I just purchased an Asustor FS12 hoping to use it just for VMs via iSCSI. A little about my setup. Upgraded the internal 4GB memory to Crucial 16GB, recommended and confirmed working in a thread I read. I purchased two TeamGroup 2TB MP44L drives and configured as RAID1. Using the 10GB port connected to a Ubiqiti Unifi 10GB aggregation switch via GTEK 10GB Copper SFP and Cat8 braided cable. Also using MTU of 9000 for jumbo frames.
I got really excited and copied over all my VMs from two Synology NASs. A couple of days later I started to see issues with the VMs, corrupt file behavior, apps not running, etc. I had to restore all my VMs from backup so I chose to put them back on the Synologys.
At this point I decided I would spend time troubleshooting the FS12 before using it in production. For testing I took a known good working VM and cloned it to the FS12. To confirm I ran sfc /scannow against it and verified no corruption as well as checking the Windows system logs as this is where I was seeing random reported disk errrors -
The IO operation at logical block address was retried
An error was detected on device during a paging operation
I opened a ticket with Asus and the first response was I needed to create my VMs fresh on the Asustor, that migrating or cloning them from another environment may cause issues. Not very helpful. In my 25+ years as an IT professional, storage is storage. If is supports iSCSI and LUNs as an industry standard, the storage should just work.
Asus support also said the drives I was using were not listed on the approved drive list. Where is the approved drive list specific to the FS12? All I can find on their website is a generic list of drives that have been confirmed working in various NAS models.
To further test I added two additional drives, a Samsung SSD 960 EVO 250GB and an Intel 660p 512GB SSD. All four drives were reconfigured as single drives with no RAID, each drive configured with single LUN, volume, and iSCSI target. I cloned a known good VM to all four drives, ran sysprep, and tested with sfc and for drive errors. All tested clean initially. Over time every drive would eventually get disk errors and sfc would fail, usually overnight. Times of the disk errors are random across every drive and VM.
Obviously, I really want this device to work so I replicated the above test between each of the following steps, all with the same results.
Removing jumbo frames, going back to MTU of 1500
Changing the network cable and port, replacing with a Cat6 and running across a 1GB connection
Changing from ext4 to btfs
Removing the 16GB Crucial memory and replacing with the OEM 4GB module
Doesn’t matter what I do, I get the same results. The VMs over time show disk errors and os corruption, eventually becoming unusable. At this point, I have to think this is related to either a bus issue, or the ADM, and I don’t think this is a hardware issue per se, but it could be.
For these reasons, if you plan on using the Asustor FS6 or FS12 specifically for VMs, I can’t recommend it. If using it for any other purpose, it may be fine., but I only need it for one thing. If you can’t trust the drives and their connectivity then you shouldn’t store data on it.
I’m open to any recommendations on further testing procedures. I am still working with Asus support. I may try and load TrueNAS Scale on it as I’ve read it works. However, my return window is getting shorter. I have about a week left to experiment, then back it goes.
Update: I created a new (not cloned) VM on the FS12 and I’m still seeing the same issues.