TrueNAS GUI not loading after exporting pool during data restore attempt.

[pdev1993]

I've got two problems I'm hoping someone on here will be able to help me with. (keep in mind, I'm pretty new to TrueNAS and what happened was out of my inexperience and stupidity). I'm kinda desperate and panicking because this affects a LOT of data I have collected over the past couple years and, while none of it is "crucial", I spent a LOT of time, money and effort getting it. Getting as much back as I can is very important to me.

Config:

OS	TrueNAS Scale Cobia, installed on 500GB SSD
Pool Configuration	1st Pool - 2x10TB drives, Mirrored, Snaphots Enabled 2nd Pool - 2x6TB drives, Mirrored, Snapshots Enabled
Memory	24GB

If you need more let me know

Background:
I was moving a large amount of media 6TB from a folder ("Media" renamed to "_Media") under dataset ("MediaA") on the first pool over to a new Dataset (also called "Media" after I renamed the folder to "_Media", we can call it "newMedia" to keep it straight here) which was a child of the MediaA dataset via the CloudCommander app (a file management docker, if you're not familiar) and had started a move process as there wasn't enough free space to do an outright full copy. I closed out of the box showing the progress, not expecting that it would end up cancelling the entire process. Afterwards, when I looked at the source folder via SMB (and in the CloudCommander app), everything was gone (cue the start of the freakout process). I was going to rollback from a Snapshot via SMB and thought I should do it from the TrueNAS GUI instead because it would be faster. When I rolled back with the Snapshot (which was from earlier this morning, before this all started, so that "_Media" folder was named "Media", matching that new dataset), I could see the Used Space for the Pool was around where it should be, however I still couldn't see the files I lost and the "newMedia" dataset was still there, with only the files that were present before the rollback. Doing some research online, I couldn't exactly find anything that would allow me to restore the files in any other way and I figured deleting the "newData" Dataset would likely destroy what little chance I have of getting that data back. I figured I'd try exporting, detaching and reimporting the pool would be worth a try (There was no visible task running regarding the Snapshot roll back)...which led to my second problem. After detaching the pool (I didn't delete any configurations, I just exported it), the GUI stopped working. It won't load, even after restarting the server. I can access the CLI and shell directly on the server, where I did try reimporting my pool but I don't think I'm doing it right, since it doesn't seem to help.

So my two problems here are:

1. The GUI won't load for use on a remote computer, the CLI/Shell does at the server itself. There are no errors displayed that I can see when the server boots.
2. I need to recover what data I can from the 1st Storage Pool. Rolling back the Snapshot seemed successful and the appropriate amount of data is being used but it's not visible in any place that I can see (even using the Linux shell to show the directory on the pool doesn't show the files I'm looking for). I can't provide any screenshots until the above problem is solved.

Please, whatever help I can get here would be greatly appreciated.

 

[pdev1993]

OK I resolved the first problem. Just needed to restart the middleware service. On to the second issue:
So to better visualize this, the MediaA dataset is where both the "_Media" folder and "Media" dataset are/were located. IF all of the data I'm looking for were removed, then the MediaA dataset would not be remotely close to 6.21TB of used space (as shown below).

The odd part here is that the only dataset under MediaA is "Media" (or "newMedia" as we're calling it here),the data usage of which isn't matching up to the overall size of its parent Dataset. Looking at the content through the shell, it's only showing some of the folders that were supposed to be copied to the newMedia dataset. It's like the dataset is hiding the rest of these folders.

 

[pdev1993]

I've got two problems I'm hoping someone on here will be able to help me with. (keep in mind, I'm pretty new to TrueNAS and what happened was out of my inexperience and stupidity). I'm kinda desperate and panicking because this affects a LOT of data I have collected over the past couple years and, while none of it is "crucial", I spent a LOT of time, money and effort getting it. Getting as much back as I can is very important to me.

Config:

OS	TrueNAS Scale Cobia, installed on 500GB SSD
Pool Configuration	1st Pool - 2x10TB drives, Mirrored, Snaphots Enabled 2nd Pool - 2x6TB drives, Mirrored, Snapshots Enabled
Memory	24GB

If you need more let me know

Background:
I was moving a large amount of media 6TB from a folder ("Media" renamed to "_Media") under dataset ("MediaA") on the first pool over to a new Dataset (also called "Media" after I renamed the folder to "_Media", we can call it "newMedia" to keep it straight here) which was a child of the MediaA dataset via the CloudCommander app (a file management docker, if you're not familiar) and had started a move process as there wasn't enough free space to do an outright full copy. I closed out of the box showing the progress, not expecting that it would end up cancelling the entire process. Afterwards, when I looked at the source folder via SMB (and in the CloudCommander app), everything was gone (cue the start of the freakout process). I was going to rollback from a Snapshot via SMB and thought I should do it from the TrueNAS GUI instead because it would be faster. When I rolled back with the Snapshot (which was from earlier this morning, before this all started, so that "_Media" folder was named "Media", matching that new dataset), I could see the Used Space for the Pool was around where it should be, however I still couldn't see the files I lost and the "newMedia" dataset was still there, with only the files that were present before the rollback. Doing some research online, I couldn't exactly find anything that would allow me to restore the files in any other way and I figured deleting the "newData" Dataset would likely destroy what little chance I have of getting that data back. I figured I'd try exporting, detaching and reimporting the pool would be worth a try (There was no visible task running regarding the Snapshot roll back)...which led to my second problem. After detaching the pool (I didn't delete any configurations, I just exported it), the GUI stopped working. It won't load, even after restarting the server. I can access the CLI and shell directly on the server, where I did try reimporting my pool but I don't think I'm doing it right, since it doesn't seem to help.

So my two problems here are:

1. The GUI won't load for use on a remote computer, the CLI/Shell does at the server itself. There are no errors displayed that I can see when the server boots.
2. I need to recover what data I can from the 1st Storage Pool. Rolling back the Snapshot seemed successful and the appropriate amount of data is being used but it's not visible in any place that I can see (even using the Linux shell to show the directory on the pool doesn't show the files I'm looking for). I can't provide any screenshots until the above problem is solved.

Please, whatever help I can get here would be greatly appreciated.

OK I resolved the first problem. Just needed to restart the middleware service. On to the second issue:
So to better visualize this, the MediaA dataset is where both the "_Media" folder and "Media" dataset are/were located. IF all of the data I'm looking for were removed, then the MediaA dataset would not be remotely close to 6.21TB of used space (as shown below).

The odd part here is that the only dataset under MediaA is "Media" (or "newMedia" as we're calling it here),the data usage of which isn't matching up to the overall size of its parent Dataset. Looking at the content through the shell, it's only showing some of the folders that were supposed to be copied to the newMedia dataset. It's like the dataset is hiding the rest of these folders.

 

[pdev1993]

OK I resolved the first problem. Just needed to restart the middleware service. On to the second issue:
So to better visualize this, the MediaA dataset is where both the "_Media" folder and "Media" dataset are/were located. IF all of the data I'm looking for were removed, then the MediaA dataset would not be remotely close to 6.21TB of used space (as shown below).
View attachment 76425 The odd part here is that the only dataset under MediaA is "Media" (or "newMedia" as we're calling it here),the data usage of which isn't matching up to the overall size of its parent Dataset. Looking at the content through the shell, it's only showing some of the folders that were supposed to be copied to the newMedia dataset. It's like the dataset is hiding the rest of these folders.
View attachment 76426

This post was meant to be a reply to my original post. This can be closed/removed.

 

[pdev1993]

This entire post can be closed. While all of this was happening, I had unintentionally mounted a Snapshot as a Dataset and not realized I could set it up as an SMB share so I could view it. It appears that all of my data that I "lost" is still there (verification pending) and I'm currently in the process of copying it over to some spare drives (admittedly probably should have done it this way in the first place). It's gonna take a while and I know I'm not out of the woods quite yet, but I feel like I can actually sleep now.