Help with VMWare Datastore issue

[Lorddalron]

Hi mate, I've been working with ESXi for years so hopefully can help you out a bit here. From what I can gather you have a server with 2 hard disks installed. These hard disks each have one virtual disk stored on them. These virtual disks can be set to a larger size than the underlying physical disk if they are set to thin provisioned. this allows for overallocating the storage but can cause problems if the actual data size gets near to the underlying physical disk size as this is a hard limit. Windows (or whatever OS you have installed) cannot see the physical disk and believes that it has however much space you have said is the size of the virtual disk which appears to be what has happened here. If the virtual disk files still exist then potentially you can get the data out of them. First off enable TSM-SSH on the ESXi host and then open an SSH session to your ESXi host. Once connected run ls /vmfs/volumes which will show you the volumes you have on your machine. there should be 2 volumes that match your datastore names (from what I can tell these are Please Media Storage and datastore1). Can yoiu please run ls -l on each of these (ls -l /vmfs/volumes/datastore1 and ls -l "/vmfs/volumes/Please Media Storage") and post the results here then I can get an idea as what you have currently. What may be the safest thing would be to manually create the Plex Media volume using command line and get it to fiull the physical disk, this way when it is running out of space this will be visibil in the guest OS as well. Note: as this is a Linux based OS everything is case sensitive so bear that in mind.

Cheers

Hi Chris, thanks for replying.

So, a couple of updates since my last message. Unfortunately the 4TB has since failed. I dont think this was the problem all along, just sheer bad luck.

Whilst I still have it up and running on the main 500GB drive, im planning on changing things up during/after Christmas as 500GB simply isn't much to play with for what I want.

As easy as it would be to simple populate a NAS with a couple of large drives, it would be a shame todisgard the Proliant given that its in good working order.

Therefore, couple of plans im thinking of:

Plan A:

Clone the existing 500GB boot drive on to a new 500GB SSD

Purchase 2 x 4TB SSD and go RAID0 again on the basis that SSD should provide a little bit more reliability. In time, I could then purchase a 3rd and 4th.


Plan A:

Clone the existing 500GB boot drive on to a new 500GB SSD

Purchase 1 x 4TB SSD and and then a single 4tb external hard drive to use as a back up solution.

Plan C:

Clone the existing 500GB boot drive on to a new 500GB SSD

Purchase 2 x 8TB SATA and implement RAID1 so that if a disk was to fail, I wouldn't have to start all over again. No backup


Plan C:

Clone the existing 500GB boot drive on to a new 500GB SSD

Purchase 1 x 8TB SATA and 1 x 8TB external hard drive. Create a single W10 machine with 8TB capacity and then configure the backup to go to the 8TB external drive. In the event the internal 8TB was to fail, I can utilise VMware to create a new vm and then migrate my media back over

All the while im typing this im not really thinking of what could happen the main 500GB SSD with VMware, therefore, im all ears as to additional "plans".

Basically I want maximum storage for as little cost as possible. The only reason im thinking of SSD was that they might or should be less prone to failure. Not sure if they would add to Plex performance, but happy to be corrected on that front.

 

[Armageus]

Plan Z

Forget about any kind of raid, clone your SSD to a new one, and buy 2 new hard drives.
Use one of them, and get a USB caddy for the other one.
Plug it in once a month or whatever and backup to it
(Also backup the VMware virtual machine if it's that hard to recreate)

 

[nade]

Sorry - for some reason had missed the earlier reply you updated.

As above, I would be concerned about using any RAID0 setup. If they are new drives, yes it might be reliable if they are new. But at the same time I think that is striping across both disks, meaning if either fails then you'll lose your data (as far as i'm aware, not done any hands on for a while)

From your latest update... if you are wanting the max amount of storage for cheapest cost, then just go with a single large disk and set it up link you have listed on PlanC.

But also as suggested from Armageus, need to consider backups if then relying on a single drive.