Dependancy failed for local file system. After July update

To add to Sam’s request - the above would need to be done with the kernel reverted back to the original non-booting newer version, which is 4.1.3-7-osmc. So follow the same process of replacing the two files, but use the two files from this version instead of 3.14.37-14-osmc.

Then in conjunction with the change in uEnv.txt it should go back to failing to boot properly but there should be a lot more on the screen than just “Depedency Failed for Local File System”. Please take a picture of whatever it is, even if it keeps scrolling.

When finished you can revert the changes to get your system running again.

[TIME] Timed out waiting for the device dev-mmcblk0p1.device.
[DEPEND] Dependency failed for /boot.
[DEPEND] Dependency failed for Local File Systems.

I’ve put the Vero in the cupboard and got the old R Pi B running OSMC again. Tried multiple cards. Can’t get logs as cant SSH in.

I am still unable to replicate this and would be very appreciative of the logs / screen captures which can be obtained from above instructions.

[TIME] Timed out waiting for the device dev-mmcblk0p1.device.
[DEPEND] Dependency failed for /boot.
[DEPEND] Dependency failed for Local File Systems.

Tried multiple cards and several updates and all the same story. Dug the old RPi B out and put Vero in the cupboard until this gets sorted.

Changing it to noquiet, the Vero just spammed “systemd[1]: Time has been changed” 20-30 times a second.

Changing it back to quiet, I got this output:

Sam,
Sorry been busy with work and haven’t had a chance to recreate this for you but looking at the latest comment as I has the same experience where on logging level 7 the screen was flooded with messaging at least 20-30 per second if not more then I backed it down to level 6 and posted my screenshot towards the top of the chat. I have to ask should it be flooding the screen? If not that sounds like a fantastic spot to start looking.

Hi

As I haven’t seen the actual output of the screen when it is in loglevel=7, the cause of the problem can only be speculated upon. There are some new fixes in the Vero kernel tree which may address this. We’ll have to wait until our website is back online before we can start building again.

Sam

So, my previous post with loglevel=7 doesn’t help at all?

No, unfortunately I can’t see the cause of the problem from that. Let’s wait for servers to come online and we can test the new kernel.

Sam

Hello all,

I am also experiencing the above mentioned issue. What I did not see mentioned up to now is that I have a Revision 1 Vero (the one were you have to extract the microSD card with your finger nail, or tweezers.)

Hope this helps with the diagnostics.

Terence

Hi

I think this issue has been resolved.

I am just investigating a nasty GPU stall bug and preparing Kodi 15.1 for the August update.

When that is out, the idea will be to download the August update or the June update and ‘upgrade’ directly to the August update, skipping the July one altogether.

Sam

My Vero is also an early one without the spring loaded SD eject but much like Sam, I don’t experience this problem. So I don’t think its that. Thanks for the info anyway.

I’m getting the same error after the August update.
After changing back to the June kernel (3.14), I first got a sadface (when creating skin shortcuts or something), then everything was back to normal.
I’ll probably have to stick with the 3.14 kernel forever :stuck_out_tongue:

I also can confirm that the August update does NOT resolve the issue!

I am sorry, but you’ve got to be kidding!
After more than a month, you still don’t get this kernel issue resolved?
Unlike others, I don’t find that funny at all!
If it was occurring on a Pi2 I would understand, but this is YOUR OWN hardware design!

I am really starting to regret having gotten a VERO in the first place!
For half of the price, I could have a PI2 running KODI, like a friend of mine has.
But I was so amazed and appreciated the work you did with crystalbuntu, that I wanted to invest into your OSMC flagship system.

OK, I get that this is not good

I am unable to replicate this issue locally. No one on our development team can replicate it with their Vero either. This makes it extremely difficult to identify and resolve this problem. As such, we have to rely on community feedback as we cannot confirm the issue is resolved at our end. It’s not a matter of having a month to fix it, it’s a matter of having a month where we could not reproduce this problem locally and only had this thread to go on. Some udelays were added, there was a patch to DTB to fix the 1.8v dummy regulator issue, we thought this must have exhausted any potential issues, but it appears you are still having an issue.

When I saw this thread initially, I was sure it was an issue with the SD card. Statistically speaking, this issue has affected less than one in several thousand Vero users if we assume that all affected users would report this issue within two months of this update being available. The only reason I’m not convinced it’s an SD card issue is that you report that upgrading is causing the issue and that your Vero otherwise works fine.

Unfortunately, the feedback here doesn’t give us enough to go on. I asked @The_Myst for some output of noquiet and loglevel=7 and he said that the log was too full. The output of ‘quiet’ doesn’t give me (or any others) an indication of what the issue could possibly be. I still haven’t seen a proper log or backtrace, although I appreciate it may be hard for you to get this. In the meantime, we’ll keep working on it, and trying to work out what may be causing this. For now, I think the most practical solution would be if one of you want to send me an SD card (Vero too will help). I’ll send you one back that will definitely work (and I am sure sales can offer something nice in return).

Thank you, and I do appreciate that. I can understand your frustration, you just want something that works. I don’t like seeing things like this on our forum, and it is not what I am trying to achieve with the Vero. I keep a couple of units aside from every Vero batch, so if there are any irregularities we can quickly isolate the situation, work out the number of users affected and act accordingly. Unfortunately, I cannot replicate this issue on any revision of Vero we have shipped. It’s pissing me off as much as it is you.

For now, I’d recommend you stick with the June update. If anyone wants to ship back their Vero and SD card for me to have a look at, I’ll return you a working one and it will greatly help us work out what’s gone wrong.

We will work this issue out.

Best

Sam

I just performed the last update. I have the “Dependancy failed” error.
I already had it las month with te previous update.
You really need to find a way to fix this Sam.
I saw the available update a few days ago and i was hesitant to perform the update not to break my Vero like last month. Finally i decided to do it. Now i really regret it.
What can i do to help you fix this?

Sam,
I have not had time to run any test lately for you (and I knew how unuseful level 7 was lol) but if you want to send me a address to mail a “problematic” vero to I would prefer you had one to test future updates against. The wife has already made me switch out the Vero with another kodi box.

Sean

Hi Sean

I have got your email. I’ll follow up on that shortly.

Sam

People,

I appreciate that this is very frustrating, but cut Sam and the team some slack here. As a developer myself I know that the first step in reliably squashing a bug is being able to reproduce it. And not for lack of trying, they just haven’t been able to.

I myself also still suffer from this issue after the August update (my CEC remote also no longer works, but that may or may not be a related issue. I can use the Vero remote instead, so no hurt there.)

@Sam: I am assuming you have access to my email address. If so, just let me know where I need to mail my Vero to. I am assuming the machine alone is enough, as any SD card with the July of August update on said machine will give the same results.

Greetings,
Terence

I am going to source a unit from @altesean and will let you know how I get on. As this issue could potentially be SD related, it would be good to get that too

Sam