Stubbs

In safe-mode, it took a long time to boot up and become accessible, but by the time it was, it seemed fine. It's hard to tell though, because even without safe mode it can perform fine, but then randomly things will start going wrong. Attached two diagnostics, both during and after safe mode. I don't understand. What remote server? Network? I didn't have any of these problems while I was on 6.11.5 stable. I virtualize my router through a pfSense VM with a quad NIC passed through. Again, worked fine on the last update. (22-05-2023) (reboot post-safemode) tower-diagnostics-20230522-1604.zip (22-05-2023) (safe mode) tower-diagnostics-20230522-1537(non-anon).zip

I recently updated from 6.11.5 stable to 6.12 RC6, and since I did, Unraid performance has shocking. The WebUI hangs, SSH connections hang, docker containers randomly flicker between working and not responding, the "Main" page often takes forever to load all my array information. It gets especially bad when I stop the array; every single action seems to take forever while the array is stopped. To recount my actions before and during the update: I unstubbed my HBA Card. I was originally passing it through to a TrueNAS VM for a test zpool, but decided to let Unraid use it because of 6.12's zfs support. I changed the HBA card's PCIe slot from the second to the third (bottom) x16 slot. I did this because since Unraid will be using this card, I don't have to worry about IOMMU groups anymore. The bottom slot was always hard to separate from other interfaces. I installed a new Intel Optane P1600X M.2 SD in my motherboard's M.2 slot. In the system log I see a lot of this. I don't know if it's relevant, but there's a lot of it: May 21 21:49:11 Tower kernel: device veth7ada777 left promiscuous mode May 21 21:49:11 Tower kernel: docker0: port 10(veth7ada777) entered disabled state May 21 21:49:12 Tower kernel: docker0: port 10(vethb79be41) entered blocking state May 21 21:49:12 Tower kernel: docker0: port 10(vethb79be41) entered disabled state May 21 21:49:12 Tower kernel: device vethb79be41 entered promiscuous mode May 21 21:49:12 Tower kernel: docker0: port 10(vethb79be41) entered blocking state May 21 21:49:12 Tower kernel: docker0: port 10(vethb79be41) entered forwarding state May 21 21:49:12 Tower kernel: docker0: port 10(vethb79be41) entered disabled state May 21 21:49:12 Tower kernel: eth0: renamed from vethecd0195 May 21 21:49:12 Tower kernel: IPv6: ADDRCONF(NETDEV_CHANGE): vethb79be41: link becomes ready May 21 21:49:12 Tower kernel: docker0: port 10(vethb79be41) entered blocking state May 21 21:49:12 Tower kernel: docker0: port 10(vethb79be41) entered forwarding state May 21 21:50:35 Tower kernel: docker0: port 1(vethbcd1caf) entered disabled state May 21 21:50:35 Tower kernel: vetha908dc1: renamed from eth0 May 21 21:50:35 Tower kernel: docker0: port 1(vethbcd1caf) entered disabled state May 21 21:50:35 Tower kernel: device vethbcd1caf left promiscuous mode May 21 21:50:35 Tower kernel: docker0: port 1(vethbcd1caf) entered disabled state May 21 21:50:45 Tower kernel: docker0: port 10(vethb79be41) entered disabled state May 21 21:50:45 Tower kernel: vethecd0195: renamed from eth0 May 21 21:50:45 Tower kernel: docker0: port 10(vethb79be41) entered disabled state May 21 21:50:45 Tower kernel: device vethb79be41 left promiscuous mode May 21 21:50:45 Tower kernel: docker0: port 10(vethb79be41) entered disabled state May 21 21:51:44 Tower kernel: docker0: port 1(vethe1280f3) entered blocking state May 21 21:51:44 Tower kernel: docker0: port 1(vethe1280f3) entered disabled state I attached two diagnostics. The one marked "initial" was right after the update when I booted the server back up. The one marked "21-05-2023" is one I initiated just now, with the whole system running horribly. (21-05-2023)tower-diagnostics-20230521-1505.zip initial-diagnostics-tower-diagnostics-20230520-0359.zip

When writing echo med_power_with_dipm, how am I supposed to find out which disks are on which host? I know sdf and sde are connected to an ASMedia 1062 controller so I have to skip them. I don't see anything in disk information telling me what their host number is.

I am just not understanding this container. I don't know why it can't be as straight-forward as others. I have functioning domains for Jellyfin, Ombi and Nextcloud, but for some reason OnlyOffice will not work. I am still stuck on "502 Bad Gateway - nginx" when I open it in a web browser. As per the instructions in this post: I removed the container and the appdata folder. I re-installed it and added that database path (which wasn't even there before). I changed the secret key to something generated. I ran: root@UnraidServer:~# cp /mnt/user/appdata/swag/keys/cert.crt /mnt/user/appdata/onlyofficeds/Data/certs/onlyoffice.crt root@UnraidServer:~# cp /mnt/user/appdata/swag/keys/cert.key /mnt/user/appdata/onlyofficeds/Data/certs/onlyoffice.key root@UnraidServer:~# cp /mnt/user/appdata/swag/keys/letsencrypt/fullchain.pem /mnt/user/appdata/onlyofficeds/Data/certs/dhparam.pem Which copies and renames the necessary reverse proxy files for this container. I am using the documentserver subdomain config from within the Swag container. I renamed the "OnlyOfficeDocumentServer" container to "documentserver" as per the instructions in the swag subdomain template. I added a CNAME called documentserver in my registrar's DNS settings. I restarted both the documentserver and swag containers. But it still doesn't work. I keep getting the old "502 Bad Gateway".

Cryptpad does not work. Even though the docker template looks correct, it just refuses to read the config.js file I placed in the /appdata/cryptpad/data/config/config.js/ directory. Config not found, loading the example config. You can customize the configuration by copying config/config.example.js to ../config/config.js Serving content for http://localhost:3000/ via http://0.0.0.0:3000. I can guarantee I have a config.js file in there. It was copied directly from the Cryptpad github page: https://github.com/xwiki-labs/cryptpad/blob/main/config/config.example.js I wanted to change the ports and http address, but the container refuses to read it.

Isn't it very risky and unsafe to put 4x 8TB consumer drives into a raidz1? https://www.digistor.com.au/the-latest/Whether-RAID-5-is-still-safe-in-2019 Bit error rate (BER) in storage systems refers to the rate at which a block just cannot be read from the disk, due to not being able to recover data from the PRML (Partial-response maximum-likelihood) and ECC (error correct code) codes on the platter. The entire disk does not necessarily fail, but the read operation to certain block/s cannot be completed. In a RAID environment, this type of failure triggers a reconstruction of the block from the remaining disks. We can assume that B2 block written to Disk 1 cannot be retrieved correctly due to BER. This block’s data can only be reconstructed from blocks B1, B3 and Bp from other drives If URE happens during the rebuilding of a failed drive: Disk 1 failed Disk 2’s C2 block URE There is no way to recalculate both Cp and C2’s data only based on C1 and C3 blocks. The entire C stripe is gone! Then is the last question: how likely that URE would fail the RAID 5 cluster rebuilding? Based on a calculation carried out in a blog (all results are verified by myself): SATA disk (URE = 1e-14) In conclusion, for consumer faced SATA drives, the rebuilding successful rate is very low even for 4 bay SATA 5 array with 1TB disks. It’s nearly impossible to guarantee a successful rebuilding with this type of disk because of its too high URE. Another alarming fact about URE is: it has no correlation to the drive’s age.

Is it reasonable to use this plugin for disks that will not be used for Unraid? Like what if I wanted to preclear a few disks to be used for a TrueNAS VM on Unraid? There is a guide on the TrueNAS forum which I assume does something similar to this plugin: https://www.truenas.com/community/resources/hard-drive-burn-in-testing.92/ Would I be better off just followin those command line steps?

Is the information in the OP of this thread outdated? My understanding of PSUs was that having a higher wattage rating did not necessarily result in higher power draw, but the OP says this: "2. The minimum capacity that can power your build. Any more will just waste power." Isn't efficiency what matters more here? Like Bronze/Gold/Platinum/Titanium? I want to buy a new Titanium PSU for my Unraid Server, and there is a good discount for one where I live, but it's rated at 850W. My Unraid Server currently has a maximum load wattage of roughly 280 W.

I'm using Surfshark Wireguard configs and I'm getting mixed results. Resilio Sync works fine on the custom wireguard network. Firefox and Linuxserver's deluge container and YTSync all don't work. Even Resilio Sync sometimes randomly doesn't work. The Surfshark .conf file looks like this (I altered the private and public keys for this post): # # Use this configuration with WireGuard client # [Interface] Address = 10.14.0.2/16 PrivateKey = nd9Aahf9ashFSHFSe1K2rt6stcGt9rM9d7hbwml2D DNS = 162.252.172.57, 149.154.159.92 [Peer] PublicKey = Nxnc8FsnaM25UQqfDXCj0wCNUsm1RgkNHd8ndsVBja AllowedIPs = 0.0.0.0/0 Endpoint = 45.248.76.221:51820 Is there anything that looks wrong here? Another thing: these .conf files work for the Windows and iOS Wireguard clients, but not for Android.

Small suggestion: I think it would be better if the sliders for "Active" and "Autostart" were a bit bigger or easier to see. I didn't know why my docker tunnel wasn't working at first. It turns out it was because it was toggled inactive by default. I didn't notice those sliders hunched in the top right, and I'm used to them being used to toggle advanced views and the like.

Can more documentation be provided for importing Wireguard configurations? There is an option for this at the top right of the VPN Manager, but I'm not sure how to get it working for Docker Containers. I imported my commercial VPN's .conf file, the Unraid VPN manager automatically marked it as "VPN tunneled access for docker" which is what I want. Then I tried changing one of my container's network type to "Custom - wg2" (my imported .conf) but the webUI is inaccessible. Are there any further steps I should be taking?

Can someone help or provide documentation on this? Unraid provides an official guide on creating a custom tunnel linked to your home network, but I notice there is also an option to import tunnels. When I import a .conf file downloaded from my commercial VPN provider, the Unraid VPN manager automatically describes it as: "Remark: VPN tunnel for docker containers only" This is actually something I'd be interested in. I tried changing one of my container's network type to Custom wg2 (my imported .conf) but the webUI is inaccessible. Are there any further steps I should be taking?

Not sure if you saw this, but Spaceinvader left a message for you in his last video: (time-coded URL)

There is some degree of truth to it.

Don't worry, I misunderstood what you meant when you said console.

Wait what? The Unraid built-in terminal wouldn't work without a GPU? I find that hard to believe. Why couldn't it just use a virtual GPU like VNC? I mean it uses that by default for VMs.

For an Unraid Server, does it makes more sense to buy a Ryzen CPU or APU? From what I understand, the regular CPUs are a bit more powerful (mainly in terms of L3 cache) and support ECC. But I heard APUs have lower idle power consumption thanks to monolithic design. Also maybe the iGPU could be used for something? I still don't know how to use it on Unraid, or if it can be used at all. I don't run a dGPU in my server, so if Vega Graphics on a G-series APU could be utilized, I'd love to know.

Does anyone know how to get an AMD iGPU working properly in Unraid? I have a Ryzen 2400G, and I would like to use the integreated GPU for hardware-accelerated transcoding, or maybe even the GPU for a VM. I have found limited information on getting this working.

Can someone tell me how I'm supposed to disable this annoying error? Usually it's meant to be infrequent and occurs maybe once a day. Now I'm seeing it all the time, constantly. From what I understand, it's a bug with Crucial SSDs and not actually a real problem. The real problem is the stupid error messages they keep producing. For the record, I attached an extended SMART test. tower-smart-20221127-0005.zip

Can it be smaller? What if I wanted to replace a 10TB drive with an 8TB drive?

Also, does the replacement drive have to be the exact same size as the failed drive? Or can it be any size above the used space?

I made probably my biggest, most costly mistake recently. I was upgrading my case to cool my drives better, but when re-connecting the HDD power connectors, I got my modular cables mixed up (they look exactly the same) and I ended up using cables from a Thermaltake 600W Gold PSU when my server uses a Seasonic 550W Gold PSU. Initially, the server wouldn't even power on (the fans would flicker once but that's it). I eventually figured out the cables were the problem, and I've since been told what I did was extremely dangerous for my hard drives and could easily brick one. Sure enough, one of them did brick, but the rest are fine. My question is, with the presumably irregular voltage they were taking in, will that likely have any long-term effects on the drives? Are they now more likely to fail in the future or have some kind of data loss? The still seem to be functioning normally now and I haven't noticed any data loss.

I currently have 30TB of storage and it shows in my array page. But my Rootshare on Windows shows 27.2TB. Does anyone know what might be causing this? I recently removed a 3TB drive from my array using this method: Could that have something to do with it?

Is it possible for a molex connector to not deliver enough power to an HDD? I recently upgraded my server from 3TB WD Reds to 10TB Seagate Ironwolfs, and the Ironwolfs kept returning errors in my server. Sometimes I'd get the parity sync working, but it would work at about 800kbps which would've taken about a year to finish. I fixed it by taking the Seagates out of the Hotswap Bay and into the case's internal 3.5" bay, and connecting them directly to both the motherboard (SATA) and the PSU (SATA Power connector). My Hotswap Bay has two SATA power connectors, and one of them is powered by a Molex connector (because that's all I could fit). Could a Molex cause a power bottleneck for higher capacity drives?

If I wanted this container and the Rclone Nacho GUI container to use the same rclone.conf file, what should I be doing? For the Rclone-Browser container, the path looks like this: /appdata/Rclone-Browser/xdg/config/rclone/rclone.conf For the Nacho-Rclone-Native-GUI container, the path is just /appdata/rclone/rclone.conf