ziogref

My unraid server was connected to a disk shelf (NETAPP DS4246) that contained all my pool devices, the shelf has suffered a failure and will take a number of weeks while I wait for new parts to be delivered. My Pool Devices (My SSD drives, 6x1tb) were not in the shelf but directly connected to the server. My 3tb SSD pool contains the appdata folder and all the docker files are on that SSD. Almost all my docker containers were only reading/writing to the SSD's (with the exception of Jellyfin and maybe some apps that end in arr). Is there are I can start most of my docker services to get things partially back online while I wait for the spare parts to arrive. As far as I can tell all the Hard Drives are healthy, so as soon as my new disk shelf arrives I would be re-adding all the array drives. I have a feeling I new to create a new config, if that's the case, do I just need to note what drives were in what slot or do I just need to note which 2 drives were my parity drives? I have my Unraid install on a Samsung Bar 128gb flash drive, I have a brand new one (unopened) would it be better copy the contents onto the 2nd flash drive as a backup before messing with my config? I assume It's just copying and pasting the files or should I use some cloning software? _______________ Part 2 of this question is restoring my server once I get my new disk shelf. I will be using the same LSI RAID card, but I will have an entirely new shelf (I opted to go new shelf for faster SAS link to the shelf). While the server was in a failing state it only marked parity 2 and disk 3 as failed (the other were spun down). When I rebooted the server all the disks were missing. Parity 2 and Disk 3 do not have the drive serial numbers written under them, all the other disks do both of these drives are Seagate EXOS 18tb SATA drives. I am a little concerned as the disk shelf lit up the drives with 2 drives with an orange light but this was prior to the reboot. (I have 2 Parity + 12 Array, all SATA, mostly Seagate EXOS) I am also worried about some custom stuff I have running, for things like IPMI control of the server fans. The fans spin at full speed (Lenovo SR650, I think 16,000rpm 2ru) and I use IPMI to spin them down to like 4000rpm. From memory this was something custom I had to download and load into a specific folder on the flash drive and I can't remember how I did it. ________________ Just looking for some advice to the steps I need to take for partial restoration and then what steps to take once I get my new disk shelf.

A little over a week ago Disk 6 (1 of 2 Seagate IronWolf Pros) dropped from the array. I re-added it and let it parity rebuild. a few days later it dropped again. So I copied the data off the drive and moved them to other disks. I then learnt about EPC and went ahead and disabled that on all 7 of my seagate drives. rebuilt the parity again and monitored if for dropping. All seemed fine so a removed the disk exclusion and all seemed well. Today Disk 5 dropped out of my array, I find it suspicious that it is the only other 16tb Iron Wolf Pro drive in my array. I also noticed disk 6 has a heap of errors marked against it, but still in the array I have the free space to move all data off those 2x 16tb drives to other disks and have started that process and I plan on replacing them if needed. I'm having a hard time trying to find out WHY they are failing. Any assistance would be great. eevee-diagnostics-20230719-2212.zip

I have docker setup using ipvlan and have a couple basic websites. I have the docker containers using real world ipv6 addresses as my ISP support full ipv6 and I give my docker containers a global ipv6 address, punch a hole in the fire wall and then anyone on ipv6 can access them. However I installed cloudflared docker container (As guided on in my cloudflare profile) to create a tunnel and cloudflare has put in either a route or an interface, I'm not 100% sure, but the ipv6 address belongs to cloudflare. So my ipv6 range that my ISP statically assigned to me starts with 2404: the range cloudflare added starts with fdd7: I restarted my server and now docker will bind to the fdd7: ipv6 address and not the 2404: address, meaning some of my containers fail to start due to them being setup in tied to the br0 interface with a static ipv6 address starting with 2404: and not having access to it. Deleting the cloudflare tunnel keeps the fdd7: ip route in place even after rebooting the server and I cannot get docker to bind to the 2404: ip range. I would like to use the cloudflare tunnel but keep docker binded to 2404: Note: It all worked before I rebooted. I had docker containers on 2404: ip ranges running across the cloudflare tunnel and it was all working. I'm still learning about docker and my only real experience is in unraid, which I know is kinda different from a "standard" docker install so if there is command I need to run in the unraid console I will need a little extra help on that. eevee-diagnostics-20230419-1125.zip