trurl

No Have you used up your Trial extensions?

Get an UPS

Are you sure you're looking at the Dashboard page? That disk should be showing a thumbs down. On mobile now so will look at Diagnostics later.

Since you are reusing old disks you should post your Diagnostics so we can check out the rest of your setup. Go to Tools-diagnostics and attach the complete Diagnostics zip file to your NEXT post. The disk in question should be showing SMART warnings on the Dashboard. Do any of your other disks show warnings? You should setup Notifications to alert you immediately by email or other agent as soon as a problem is detected.

You can go directly to the correct support thread for any of your plugins by clicking on its Support Link on the plugins page.

Does it show up in the BIOS? Go to Tools - Diagnostics and attach the complete diagnostics zip file to your NEXT post.

You can go directly to the correct support thread for any of your dockers by simply clicking on its icon and selecting Support.

I see you attached it to the original post. The forum doesn't show us there is any new content when you do that. That is why I usually say: But, since you "thanked" me, I got a notification about this thread anyway and that is why I came back to it. Now that I have posted in it, others will also see there is new content in the thread. Any particular reason you chose 19G for your docker image size, instead of a more round number? I usually recommend 20G, but 19 is OK and it doesn't look like you are in danger of filling it. Just wondered if there was a reason for that choice. But docker image may indeed be the problem, see below. Your appdata and system shares are setup wrong. You want appdata, domains, and system on cache and set to stay on cache, so your dockers and VMs performance won't be impacted by the slower parity writes, and so those won't keep array disks spinning. Your syslog has a lot of stuff about trying to move appdata from cache, which you shouldn't be doing anyway, and it is failing due to duplicates already on the array. Not sure how you got duplicates. That will all have to be cleaned up. There is a dump in syslog which seems to be related to docker image. Probably need to delete and recreate it. Then you can add your dockers back just as they were using the Previous Apps feature on the Apps page. But it might be better to just delete docker image and don't enable docker again until you get those shares fixed.

Can't say what was happening since syslog is only since reboot. Those speeds sound like you may have had connection problems. It starts with the longer outer tracks, so more data per revolution, and then continues to the shorter inner tracks, so less data per revolution. Completely normal for the outer tracks to be faster data rate than the inner tracks. Also, those smaller disks will be slower than larger disks due to lower data density. Once it gets past the 2TB mark I would expect it to speed up again as it finishes checking the rest of the 8TB parity, then get slower as it goes. All about the data density.

On the other hand, the only consequence of not keeping files together is a possible short delay while another disk spins up. So I just let it split them.

Unraid will not move existing files except in accordance with the Use cache setting (between cache and array). And Split Level has precedence over Allocation Method or Minimum Free. Also, Unraid has no way of knowing how large a file will become when it chooses a disk to write it to. So, it will choose the disk that Split Level says it should choose. Long story short, Split Level CAN make you run out of space on a disk.

Parity2 and parity1 are not interchangeable. Either provides protection for a single drive failure. Together they provide protection for 2 simultaneous failures. You can have parity2 without parity1, but you can't put parity2 in as parity1 unless you rebuild it as parity1.

You shouldn't access your server remotely except through VPN. Your server shouldn't have a public ip or ports opened to it. Probably you are being hacked already.

How did you determine the drive was bad? Many things can cause a disk to become disabled and bad disks aren't the most likely reason.

Go to Tools - Diagnostics and attach the complete diagnostics zip file to your NEXT post.

Do you have another flash drive you can try? Boot from a USB2 port if possible.

memtest must run as the only thing, no OS or anything else running. Not sure who you think has made an oversight here. This is the way memtest always works on everything, and Limetech didn't write memtest.

Usually growing Docker image is due to an application writing to a path that isn't mapped. Very rare to need more than 20G and never much more. If it is growing something is set up wrong.

Not related, but why have you allocated 100G to docker image? Have you had problems filling it? Maybe just a coincidence, since your diagnostics includes the results of ifconfig and ethtool and they look OK. Have you done memtest?

New Config will rebuild parity unless you tell it not to. And you MUST rebuild parity in this case.

Save us the trouble of asking and save yourself the waiting

Go to Tools-diagnostics and attach the complete Diagnostics zip file to your NEXT post.

https://forums.unraid.net/bug-reports/stable-releases/report-guidelines-r68/

Not sure 6.5.6 is available anymore. In any case, the version you have is so old that you should probably just work through the upgrading wiki link you quoted, and also see this additional one: https://forums.unraid.net/topic/66327-unraid-os-version-650-stable-release-update-notes/