Keichi

Hello, Last week, I explained that an ASM1061 controller was not working in UNRAID, for my configuration: https://forums.unraid.net/topic/102010-recommended-controllers-for-unraid/?do=findComment&comment=1267945 Unfortunately, I didn't keep the diagnosis. I acquired this other controller, based on the ASM1062 with 4 ports: https://www.amazon.fr/dp/B07TD57RGZ I made a mistake (in my opinion, with 4 ports, it multiplies the SATA ports, but on this trial, only one disk was connected). I had the same problem: - the disks on my HBA got nothing (my array -last time, disk1 was disable because connected to the motherboard, so i move it to the hba to avoid another reconstruct-) - on the other hand, the disks connected to my motherboard (ASRock X570S PG Riptide) were duplicated (in this case, the cache pools). I've ordered a third card (https://www.amazon.fr/dp/B0BNCHQBK8, 4 ports, ASM 1064) recommended by JorgeB (here: https://forums.unraid.net/topic/102010-recommended-controllers-for-unraid/) and I'm getting it tomorrow, so I'll see what happens (the aim being to plug in 3 SATA doisks for cold backup, so speed doesn't really matter to me). But in my opinion, the problem will be the same.... Would you have a good card that plugs into a PCIe 1x 3.0 port to recommend? Thanks a lot! I add the diagnosis and a screenshot. K. ASM1062 - 4 ports - yggdrasil-diagnostics-20230605-1850.zip

Ok, thanks! Still 8hours in rebuild, i will see after. Or maybe try to replug the card after a full backup, if i can reproduce it and help the commonity. K.

Hello JorgeB, I did not start a new thread, because i knew you will ask for that. I forget to download the diagnostic before rebooting, so i don't know, sorry. Problems with unraid are unsual, I didn't have the reflex to download the diagnostics K.

Hello JorgeB, I bought this card : https://www.amazon.fr/BEYIMEI-contrôleur-dextension-Adaptateur-Ordinateur/dp/B07PRRQ41J/ref=cm_cr_arp_d_product_top?ie=UTF8 It is a ASM1061, which is supposed to work with Unraid. After plug-in it (with one SSD attached), four drives (connected to the motherboard) went in error state, and one of these (non-pool drives) were "disabled" (following your advice, i ran a filesystem check and it is now rebuilding). I don't undertand why. I have 2 CGs (PCIe 3.0 x8 and x4) and one HBA LSI 9200-8i (PCIe 3.0 x2, if i understand well my ASRock X570S PG Riptide motherboard), and two PCIe 1x port are masked (because of CG), so only have one PCI 1x availaible. I thought it was a good option. Do you think i can try your "4 ports: Asmedia ASM1064 (PCIe 3.0 x1)" (like this one : https://www.amazon.fr/IO-Crest-ASMedia-Non-Raid-Support/dp/B08J2PK1HF), or i will have the same issue ? Thanks for your time, K.

Hello, As everyone, i was seeing some warnings in the syslog. Thanks to this thread, i updaded my 9211-8i from : Num Ctlr FW Ver NVDATA x86-BIOS PCI Addr ---------------------------------------------------------------------------- 0 SAS2008(B2) 10.00.05.00 0a.02.00.18 07.19.00.00 00:04:00:00 To : Num Ctlr FW Ver NVDATA x86-BIOS PCI Addr ---------------------------------------------------------------------------- 0 SAS2008(B2) 20.00.07.00 14.01.00.08 07.39.02.00 00:04:00:00 Without any issue. Thanks to the author! K.

Hello Jorge, It is in fact what i try to do : eth0 : 192.168.0.7, member og br0 eth1 : 192.168.1.1, member of br1 And my Windows VM use br0 with virtio. I can access in the VM the webGUI with 192.168.0.7 but Files Explorer is unable to browse shares. It there any differences bet ween .1 and .10 ? Thanks

Dear community, My Unraid server is connected to my 1Gb wifi router. Recently, i bought a 2.5GbE card and an 2.5GbE USB adapter for the computer, in order to make direct connection and have better speed. I configure the 1Gb wifi with eth0 and the 2.5GbE with eth1. As expected (i guess...), Fix Common problem is unhappy with the error : Even if my computer correctly use the 2.5GbE, and others devices on the network the 1Gb Wifi, my VM are now unable to access the Unraid server files but can access the webGUI with no issue. I know it is not well configure, but i don't know how to do it. Can you provide me some help ? yggdrasil-diagnostics-20220911-1146.zip Thanks, K.

Hello, no one for help ? Thanks

Hello, I am trying to do a simple thing : copy my Google Drive in /mnt/user/GoogleDrive. But i missed something... Here the details if you can help, thanks. I successfully installed this Docker with that : I can navigate from the docker to my /media path I can see my Google Drive on the WebGUI : But in the log, i have : I try a lot of things for RCLONE_MEDIA_SHARES, but i think i don't understand what it is. Thank you for your assitance, K.

Hello everyone, I understand that we can get the running status of UserScripts inside Unassigned Device. But do you now if it is possible to automatically launch a userscript when a device is connected ? I backup all my NAS with spare hard drives and create a few userscripts to launch some rsync. Thanks !

Same issue for months now....

Same issue for months now....

Same issue for months now....

Ok, thanks Itimpl, still have the warnings, even with "0" in the field, which should disable it. I put the emails in spam and waiting for an UNRAID update then

No one to help me ?

Thanks, i edited my post with the diagnotic !

Hello everyone, For the last ten days, Unraid sends me two warnings/mails per day (x10 drives !!!) because of disks on high usage, which is incorrect. And the next minute, another warning saying that everything return to normal utilization. It seems to started unexpectedly, everything was working good, and i don't think i made any change. Do you know how can i avoid it ? Thanks ! yggdrasil-diagnostics-20210817-1337.zip

Hello Squid and thanks for the quick answer. 1/ SSD : i didn't know about VM backup plugin, and i guess it can do the job. 2/ Cache Pool : I understand that it is the best choice. Only interrogation i have left is about Mover. From what i read, it is said that the mover do not "move" the file in use, and i guess the vdisk will always be in use. So, it seems good in you have 2 SSD in cache pool with protection, but not so good if you use only one SSD, so without protection. Or i use one SSD in Cache Pool along side with VM backup plugin ? That seems to be the good solution ? Thanks !

Hello, I have a UNRAID server with 10 mechanical drives (and different "quality", replace the old ones one by one). As noticed, the write on the array rarely exceed 80Mb/s, so my 1Gb/s network is never the bottleneck. Right now, my VM is inside the array and i would like to move it to a spare SSD i have. I don't really know what is the best pratice : - Unassigned device with rsync backuping it every night on the array - Cache pool only for VMs (did it move everynight inside the array ?) - The new "pool" feature i don't really fully understand yet ? I don't want to lose the array protection, even if it is only every 24h, like daily backup without shuting down the VM. Thanks. Thanks !

Yes, that is what i try. Sadly, not all Docker from DockerHub start on UNRAID, so sometime, i don't know. I will look into Emby, thanks. Yes, i suppose : And i can't launch ffmpeg from the console : I will continue to installa and try several dockers, Unraid is really flexible, just need to dig a little deeper and understand how everything is working Thanks again!

Handbrake doesn't support h265 NVEnc with 10 bits sadly. Thanks ! I missed that, but even with it, FFmpeg-mkvdts2ac3 doesn't have hevc_nvenc encoders I will reach the dev to see if it is possible to add it. Jellyfin comes with an old (4.1) ffmpeg. But thanks for the idea! In fact, i use ffmpeg with libx265 which is great with a lot of options, but CPU encoding is really slow (~2 fps/s) so i bought a GeForce GTX 1650 SUPER with all the h265 features of Turing architecture, hoping to use NVEnc, but seems harder than expected on Unraid. I can't find a way to have as many seetings as libx265. ffmpeg -i Japan.mkv -pix_fmt yuv420p10le -c:v libx265 -profile:v main10 -preset slow -crf 23 -x265-params "no-info=1:level-idc=5:allow-non-conformance=1:wpp=1:tu-intra-depth=1:tu-inter-depth=1:cutree=1:early-skip=1:limit-refs=3:cu-lossless=1:aq-mode=1:b-intra=1:constrained-intra=1:rc-lookahead=72:rc-grain=1:weightb=1:weightp=1:deblock=0,0:sao=1:sao-non-deblock=1:selective-sao=4:subme=7:me=3:rect=1:amp=1:rd=4:psy-rd=0.8:psy-rdoq=1.80:rdoq-level=1:signhide=1:temporal-mvp=1:strong-intra-smoothing=1:bframes=8:ref=4:b-adapt=2:qcomp=0.7:cbqpoffs=-1:crqpoffs=-1:qpmax=54:qpmin=0:analyze-src-pics=1:no-fast-intra=1:pass=3:stats=1:analysis-reuse-file=1:multi-pass-opt-analysis=1:multi-pass-opt-distortion=1:opt-qp-pps=1:opt-ref-list-length-pps=1:multi-pass-opt-rps=1:scenecut-aware-qp=1:scenecut-window=650:fades=1" -c:a libfdk_aac -profile:a aac_he -vbr 5 Japan-enc.mkv Don't want to start a discussion about quality either, i just do my encode, compare to the source, and tweak my setting to find a good balance between size and quality. So it is purely subjective to my eye. I will look on how to dockerised NVEnc which i was hoping to use. Thanks for your time ich777, i do not disturb you more, your plugin is perfect and work flawlessly, thanks for your work, it is something essential to have on Unraid ! K.

I don't use Ember or Plex, as i just want to transcode manually, like using CLI. So i install Unmanic which use ffmpeg 4.3 and according to the help, the options i can pass using HW transcoding are really limited (no bframes, no multipass, no -mv precision, no aq...). I try also FFmpeg-mkvdts2ac3 (with --runtime=nvidia & NVIDIA_VISIBLE_DEVICES) but it does not have hevc_nvenc installed. That is why i said there is no "full spport". But i can be wrong, i am a newbie in transcoding. But what i can say is that i can't reach the level of quality i get with CPU encoding, even with bigger files. Thanks for your time ich777!

Hello ich777, Thanks for the quick reply. In fact in the installation procedure, some examples of Docker are given using this plugin (Ember, Plex,...), but none of these took advantage of the full support of NVEnc encoding (with h265). I guess that some people used it for encoding and maybe have thought about the support of NVEnc in UNRAID. It is not related to the plugin, but is possible thanks to this plugin, that is why i posted here. But maybe i made a mistakes and need to create a new topic, sorry. K.

Hello, I successfully installed the plugin. It works correctly, with Unmanic. Thanks to all of you! However, the NVEnc implementation in ffmpeg is very old and I can't encode properly using "hevc_nvenc". Do you know if there is a way to use NVEnc directly (https://github.com/rigaya/NVEnc) on Unraid while taking advantage of my GeForce GTX 1650 SUPER? I can't find any NVEnc Docker and my hardware being too old, I can't install a VM on the NAS and therefore run Windows taking advantage of the GC Thanks to all of you ! K.

For the log, i didn't think about it and it seems to be because the file is know : crond: USER root pid 1406 cmd /usr/bin/flock -n /tmp/my.lockfile /bin/bash /ffmpeg-dtsac3/encode.sh -i /watch -o /output -m /copy 2020-12-20-12:54:00 Not a file. HDM6.mkv Allready exists in files.txt After renaming the file, ffmpeg start right now with the encode. BUT Even with ffmpeg v4.3.1, i still have my issue : I think this is because libx265 is in version 3.1.1+1-04b37fdfd2dc. The latest is 3.4. According to the Version History : - selective-sao was added in 3.2 - scenecut-aware-qp was added in 3.3 - scenecut-window was added in 3.3 Thanks !!