trurl

Do you mean from outside your LAN? Shouldn't do that without VPN. These sound like a problem with your server being able to access the internet. No matter which version you are running if your server can't access the internet, and get reliable DNS lookups, these things will not work.

Shouldn't be necessary to move them manually if you have the "Use cache disk" setting correct. See this FAQ for more detailed explanation of that setting: https://forums.unraid.net/topic/46802-faq-for-unraid-v6/?do=findComment&comment=537383 As you can see, cache-yes is the only setting that will move files from cache to array. After mover has finished post new diagnostics.

To answer the question in the title, you need the config folder. But it's not clear whether that would help with the problem. Go to Tools-diagnostics and attach the complete Diagnostics ZIP file to your NEXT post.

Unraid monitors these things for you but you should set up Notifications to alert you immediately by email or other agent so you will know without even opening the webUI.

Post new diagnostics. Looks like your cache was completely full. You have a cache-prefer share anonymized as S---f , what is that for?

Your docker image is corrupt. Why have you allocated 40G? Have you had problems filling it? This is usually due to misconfigured docker application(s). 20G should be more than enough and making it larger won't fix anything, it will just make it take longer to fill.

Each parity disk must be at least as large as the largest single data disk. Since disk3 is 10TB parity2 must be at least that large.

Go to Settings - Global Share Settings and make sure all disks are included in User Shares.

Since OP never answered the question Maybe you will.

You can get to the console inside the container by clicking on its icon and selecting Console if you want to try working at the command line inside the container. But the docker image doesn't have anything that can't be downloaded by installing again. The things you don't want to lose aren't in the docker image. The template with all your mappings etc is on flash and Previous Apps on the Apps page will use the template to reinstall exactly as it was. And anything the application itself needs to remember is in its appdata. We often recommend just blowing away docker image when people have filled or corrupted it since it is easily recreated and contains nothing that won't be downloaded during install of your containers

I blame aggressive salesmanship by search engines. While looking for that OCULink to 4 x SATA cable I was given all sorts of results to places that didn't have it even when I searched on the exact part number. No real problem and you've been more than helpful.

I pulled the trigger before going to bed last night. Amazon had not shipped so I cancelled the RAM. So that blows all my "savings". I ordered the slower (cheaper) Timetec.

You can just use the user share settings to control whether the share gets moved from cache. The default for dockers and VMs has those staying on cache.

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

The last contents of your syslog in those diagnostics was this: Jun 2 07:13:19 Tower rpcbind[4721]: connect from 192.168.1.68 to getport/addr(mountd) Jun 2 07:13:19 Tower rpcbind[4726]: connect from 192.168.1.68 to getport/addr(mountd) Jun 2 07:13:19 Tower rpcbind[4731]: connect from 192.168.1.68 to getport/addr(mountd) So that might correspond to when the last time the server was up.

I had a backup/test server in a Node 304 that I got the mobo out of for my current temporary setup. I had no end of connection trouble with that one every time I ran a parity check. I don't know if it was those suspended disks, or maybe the locking cables weren't playing nice with my WD disks, or both. I may try to get it going again after I get my main server rebuilt.

Exactly the sort of feedback I was hoping for, thanks. I almost asked you why you weren't using registered ECC. And Amazon lets me save even more money on those Crucial sticks. Crucial 32GB Kit (16GBx2) DDR4 2400 MT/s (PC4-19200)

Still these filling your logs. Does the traffic go away if you make this stop? Does the traffic go away if you stop plex?

Many people find booting from a USB2 port to be more reliable. USB3 speed isn't needed since the flash drive is mostly used for loading the OS into RAM at boot and it runs in RAM (think of it as firmware). Any configuration changes you make in the webUI are also recorded on flash so the can be reapplied at boot, but flash isn't used much and doesn't store much.

Do you use plex DVR?

And it looks like I can step down to the E-2146G without losing much and so impress the wife with how much money I'm saving.😉 Looks like Amazon has everything as cheaply as anyone, except those CPUs. They are getting them from Provantage and marking them up a bit. And they don't have the OCULink cable. Working strictly from the C246 WSI CPU and Memory lists. Here is what I have so far and I think it is all I need since I have the case and PSU already: AsRock Rack C246 WSI Mini-ITX Server Motherboard Intel XEON E-2146G Processor Noctua NH-L9x65 2 x Hynix 16GB DDR4 PC4-19200 2400MHZ ECC REG DIMM Supermicro 50cm OCuLink To 4 SATA Cable

Cache-prefer means "prefer to keep the files on cache and if any overflow to the array, move them to cache when there is room". Go to Tools - Diagnostics and attach the complete Diagnostics ZIP file to your NEXT post.

Thanks for all this info and the contact. Some of these IPMI considerations are making me think I might be more comfortable with the WSI now though. And it would save me a little money. I just remembered that NH-L9I died some time ago and I went back to the stock cooler. So I will get that NH-L9x65, certainly looks like it would do more cooling.

1. Parity must be at least as large as the largest single data disk, so in your example of 6x8TB data disks, a single 8TB parity disk provides redundancy to rebuild any of those data disks. But note that parity isn't a backup in any way. Parity in whatever system it is used is just an extra bit that allows a missing bit to be calculated from all the other bits. A parity disk allows a single disk to be recovered from the parity calculation by reading parity plus all the other disks. Lots of ways to lose data besides a single disk failure, including simple user error. Parity cannot help if you accidentally delete a file for example. You must always have another copy of anything important and irreplaceable. 2. Since each disk in Unraid is an independent filesystem (no striping), each file exists completely on a single disk. Read speeds are at the speed of the disk being read. Write speeds are somewhat slower due to realtime parity updates. If you are copying a very large number of small files all systems will be slower simply due to filesystem overhead. 3. I don't see any dockers listed for Roon Server, but I assume you could run it in a VM.