ChatNoir

If you really need more PCIE lanes (several discrete GPUs, HBA, etc), consumer grade plateforms will be a limiting factor since they generally only provide 20 PCIE lanes. "Enthusiast", Workstation or Server grade platforms offer more on that front but at a generally higher price. On the other hand, you are already considering pretty expensive MBs ...

It might be that, did you try to power it through a Molex 4pin + Sata adapter ?

Yes, you can use a random usb key as an array drive in your situation.

I found this on the appstore, looks like a lancache: Don't know if it would do what you want, but you can ask there.

The mover cannot touch used files. Did you deactivate the Docker service before running Mover ?

As for the complexity of the FAQ entry, you should be able to ignore everything after the image if you are on 6.8.3 or above. It is not really difficult: Stop the array add the second disk to the pool start the array Unraid will automatically balance for RAID1, wait for the balance to complete balance to RAID0 if you don't mind the lack of redundancy. It is easy, just select RAID0 from the drop down list and hit Balance. wait for the balance to complete You have a 2 devices RAID0 cache.

Hello, you should attach your diagnostics in your next post, there are much more information than an extract of the syslog. (Tools / Diagnostics) While it may not be the cause for your disks issues, I see that you are on a Ryzen platform so you should look at this: You are running your memory out of the CPU specs and C-states could causes issues too.

Hello, you should attach your diagnostics in your next post, there are much more information than an extract of the syslog. (Tools / Diagnostics) I see that you are on a Ryzen platform so you should look at this:

You should attach your diagnostics in your next post (Tools/Diagnostics).

Anyhow, the regular blkdiscard seems to have removed the partitions on the drives. The RAID10 balance / conversion is finished. Thanks @JorgeB !

I only have those options available: root@mediahub:~# blkdiscard --help Usage: blkdiscard [options] <device> Discard the content of sectors on a device. Options: -o, --offset <num> offset in bytes to discard from -l, --length <num> length of bytes to discard from the offset -p, --step <num> size of the discard iterations within the offset -s, --secure perform secure discard -z, --zeroout zero-fill rather than discard -v, --verbose print aligned length and offset -h, --help display this help -V, --version display version For more details see blkdiscard(8).

Still on beta25. So not sure if it is necessary but I wanted to use it just in case. The -f option is not recognized (or I am doing something wrong). root@mediahub:~# blkdiscard -f /dev/sdf blkdiscard: unrecognized option '--f' Try 'blkdiscard --help' for more information.

I am not great at linux commands, is it simply a blkdiscard sdX or should I use options ?

Emulated drive means that the data is only available by computation of data from all other drives and parity. It also means that any other issue on another drive might cause data loss. You should fix this quickly. You should attach your diagnostics (Tools / Diagnostics) in you next post.

Back on the topic. I am moving all my files from local_data to cache with Midnight Commander. I would like to be sure of the next steps. I thinking: scrub cache to start with a clean slate stop array remove drives from local_data start array stop array assign freed drives to cache start array wait for the end of automatic raid1 balance convert/balance for raid 10 Are there mistakes, missing or useless steps ?

It is in the base kernel, but 6.8.3 was built on a pretty old one. The issues have been solved and 6.9.0 is up to date.

Welcome, I think it is a good idea to stay around, see what people are doing, what problem they are having so you can start on the good track and be able to make the good decisions or at least know what to ask for.

Glad it worked. I have a screen shot of the main page with all the drive visible stored on my Onedrive. That way it is always available. But you can also just put a sticker on the physical drives, etc.

De quelle type de modification tu parles ? Qu'est-ce qui est réinitialisé ? (je ne connais pas Jeedom)

I would wait for a confirmation from someone with more knowledge than I have. But I think the solution is to put all drives as data and the drive with no file system is your parity drive. I would only work with single parity.

using Community Application, I do find the Chinese language (using beta25)

Salut, Peut-être un soucis dans tes chemins sur la config du docker ?

You would need to install the Community Applications plugin : Then go to Apps and install the Chinese language pack.

I am no expert, but I think you are mixing things. The system share is where the file docker.img is stored. In there is store the generic images pulled from the internet. The path you mention looks OK for me. If you want to have it on your cache drive, you have to make sure that the system share is indeed on your cache (Prefer would be good). The appdata share is where your dockers are writing the files specific to your system. They are generally best on an SSD too, so the setting you have currently (Prefer) looks OK.

Ok, you are right. The visualization section was open and hiding all the good stuff below ... Pretty dumb from me.