

Is there any update on this issue?

Hi,

@fxschaeffer received a replacement PSU and SD card today to try. I will chat with him on Skype to see how he gets on.

There was also a bug in Linux 4.1 upstream which mirrors the symptoms in this thread. I have built a new kernel with this fix included and if @fxschaeffer’s replacement peripherals don’t help, then I’ll get him to test the new kernel as well.

Cheers

Sam

Thanks for the update, @sam_nazarko! I am anxious to hear how he gets on.