dchamb

Thanks for the response. My willingness to give up parity 2 was out of necessity, at least it was my understanding that parity 2 would be invalid if the disk position were changed. Mainly I want to know if I copy a disk from the array to UD (reiserfs drive is 14TB and current parity 2 drive is 18TB), can I add that disk (18TB) to the array while deleting the original source disk (14TB). I will have a new disk to add as the replacement parity 2 drive which will be the same size as the main parity drive.

Thoughts, anyone?

Thanks JonathanM, This is very helpful. I am still trying to decide which of the scenarios apply to my case. I am leaning toward eliminating both parity drives, but I am not sure as to the difference between "Share based no inclusions" and "Share based with inclusions". I thought I could start with moving the current parity 2 drive to UD, formatting it as XFS, use Rsync to copy the larges reiserfs data drive to the new XFS drive and wrap up with deleting the old reiserfs drive from the array and adding the new XFS drive to the array (not swapping reiserfs for XFS). Would this not work? and can I just build the parity drive from scratch?

Hi JonathanM, I've been looking but maybe I'm not looking in the right place. The instructions I find all say to shutdown the Dockers during the transfer and I really can't keep them down that long. Also looking to verify if parity rebuild will start automatically if I remove a data drive and add a new data drive to the array.

I have 3 drives that are still reiserfs and need to convert to xfs. My primary parity drive is 22TB (sdg) and second parity is 18TB (sdc). No data drive is larger than 18TB. My preference is to leave the dockers running while making the conversion. I would like for some feedback on the approach to convert the drives: 1. Take the second parity disk (18TB) offline and move to Unassigned Devices. 2. Format this drive for xfs. 3. rcopy the largest data drive (14TB) known as sdb to the newly formatted xfs drive in UD. 4. After the rcopy completes, shutdown the dockers long enough to the rcopy again to verify any changes are transferred to the new xfs drive. 5. Shutdown the array and add the new xsf drive to the array and remove the sdb drive from the array. 6. Restart the array. Parity should rebuild. (?) 7. Restart the dockers 8. Do same procedure for the remaining reiserfs drives using the drive left behind in UD as the target drive for rcopy. I am not very familiar with rcopy so if someone could provide the correct command syntax I would appreciate it. Thank you, Dale

Thanks apandey, I thought that might be the case but I was concerned about something I read regarding parity2 (implying it was position dependent). Maybe it was referring to your point that the drives are in the same slots as shown in the unraid UI. I do plan to reuse the same usb. In the new system, there will be an NVMe drive that I plan to use in addition to the 1TB Samsung existing cache drive. When I bring up the new configuration I expect the existing cache drive will show up on the UI and the new NVMe drive will show as an unassigned drive. I can add that later to my cache pool.

I am going to move my unraid server to a new Intel Core i7-12700K Alder Lake system and ASUS X690-R Prime motherboard. I have 2 parity drives, 4 array drives (3 reiserfs/1 xfs), and a 1TB Samsung SSD (cache). Anything I need to do to move the drives to the new hardware? Do I need to keep the cable and drive assignments the same? Will it affect the second parity drive?

Can an array support a mix of reiserfs and xfs drives?

Okay. I just updated to 6.11.3 and so far everything looks good!

I am currently running 6.10.3 with no problems. What would I gain/risk by going to 6.11.3? I am about to install a 22TB WD Gold drive to replace my 18TB parity drive.

Is the HDMI dummy plug necessary if you have a physical monitor plugged into the port through a KVM switch?

I'm just avoiding HW transcoding altogether

Thanks! This helps with clearing up the confusion. And the Plex problem is in Plex's arena to fix, right? Or are we waiting for a kernel fix for unRaid or both?

I understand it has discussed frequently and it also appears there is discussion to the contrary. I don't want to tear this system apart only to have regrets.

It's getting really confusing. I have an Alder Lake system with an Asus 690 m/b ready to go for transitioning from my very old Ivy Lake system. I run primarily Plex but without HW transcoding. Would you make the switch if you were me?

I did upgrade my old system to 6.10.3 and everything has been fine for over 11 days.

Software or hardware transcoding?

Ok help me understand how the GPU is enabled without actually hardware transcoding. Is there a way to have unRaid just totally ignore the GPU?

I am getting ready to do the same thing. My understanding is that iGPU transcoding will still be a problem until unRaid 6.11. So I am steering clear of HW transcoding until then. Others have advised me not to spend the money on a NVIDIA card since 6.11 is near. Although I'm not sure what "near" means.

I decided to pull the trigger on an i7-12700K and ASUS Z690-P Prime motherboard. Trying to decide if I should upgrade my current system to 6.10.3 from 6.8.3 before moving the drives over to the new configuration, or start the new system with 6.8.3 and upgrade there. BTW, I did locate one stick of memory on my current system that shows errors on memtest. I replaced the memory and everything so far seems solid.

So does that mean if I go with Alder Lake I won't be able to use Plex, or just Plex with HW transcode?

So no point in using hardware transcoding with the 12600K since it is powerful enough?

No I am not on Alder Lake. I am looking to refresh my tech since it is 10 years old now (on an i5-3570 CPU). I tried upgrading to 6.9.2 several months ago without hardware transcoding turned on and my system crashed within 24 hours. Tried several times but could not get it to stay running for more than a week. I rolled back to 6.8.3 and it was running fine until today. Now I am thinking I may need to accelerate plans for the hardware upgrade. My latest post on the current situation:

Today I noticed that I wasn't able to access shares from Windows, Plex could not access the server and the webgui was showing zero processor activity. Yet I could move from tab to tab on the webgui. I am running 6.8.3 and it has been rock solid until now. I have made no hardware changes. I had to force shutdown since the shutdown command was unresponsive. Naturally when it booted back up it was running a parity check. For a short time I was able to access the shares via Windows, but then after a couple of hours, no access again. Webgui shows CPU being used. Fortunately I was able to get a diagnostic dump. Can someone look at this at tell me what could be going wrong? Thanks Dale tower-diagnostics-20220629-1250.zip

Because the iGPU issue will get resolved? Seems like this problem has been around for a year now and I haven't seen any problems reported on NVIDIA.