Can you also confirm that snapshots are not available on XFS? is that a btrfs feature only?
Unless you use lvm with ext4 or xfs, btrfs or zfs are your only options with snapshots.
OMV on ReadyNas Pro 6
-
-
Is that Logical Volume Management?
Yes
I assume it's something not in the UI and also which would be needed before the volume is created?
You are really worried about things not being in the UI. I won't suggest anything outside of the UI without explicitly telling you it is outside the UI. That said, there is an LVM plugin. So, you could do it from the web interface but it is nowhere near as feature complete as the btrfs snapshot stuff.
-
-
Thank you.
Just for the records, I can deal with a CLI. I just don't want to build a system which relies heavily on CLI - now and in the future when updates come out. My usage of the NAS is simple and as such I'd like to keep it simple. For other tasks, by all means I can spend days on a CLI
I was really not using snapshots on my ReadyNAS - I discovered them fairly recently. I'll take a look at the plugin. If it does the basics, it might work.
I am 50% into restoring my data on the NAS - would LVM require me to re-create the volume?
And I appreciate you filtering the info to suit my needs!
-
would LVM require me to re-create the volume?
Yes. You would create volume groups/logical volumes/physical volumes on top of mdadm raid and the ext4/xfs/whatever filesystems would go on top of the logical volumes.
-
Hijacking this old thread to ask if latest version of OMV currently 7.4.x is actually stable on the ReadyNAS Pro 6 (BE/Pioneer) as I have a Pioneer unit with the dreaded Marvell Yukon 88E8053 network card which dies out after a while due to some bug in the sky2 driver that got reintroduced on the more recent Linux kernels.. (as well as the msk driver that also brought the same bug back in the more recent FreeBSD releases :/)
-
-
I don't have one but if the bug is in the 6.1 or 6.11 kernels that OMV 7 uses on x86 systems, then you probably won't have much luck.
-
-
I just wanna add, that I have finally figured out the winning combination. You have to use the Proxmox kernel in order to get rid of that dreaded Marvell sky2 bug!
With the regular Debian 6.1 kernel that comes with OMV, the issue is still there and the sky2 driver crashes after a while. However, as soon as I installed Proxmox 6.11 kernel, the problem went away !
-
-
oh that sucks - you're saying I cannot upgrade my Pro6 to OMV 7??
You can upgrade to OMV7 no problem, but after you upgrade, make sure to install the extras if not already done and then install the Proxmox kernel and make it the boot defauly. I can 100% vouch for this one in particular:
CodeLinux version 6.11.11-1-pve (build@proxmox) (gcc (Debian 12.2.0-14) 12.2.0, GNU ld (GNU Binutils for Debian) 2.40) #1 SMP PREEMPT_DYNAMIC PMX 6.11.11-1 (2025-01-17T15:44Z) ()I also wanna thank you for pointing me in this direction as I had pretty much almost given up on ever using this ReadyNAS 6 Pioneer unit in any capacity due to the damn Marvell Yukon-2 EC network card that has this bug in the sky2 linux driver as well as in the msk(4) FreeBSD driver as well. Everything I tried kept crashing eventually : TrueNAS SCALE and XigmaNAS included. I will try the beta of TrueNAS SCALE today or tomorrow as it is based on Linux kernel 6.12 just to see if by chance it is ok but at least we have a 100% rock stable version of OMV that we can use with the Proxmox kernel that I mentioned above
Cheers!
-
No, thank you for the extensive reporting - I found your posts online. I wouldn't have been able to do that. I didn't know what "tainted kernel" meant!
The thing is that I am not a Linux person and I'd hate to have to fiddle with Linux more than necessary - the idea of having to replace the kernel doesn't appeal me, not to mention I have no idea how to do that.
My NAS has 12TB of storage, I don't feel like doing too many tests - yes, I could just replace the drives and test with empty ones but when I had those issues it would take hours/days/weeks/month for the NAS to disappear from the network. You've probably saved me a huge amount of time.
Silly question: what is the Proxmox kernel?
Please do let me know how 6.12 goes - and let me know if I can report the issue somewhere so to double the number of reports!
-
-
So in OMV7, if you install the OMV-EXTRAS, you will get an option in the GUI to install additional kernels. So you can just install the Proxmox kernel and that's it. Config remains the same. The only difference is that it will be stable
-
interesting. we need to find out what's different in the Proxmox then
It would be a shame if the bug remained in all future versions... -
Did some further stress tests and I installed latest beta of TrueNAS Scale 25 which comes with kernel 6.12.
Turns out the rx error still pops up under extremely high load even with the latest kernels 6.11 and 6.12.(Torrentting constantly at 500+ Mbps with) and pushing the CPU pretty high. I was seeing it as well with Proxmox kernel 6.11 as well but again very very few such errors compared with kernel 6.1 where once the errors would start, it would result in a crash of the sky2 driver and the system would become inaccessible until rebooted.
So for me, running OMV with kernel 6.11 or TrueNAS SCALE 25 beta with the 6.12 kernel, has reduced the error where i highly doubt I will actually see the box become inaccessible on a regular basis like it used to. I have also setup the 2 network ports on the ReadyNAS in a LAG just in case one of the 2 ports does eventually lead to the sky2 driver crashing, the other one will continue to send/receive data albeit with heavy packet loss while I go about to reboot it.
The last test that I want to perform is to install TrueNAS CORE latest version which at this point is pretty much abandoned but since it is based on FreeBSD 13.3, and its msk(4) driver supposedly includes a built in function to deal with RX FIFO underrrun errors, should hopefully be more stable..
I'll update the thread once I have it up and running and under heavy stress tests..
-
-
Thanks for the extra tests.
I had a look again at TrueNAS but I keep finding mentions that it won't be so happy with only 8GB of RAM
It's good news that 6.11 and 6.12 seem to mitigate the issue somehow.
Can I ask how to LAG the two interfaces?
Looking forward to further feedback, thanks for taking the time to report here!
-
Haven't had the time yet to test TrueNAS CORE as honestly I think I'll stick with the TrueNAS SCALE 25 for now as kernel 6.12 is very very stable and I have yet to have any issues with it. Just waiting to increase the RAM to 8GB but I find it hard to find it on ebay for a good price.
It works fine with TrueNAS with 4GB no problem as long as you don't enable ZFS deduplication. I've had zero issues when it comes to RAM usage so far so with 8GB RAM you will be more than fine with it.
In order to bond the 2 interfaces together into a LAG, you first need to have a managed switch and configure 2 ports on that switch to a LAG first. (Link Aggregation aka 802.11ad) once you do that, you go to the TrueNAS config or OMV config for the network and create a Link Agregation or a Bond from your 2 existing network interfaces. Once you apply and save the settings, you plug in the 2 ports into the ports that you configured on your switch and reboot the NAS. This is much easier to do with the VGA coable connected to the box in case you need to troubleshoot it or to revert the changes..
Cheers!
-
-
I'll update the thread once I have it up and running and under heavy stress tests..
Hi adisor19
I was wondering if you could update on your tests? One of my NAS is still running on an old OMV and it might be time to update. But it's the one with the problematic NIC!
Thanks!
-
Wondering if you figured out fan control on the ReadyNAS?
Chris
Participate now!
Don’t have an account yet? Register yourself now and be a part of our community!
