TheFreemancer

Yes I have tower-diagnostics-20220915-1009.zip

tower-smart-20220915-1037.zip tower-diagnostics-20220915-1039.zip I've rebooted the server and the drive wasn't showing up. I replaced the SATA cable and it showed up again but good 15 UDMC errors, I remember this drive had UDMC errors before. The drive is showing up again but still disabled. There's an option I never seen before "Read-check", I'm starting it as I don't think it will hurt, if I shouldn't do it I'll stop it. EDIT: I remember something similar when a drive got disabled and the solution was to rebuild on top of that particular drive. But this is dangerous because if it starts to rebuild and the drive actually fails, all data is gone if I remember correctly. Tbh, I don't remember how to do this and I'm not sure if this is the right thing to do.

This happened to me today, also running 6.10.3. I'll reboot my server and see if it come back online again.

Just now I received a pushbullet notification for an error with a particular drive. I haven't done anything, didn't reboot nothing as I'm coming here to ask what should I do and since some logs disappear after I reboot I've done nothing so far. Here's the log I got: Sep 1 15:05:44 Tower kernel: sd 5:0:0:0: [sdf] 15628053168 512-byte logical blocks: (8.00 TB/7.28 TiB) Sep 1 15:05:44 Tower kernel: sd 5:0:0:0: [sdf] 4096-byte physical blocks Sep 1 15:05:44 Tower kernel: sd 5:0:0:0: [sdf] Write Protect is off Sep 1 15:05:44 Tower kernel: sd 5:0:0:0: [sdf] Mode Sense: 00 3a 00 00 Sep 1 15:05:44 Tower kernel: sd 5:0:0:0: [sdf] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA Sep 1 15:05:44 Tower kernel: sdf: sdf1 Sep 1 15:05:44 Tower kernel: sd 5:0:0:0: [sdf] Attached SCSI disk Sep 1 15:06:15 Tower emhttpd: ST8000DM004-2CX188_ZCT128DJ (sdf) 512 15628053168 Sep 1 15:06:15 Tower kernel: mdcmd (5): import 4 sdf 64 7814026532 0 ST8000DM004-2CX188_ZCT128DJ Sep 1 15:06:15 Tower kernel: md: import disk4: (sdf) ST8000DM004-2CX188_ZCT128DJ size: 7814026532 Sep 1 15:06:15 Tower emhttpd: read SMART /dev/sdf Sep 1 15:10:02 Tower emhttpd: shcmd (259): echo 128 > /sys/block/sdf/queue/nr_requests Sep 4 13:18:47 Tower emhttpd: shcmd (8734): echo 128 > /sys/block/sdf/queue/nr_requests Sep 4 13:19:18 Tower emhttpd: shcmd (8772): echo 128 > /sys/block/sdf/queue/nr_requests Sep 4 13:19:24 Tower emhttpd: shcmd (8789): echo 128 > /sys/block/sdf/queue/nr_requests Sep 4 13:19:28 Tower emhttpd: shcmd (8810): echo 128 > /sys/block/sdf/queue/nr_requests Sep 4 13:19:36 Tower emhttpd: shcmd (8840): echo 128 > /sys/block/sdf/queue/nr_requests Sep 4 13:19:36 Tower emhttpd: spinning down /dev/sdf Sep 4 13:19:42 Tower emhttpd: shcmd (8876): echo 128 > /sys/block/sdf/queue/nr_requests Sep 4 13:19:51 Tower emhttpd: shcmd (8924): echo 128 > /sys/block/sdf/queue/nr_requests Sep 7 09:28:49 Tower emhttpd: read SMART /dev/sdf Sep 7 18:28:50 Tower emhttpd: spinning down /dev/sdf Sep 10 09:57:19 Tower emhttpd: read SMART /dev/sdf Sep 10 20:47:15 Tower emhttpd: spinning down /dev/sdf Sep 12 10:44:08 Tower emhttpd: read SMART /dev/sdf Sep 13 03:43:08 Tower emhttpd: spinning down /dev/sdf Sep 13 11:16:35 Tower emhttpd: read SMART /dev/sdf Sep 13 20:17:37 Tower emhttpd: spinning down /dev/sdf Sep 14 11:15:44 Tower emhttpd: read SMART /dev/sdf Sep 14 19:18:15 Tower kernel: sd 5:0:0:0: [sdf] Synchronizing SCSI cache Sep 14 19:18:49 Tower kernel: sd 5:0:0:0: [sdf] Synchronize Cache(10) failed: Result: hostbyte=0x04 driverbyte=DRIVER_OK Sep 14 19:18:49 Tower kernel: sd 5:0:0:0: [sdf] Stopping disk Sep 14 19:18:49 Tower kernel: sd 5:0:0:0: [sdf] Start/Stop Unit failed: Result: hostbyte=0x04 driverbyte=DRIVER_OK Is the drive dead? Replace? I can't access any SMART data because the drive is offline. Drive is connected using SATA, as I've googled some questions with this error on the forum and it was related to USB. So, what do I do?

I do have a cache ssd (240gb) but that's for downloading only. When I transfer files to the array those surpass 240gb very easily and even a 1TB cache would be filled in a day in some cases. So no mover here. My workaround is have an unassigned device with a 4TB drive that I work and when it's done move to the array. But sometimes I need a file that's on the array and have the spin the drive up, work on unassigned, drive spin down during the nighttime and in the morning, spin up again. So, just a few more options would help.

I believe this is easy to implement but not sure if many people would use it. Sometimes I work on something on my NAS, go to bed, eat breakfest and start working again. Part of my work is in a drive I use a lot, some are not but I can use it in the middle of the morning and this routine isn't very fixed. So, to me a 12h, 16 and 24h inactivity spindown option would be awesome because it would spin down the drives when I'm off for the weekend. Or if I don't need that content so often. The 9 hour limit, to me, causes more spin downs and ups than I would like, so I'm leaving my drives 24/7 spinning. Can we get those options or a custom menu?

So, I'm not exactly a dev but I just tested it this does work. The problem, when DSD is sent as x-dsd many renderers like Jriver, BubbleUPnP, USB Audio Player, even receivers that should support DSD properly over DLNA are actually receiving PCM 352Mhz instead of 2.8Mhz. I forked somebody's copy of MiniDLNA (since the original source isn't on github) and made the changes here: https://github.com/fandangos/minidlna/commit/0f402a51120ca19d15d70291ae3d9cd629d76062 I tested it inside Unraid but since I compiled it on Ubuntu it required a lot of dependencies. After copying all to binhex-MiniDLNA docker container alongside my compiled binary - deps went to /usr/lib - binary went to /usb/sbin/minidlnad I can say it worked as it should I created a snapshot from binhex-minidlna container on my server but this resulted in a 1.5GB tar file and this isn't pratical to share. So, if anyone knows how to build a docker image I can share here for anyone interested in pure DSD playback.

Nice! Thank you

Hey Binhex, do you have a guide of how I could compile this from source?

Quick question, can you buy just the enclousure of the Storinator? My system is ready with all the drives, mobo, sas card etc.. I would like to have a more elegant solution that is easy to add or remove drives.

I don't. I actually keep the unraid tab open and look at it a few times a day since I work from my PC.

One of those is WD. I don't see 200, only 199. 1 is Raw read error rate which is zero. and 199 which is UDMA CRC error count that is 5 and this happened when I swapped the sata cable. This value is like that for a few months now.

Hey big hard drives are not cheap in my country so I'm having a good chance that a friend is bringing me one from Europe. So, my array is: Parity 1x 16TB Data: 5x 8TB 2 of those 8TB already shows Reported uncorrect 4 and another one 8 but doing a SMART extended self-test completes without error. So, basically I have 3 questions: 1. Should I replace one of those 8TB with a 16TB one or can I just add the 16TB into the array for more space? 2. IF one of those drives die, and let's say I have 10TB free on this 16TB one. Can I recover the data into this disk and just remove the old drive from the array? Or is this a bad pratice? 3. I think this is correct to assume but, it's ok to replace a 8TB with a 16TB one, as long as it is smaller than the parity drive. Correct?

Great, thank for opening the issue. I recently tested a PR that fixes DSD sample rate output. Let's see if we can fix the crashes now.

Thank you very muich for this. Do you know how can I start the docker with --debug attribute? I'm having some crashes and adding it as extra parameters and post arguments don't work. Do I need to edit docker entry point.sh?

I only use sata cables with that metal lock for this reason. It's rebuilding now at 96% no errors. This situation poped up when I was replacing the CPU cooler fans, so I might have just touched the cables and moved it slightly or something and when powering it back up again it gave the write error that resulted in all this.

I'm rebuilding it now. Could you please check the SMART that I'm attached just for assurance that the drive is ok or at least okish? Really would like a second opinion on that Raw error data.

I'm reading a few similar threads and I can't believe I'll have to rebuild my entire drive just because of a cable connection issue. One of my 8TB drives is showing UDMC error of 6. When I do a SMART test it passes without any issues. It's a few months old ST8000DM004. Is there anyway around this without having to rebuild it? I have attached the SMART test just in case I missed something. tower-smart-20211228-1615.zip

How can I uninstall the drivers installed with this container? After installing it I can't start my Windows 10 VM with GPU passtrough, it crashes the entire server. Tried removing the nvidia drivers plugin but no success.

First, I would like to thank you. This is amazing. I'm able to play without any kind of conversion, just direct stream, dsf files (sacd) from unraid to my Denon receiver using my phone with Bubble UPnP and it's easy AF! Thank you. My only question is, would be possible that I can do a scan on demand? Some command line to trigger the scan that I can put on user scripts and do from my phone?

Just wanted to ask this to be sure. I have a a share: Music, that have files on all my drives. I've added a new drive and I want to concentrate all the content from inside Music to it's drive. I'm moving everything with unBALANCE. After everything is move to the new drive, can I go into each drive and rm -rf all the music folders that are empty? I want to do this to avoid disks spinning up whenever I access the Music share from another device.

I have no idea where to ask this, so I'll try here. I have SABNZBD downloading to a folder named INCOMPLETE, this folder is on cache SSD set to use only the SSD. When the download ends it starts unpacking the files and it is set to /mnt/disks/2TB/Downloads as RW Slave. This is the unassigned device. The thing is, whenever a file is unpacked from cache to the unassigned device, all my hard drives start up. In file activity I can see only activity on the unassigned device. So, any idea why?

So I have setup everything now the docker claims 40+k points and just the first 9k are registered. Also, if I restart the container it gets reassigned and starts folding if I leave it be, nothing happens. Very different behavior compared to my windows machine folding. Getting a few warnings and errors: ERROR:WU03:FS01:Exception: Failed to connect to 40.114.52.201:80: Connection timed out ERROR:WU03:FS01:Exception: Could not get an assignment WU02:FS01:Sending unit results: id:02 state:SEND error:NO_ERROR project:11779 run:0 clone:8658 gen:5 core:0x22 unit: is this expected?

I've followed the SpaceInvader tutorial for Hardware transcoding. I have a GTX 1060 here which is capable of hevc decoding and encoding. I can do hardware encoding and decoding of h.264 just fine. But when I try to play hevc content like a uhd bluray remux that is 4k hevc it uses software using my cpu. It can see the gpu on emby advanced tab, it even shows it for hevc encoding and decoding both checked but it uses software for some odd reason. Is there anything special I need to do to support HEVC?

Yes it does. That was an easy fix. Now just needs to fix Emby.