Cant get into my Rockstor machine anymore Ive made a mistake

Ive come a across an issue where i cannot get into my machine any more.
I changed the “root” share owner from root to my username and since then Im not able to access it from my GUI. I can access it from the machine itself but i dont know how to fix my issue, can anyone help?


I made a rockstor iso usb and ran the troubleshooter and this is what came up.IMG_20200622_194121931

@Emmanuel_perez Hello again.

Yes, we haven’t ‘surfaced’ this ‘root’ subvolume for quite some time now as it turned out to be a bug and lent itself to this kind of mistake/miss-hap. It may also be that your entire root system, not just the /root dir has had it’s ownership changed. And given this is a tricky thing to put back, without boot to snapshot which we have in our new beta testing openSUSE variant by the way, your best bet I think is to do a re-install.

Sorry to be the bearer of such news, and from your screen shots it does look like your system disk config is toast. This ‘surfacing’ of the ambiguous root subvol was fixed as a bug in June 2018:

And released in the Stable Channel 14th June 2018:

If you are familiar with the linux command line (or interested in learning/persevering), ideally careful & patient, then it may be our new beta stage testing channel rpm will be of interest. If you are OK jumping through a hoop or two to get it all setup. Otherwise if this is not that urgent then we should have a new installer out fairly soon based on the work in that openSUSE based testing channel. Just remember to disconnect physically all your data drives with the machine powered off before doing any re-installs as that way you can’t accidentally overwrite your data pool/s.

Hope that helps.

Thank you very much for taking the time to respond to my issue. I myself am a visual learner and its hard to pick up linux for me. As you can see that I start tinkering with things that im not supposed to. Id like to get my machine up and running again.

If I do a new install, would i be able to get my raid array back up and running? Thats what im worried about the most. Ive got 4 drives in raid 10. Also have two other drivers raid 1 for plex.

Also, the fix that would have prevented all of this, would that have been included in the paid version of rockstor?

That was really the nature of the bug, in that it exposed a system critical subvol. But of course you do have all powers of the system anyway so you will always be able to break stuff. The confusion here is that, on CentOS btrfs arrangement, there is a root subvol which contains the system, but there is also a root directory within linux systems “/root” that is the root users home. Those two elements were confused in the earlier code and provided for creating a subtle set of bugs. So I just removed it from the Web-UI as I believe it was initially a mistake that it emerged in the first place.

Yes, you would import them after the install. But best to pick you update channel and apply all updates before you do that though as then you get newer import code. The following may be of help for the reinstall process:

Reinstalling Rockstor:

That fix was added after we had dropped the CentOS based testing channel development, so yes it only appeared in our Stable channel updates subscription service. But since we have now restarted the testing channel, but for our pending openSUSE based variant only, it is now in there also. See the following for the background on all of this:

Specifically the “Quick history” subsection.

Hope that helps.

@phillxnet Ive got my system up and running!
I swapped the OS HDD and reinstalled Rockstor into it. Then just reactivated my drives and re did my SMB shares and rockons.

I really appreciate your help. Also I would like to upgrade to stable updates soon. Do i just pay for the subscription and it’ll update itself or is there anything else i need to do?

Thank you for taking the time to help me out with my issue

@Emmanuel_perez Thanks for the update and glad you got this sorted.


The Update Channels doc section should cover this hopefully:
and more specifically the Stable Channel subsection:

Esentially you just subscribe the the Stable Channel updates service from within the Web-UI, you can then update the rockstor related rpms from that channel.

But given there are now hundreds of MB’s worth of upstream changes it can take quite some time, like 20 mins or more, depending on your system/internet speed. Assuming you haven’t already applied these of course. So having first subscribed it is now advisable to invoke this update from the command line as then you can see it’s progress. The Rockstor way of monitoring this ends up timing out unfortunately. This has lead to folks rebooting part way through what is actually a distribution update and that usually leaves thing in a little bit of a mess. So either wait a goodly while before you reboot or go the command line route via:

yum update

In both cases the Rockstor Web-UI will likely be unavailable for a few minutes as the rockstor package stops the related services and then starts them up again after it itself is updated.

One more caveat, if you are moving from the now legacy/discontinued CentOS based testing channel then you may be affected by this:

Not a problem though if you are updating directly from the admittedly old but functional install ISO.

Hope that helps. The stable channel subscriptions model is our attempt at sustainable open source software development and is currently critical to our efforts as they stand.


Ive got it! @phillxnet Ive updated to stable channel and “quotas” come up on my pools. Thank you for your help!