Greywall

Thanks @Lolight, bumped up the values by 10 degrees. Seems to have quieted down the warnings for now.

Your use cases pretty much match mine, so I think you'll be good with the build. No NVR yet tho, I have to get some cameras and get them installed first. I bought the i5 since it was on sale with the mobo as a bundle. But I chose to go with 12th gen mainly to stay with DDR4 memory. Didn't feel like the extra cost of DDR5 was worth the minor performance increase. Now if this was my gaming rig, I might have. My current gaming setup has a 12700k + DDR4 and I'm very happy with its performance.

@underdeveloped-will8428 It's been a rock solid build, pretty happy with it so far. That case is huge though, we moved and it surprised me how heavy it was once I had everything in there. Although, I'm still running it without a GPU, just been using the Intel iGPU/QSV for plex and transcoding. One minor thing I'm running into is on heavy file I/O activities like media processing, the NVME's throw too hot notifications. Hitting temps around 50c. The mobo did not have a heat sink for the 2nd NVME so I added one, but all that did was get both NVME temps to sync up. It's something I suspect I need to tune with my cooling settings. Thinking since the fan control is tied to CPU temps, I suspect fan speed isn't increasing on SSD temp spikes. In situations when you have high SSD I/O but low CPU load, I think it can lead to less than sufficient cooling. Something I need to dig into when I can shut the server down and get into the BIOS. Also, this was my first foray into the VM features of Unraid. I wasn't aware that you needed to dedicate specific cores to each VM. With this in mind, I'd have splurged for something with more cores just to not have a limitation on VMs. Although for right now, I'm only allocating 2 cores for a HomeAssistant VM so it's not really an issue. Just a thought on future extensibility on the VM side.

Thanks for the insights. I ended up going with the following: Got the ASRock for its 8 SATA ports. It would give one up if I end up adding a 3rd NVME disk on the mobo, but don't see a reason why I'd need one. Went with the i5-12600k paired with the ASRock for a combo deal. Ended up being a much better price point than the 12700k. Should be decent for transcoding with the UHD 770 iGPU. Lots of recommendation on Fractal's Meshify 2XL case. Surprised me with its height, but it's been a joy to work with. Right now have the drives from my old box mounted: 1x 6G parity, and 2x 4G drives. Will adjust drives as needs evolve. Had a good experience with Thermaltake GF A3 power supply on a recent desktop build, got the same again. Agree with not skimping on power supply, everything else benefits from solid power delivery. Held off for now on the addition of the 1070, as my CUDA use case is a year out. Will see what makes sense at that time. Besides no sense drawing the extra power if it's not going to be used just yet. Good leads on power consumption, will go through and check them out. Thanks!

Hey everyone, Moving to a new house and thinking it's about time I upgraded my old server. It's been running for the last 6 years on an older i7-3770K CPU and doing some light Plex work during this time. Looking to put together a new server with new cpu/mobo/case/ram/etc. At the new place I need the server to support: Plex + hardware iGPU transcoding Docker containers related to Plex Occasional image processing win10 VM (CPU + CUDA dependent) Home Assistant + DB (docker) NVR for IP camera recording (probably Frigate) Minecraft Server for the kiddos Hardware wise, I only want to keep my current drives. Everything else is going to be new. I have a 1070 I can add to the setup for CUDA support, but trying to nail down cpu/mobo combo to support the above. Started looking at a 12700k + Z690 combo. Something with DDR4 ram, 6 SATA onboard, NVME for cache, and a few PCIE slots. I've been buying Asus mobos for my desktop setups, but open to looking at others to get the right number of ports and slots. Appreciate a quick look to see if I'm headed down the right path. At the same time I'm wondering if I'm going to be throwing too much at a single setup. Also if I can downgrade components, it would be nice to reduce the power draw. (probably not though with the 1070?) Recommendations? Thoughts? Thanks!

Sorry, missed this question earlier. Yep, all three mount commands resulted in the same error.

Yeah, I'm kinda grasping at straws... not really sure about what's the best thing to do at this point. Been googling stuff and trying them out. Is my general thought process correct?

I have had this Unraid server running for 3-4 years without any issues. The config and hardware has been static since initially built/setup. Early this morning, ran into an "Unmountable: No file system" disk issue with disk 1 of a 2-disk cache pool... Restarted the array in maintenance mode and attempted a btrfs check, but got the following in the Web UI: bad key ordering 126 127 bad key ordering 126 127 bad key ordering 126 127 bad key ordering 126 127 ERROR: failed to read block groups: Operation not permitted ERROR: cannot open file system Opening filesystem to check... Attempted mounting the filesystem per the FAQ post but got error: root@Homeserver:/# mount -o ro,notreelog,nologreplay /dev/sdb1 /x mount: /x: wrong fs type, bad option, bad superblock on /dev/sdb1, missing codepage or helper program, or other error. Attempted doing an xfs_repair, but got issues with finding superblock: root@Homeserver:/# xfs_repair -v /dev/sdb1 Phase 1 - find and verify superblock... bad primary superblock - bad magic number !!! attempting to find secondary superblock... ...........Sorry, could not find valid secondary superblock Exiting now. root@Homeserver:/# Rebooted server and ran diagnostics (attached). The following is from syslog after reboot: Aug 30 10:20:42 Homeserver emhttpd: shcmd (40): mkdir -p /mnt/cache Aug 30 10:20:42 Homeserver emhttpd: cache uuid: b4197f0b-68d1-4496-9837-6e480fcf1bbc Aug 30 10:20:42 Homeserver emhttpd: cache TotDevices: 2 Aug 30 10:20:42 Homeserver emhttpd: cache NumDevices: 2 Aug 30 10:20:42 Homeserver emhttpd: cache NumFound: 2 Aug 30 10:20:42 Homeserver emhttpd: cache NumMissing: 0 Aug 30 10:20:42 Homeserver emhttpd: cache NumMisplaced: 0 Aug 30 10:20:42 Homeserver emhttpd: cache NumExtra: 0 Aug 30 10:20:42 Homeserver emhttpd: cache LuksState: 0 Aug 30 10:20:42 Homeserver emhttpd: shcmd (41): mount -t btrfs -o noatime,nodiratime -U b4197f0b-68d1-4496-9837-6e480fcf1bbc /mnt/cache Aug 30 10:20:42 Homeserver kernel: BTRFS info (device sdb1): disk space caching is enabled Aug 30 10:20:42 Homeserver kernel: BTRFS info (device sdb1): has skinny extents Aug 30 10:20:42 Homeserver kernel: BTRFS critical (device sdb1): corrupt leaf: root=2 block=243354976256 slot=127, bad key order, prev (576460982172839936 168 4096) current (229869420544 168 4096) ### [PREVIOUS LINE REPEATED 1 TIMES] ### Aug 30 10:20:42 Homeserver kernel: BTRFS error (device sdb1): failed to read block groups: -5 Aug 30 10:20:42 Homeserver root: mount: /mnt/cache: wrong fs type, bad option, bad superblock on /dev/sdc1, missing codepage or helper program, or other error. Aug 30 10:20:42 Homeserver emhttpd: shcmd (41): exit status: 32 Aug 30 10:20:42 Homeserver emhttpd: /mnt/cache mount error: No file system Aug 30 10:20:42 Homeserver emhttpd: shcmd (42): umount /mnt/cache Aug 30 10:20:42 Homeserver kernel: BTRFS error (device sdb1): open_ctree failed Aug 30 10:20:42 Homeserver root: umount: /mnt/cache: not mounted. Aug 30 10:20:42 Homeserver emhttpd: shcmd (42): exit status: 32 Aug 30 10:20:42 Homeserver emhttpd: shcmd (43): rmdir /mnt/cache Guessing the disk is beyond repair and I would need to something like: remove bad disk from pool attempt to wipe/format/salvage disk attempt to use what's on cache pool disk 2 to restore content to disk 1? rejoin disks into pool? Before I start going down these paths, thought it would be best to consult the sages on forums and avoid any (more) bad moves I could make. Thanks in advance! Mark. homeserver-diagnostics-20200830-1044.zip