Search the Community

hello I connect to this topic, I would like to assemble a new nas baremetal, I would like to install an amd ryzen 5600g cpu, I am not interested in transcoding. do you think it is compatible with dsm 7.1? where can i find the list of compatible cpu?

If the NVMe / HBA / transcoding that I inquired about are all factors you are not interested in, I recommend a Device Tree-based model that makes disk recognition easier. I recommend the DS1621+, which does not require transcoding and has ample bays. You don't need to worry about SataPortMap etc. It maps disk ports on its own.

at the moment I'm testing it on a lenovo pc with i5 cpu and 8gb of ram. So if everything works, I assemble a new case for nas with motherboard, maybe if you can recommend something like hardware .... I am not interested in transcoding, I have to make local and smartphone backups, replace google photos and synchronize the photos of the smarthpne with the nas. also make backups on cloud services and then put on 4 vm servers of few resources. Because I want a fast system.

I'm not sure about the cpu model alone, Any plans to use NVMe or HBA? What is the motherboard model? Any plans for video transcoding? There is too little information.

I test it with Intel G2030 on baremetal. Need to active transcoding or some codecs like AME? I can´t detect persons and vehicules.

If so, it is 12 core 24 ht core. It is too resource-consuming to use only with bare metal. At that level of performance, it is a workstation with excellent performance for rendering or encoding. I'm also a HP Z600 user, and I have the same specs. I didn't use it as XPENOLOGY because the specs were a waste. If there is no transcoding, XPENOLOGY will work well with a Celeron-class CPU. If you have PROXMOX in mind, XPENOLOGY does not need to allocate many resources.

Thanks for the reply @kekesed97 I have 2 Baremetal setups right now and as of now i do not use transcoding or surveillance station. I am really not sure what.... This is why i posted this :). I want to setup something that can support Hot Patching / updates as i only have 1 server so if it reboots there goes all my VM's etc.... I have never played with Proxmox so i am not sure if this is an option to VMware. (Need to Google) I see TrueNas as well but this will then replace XPEnlogy which i do not want to do. Thanks

What is your goal of your setup? The benefit of using VM is you can emulate certain hardware features, or to workaround with some HDD detection problems (using SATA passthrough). I don't know the details, that's what I heard. And using VM also enables you to deploy more than one Xpenology instances rather than a single instance in baremetal (one VDSM is free, but you would need a licence for more). And some of the users said that it boots faster as you don't need physically turn off the motherboard. Are you using transcoding feature? Do you need surveillance station? And the last but not least, never apply changes directly to your production machine. Test it using some dummy HDD you don't care about to see if your target config can satisfy your need. Hope that helps. Cheers!

I have the same issue on my "primary NAS." I got it working on my test system by changing the VT-d value (I forgot if I turned it on or off to get it working). It's not a selectable option on my current setup; it's disabled and listed as unsupported. I tried using the patched library files and reindexing but still no go. I'm toying with the idea of migrating from a DS918+ setup to a DS3622x setup (I read somewhere that that DSM OS has some differences and Photos is more reliable there). I would lose Intel Quicksync as a result so not sure how my circa-2015 i3 would handle transcoding as a result.

The killer e2500 seems to be included in alx ext in REDPILL. https://raw.githubusercontent.com/PeterSuh-Q3/rp-ext/master/alx/rpext-index.json https://manpages.ubuntu.com/manpages/bionic/man4/alc.4freebsd.html DS918+ / DS920+ are required for igpu transcoding, The DS920+ unfortunately doesn't work at all with the HBA. REDPILL's mpt3sas ext for HBA has some problems with DS918+. There is a problem that the serial number is not displayed correctly in Disk Manager. It is expressed as SDA/SDB/SDC, etc. Not recognizing the serial number can be a serious problem if one of the RAID disks needs to be replaced.

Which makes me think I should put windows on it and look at cpu-z, but if I remember fma3 instructions were release at first only starting with Haswell cpus. The problem is that my asrock for 2 cpu slots motherboard fried recently and I have left with 2 cpus 128gb ram, and those motherboards are rare. Only found for single cpu motherboards, some chinese brand. And its interesting for me to resurect the old power of those cpus, as well as to test new loader with new dsm and especially dva but it all takes time and money. And I really hope that by the time I will get it all together again the solution of some sort will be found, since 918+, 3615 or 3617 is not that interesting anymore if its possible to get transcoding or 8 licenses on ss with gpupassthrough for perspective view or as for fail over.

Ok, that’s really good to know. So the drivers are fine 👍🏻 wrt video station, it has own issues and ffmpeg needs to be patched. I haven’t had much success making that work. For me, plex/jellyfin hw transcoding was critical and those have been tested to work.

the gpu passthrough was working fine on Windows (detection, drivers install and so on) so that could not be an Hypervisor problem (VMWare ESXi 7.0f3), vt-d and vt-x enabled, could not figure out why it says "Killed" in Syno terminal, and it still does. i changed the ARPL boot from bios to efi (hypervisor was also already efi boot) and thats the only thing i did actually (and fix the *.bin wrong header files problem in arpl-addons), recreated the docker container with the /dev/dri mapped (permission correctly fixed), enabled the hw transcoding and encoding and now it shows the damn (hw) while streaming. i have to spin up a DVA test vm to see if face recognition also works in Surveillance Station, seems that the ds920+ i'm using does not have it, bummer. i did not test Video Station but i dont actually use that, i guess it "should" also work.

But synology Drive app is working fine. Mobile and desktop upload, generates thumbnail for video. It don't requires transcoding or how it is transcoding?

- Outcome of the update: SUCCESSFUL - DSM version prior update: DSM 7.1.0-42661 Update 3 - Loader version and model: Tinycore-Redpill v0.9.0.2 - DS918+ - Using custom extra.lzma: NO - Installation type: BAREMETAL - Additional comments: Upgraded to DSM 7.1.0-42661 Update 4 via GUI. Booted into TCRP and run: ./rploader.sh update sudo ./rploader.sh postupdate apollolake-7.1.0-42661 exitcheck.sh reboot Transcoding is still working: cat /usr/syno/etc/codec/activation.conf {"success":true,"activated_codec":["hevc_dec","ac3_dec","h264_dec","h264_enc","aac_dec","aac_enc","mpeg4part2_dec","vc1_dec","vc1_enc"],"token":"12345678XXXXXXXXX"}

Thank you Peter Suh. M.2 is not that important to me or this plex server if it will jeopardize the raid 5 btrfs. As I see it in the chart quicksync transcoding is not an option on any other platform except DS918+ I use 8 hdds. 4 8tb 2 10tb and 2 14tb. On the H200. Logically I am out of options unless you can enlighten me with an alternative. Other than Truenas Scale (drives are a mix). or Unraid (poor performance IMO) Thank you again. CD

If an M.2 SATA type SSD is used, it is informed that it corresponds to Muxed in which the SATA port is shared as expected. However, it is said that the port sharing phenomenon can be avoided if the setting is maintained in the M.2 PCIE nvme method. I'm not sure exactly what the setting below means, but If you need to boot to NVME, they are told to change this setting. Advanced => Onboard Devices Configuration => M.2 and SATA Express Mode configuration => M.2 , SATA Express The DS918+ was chosen because of the need to use transcoding and HBAs, but it's not the best. As mentioned last time, HBA is unstable on DS918+. Do the disks use RAID? If RAID is used, when replacement or recovery of a failed disk is required, Synology Disk Manager recognizes the replacement slot based on the disk's serial number. However, DS918+ can only display disk serials as sda/sdb/sdc, etc. This can be a very serious problem because it is impossible to replace a bad disk among RAIDed disks.

What is your motherboard model number? This is to check the possibility that the SATA port and nvme are muxed. If nvme is muxed, you should give up nvme activation on ds918+. Is there a compelling reason to use the ds918+? Is it because of transcoding? Depending on your answer, we can recommend switching to another model. You can look again how to activate nvme.

I have my own fork and patch with all versions to DSM 7.1 42261 Update 4. GitHub - apriliars3/synocodectool-patch: This patch enables transcoding without a valid serial number in DSM 6+

+1 This is not working on DS918+ DSM 7.1-42661 Update 1 Neither Transcoding or Face Recognitions. What I see, Api authentication failed for valid license or no synology account. Which why this is not processing the plugin activation. I hope someone will comes around with something and make it actually work. Thanks

- Outcome of the update: SUCCESSFUL - DSM version prior update: DSM 7.1-42661 Update 3 - Loader version and model: Automated Redpill Loader 0.3a7 DS920+ - Using custom extra.lzma: NO - Installation type: VMWare ESXi 7.03f @ custom Z490-i build (i3-10100T) - Additional comments: Upgraded: via GUI, no fancy stuff to do. Everything works great, CometLake Sata and iGPU in passthrough for real smart data and transcoding.

Hello, Does someone knows how to add --runtime=nvidia in docker ? I've read since docker 20.10 there is no need anymore for docker-nvidia2 https://docs.nvidia.com/ai-enterprise/deployment-guide/dg-docker.html But I'm not sure our Synology docker package actually handle it. When I try to start a container with --runtime=nvidia it fails to start. I've found some infos to add the conf in /etc/docker/daemon.json : https://forum.openmediavault.org/index.php?thread/31206-how-to-setup-nvidia-in-plex-docker-for-hardware-transcoding/&postID=233677#post233677 But file does not exist on Syno, and if I create it, it does not seems to work. With @Zowlverein instructions, there is no nvidia-container-runtime in /usr/bin/ Only nvidia-container-runtime-hook symlinked to /usr/bin/nvidia-container-toolkit I tried it : # cat daemon.json { "runtimes": { "nvidia": { "path": "/usr/bin/nvidia-container-toolkit", "runtimeArgs": [] } }, "default-runtime": "nvidia" } Restarted docker and tried the --runtime parameter : # docker run --runtime nvidia --gpus all nvidia/cuda:10.2-runtime nvidia-smi docker: Error response from daemon: Unknown runtime specified nvidia. See 'docker run --help'. # docker system info Client: Context: default Debug Mode: false Server: Containers: 6 Running: 0 Paused: 0 Stopped: 6 Images: 5 Server Version: 20.10.3 Storage Driver: btrfs Build Version: Btrfs v4.0 Library Version: 101 Logging Driver: db Cgroup Driver: cgroupfs Cgroup Version: 1 Plugins: Volume: local Network: bridge host ipvlan macvlan null overlay Log: awslogs db fluentd gcplogs gelf journald json-file local logentries splunk syslog Swarm: inactive Runtimes: io.containerd.runc.v2 io.containerd.runtime.v1.linux runc Default Runtime: runc Init Binary: docker-init containerd version: 3fa00912415f3e9c6f82dd72119179d599efd13b runc version: 31cc25f16f5eba4d0f53e35374532873744f4b31 init version: ed96d00 (expected: de40ad0) Security Options: apparmor Kernel Version: 4.4.180+ Operating System: (containerized) OSType: linux Architecture: x86_64 CPUs: 6 Total Memory: 31.27GiB Name: DVA3221 ID: ZKJT:3KVD:BSZU:X2SS:HRLW:D4PJ:VPZ3:XO5D:ZXR7:A2H2:CYPT:HLMR Docker Root Dir: /volume1/@docker Debug Mode: false Registry: https://index.docker.io/v1/ Labels: Experimental: false Insecure Registries: 127.0.0.0/8 Live Restore Enabled: false WARNING: No kernel memory TCP limit support WARNING: No cpu cfs quota support WARNING: No cpu cfs period support WARNING: No blkio weight support WARNING: No blkio weight_device support WARNING: No blkio throttle.read_bps_device support WARNING: No blkio throttle.write_bps_device support WARNING: No blkio throttle.read_iops_device support WARNING: No blkio throttle.write_iops_device support Some help would be appreciated Edit : I see there is a /volume1/@docker/runtimes/ folder, but it is empty.

Does it need access to some part of the CPU. On a bare metal or on the real Synology I thought to remember that it requires transcoding capabilities. As an example the DS918+ with an intel J3455 chip uses the Intel HD Graphics 500GPU for this. In order to get this working you need somehow passthrough the iGPU to the XPEnology VM I think to get facial recognition working on unRaid.

Maybe it needs legit SN/Mac for it to work with video station? Not sure, I use plex and hw transcoding is working properly fine.

Hello boys! Can you please tell me if it is possible to enable hardware video transcoding in Videostation or Emby on Xpenology assembly for dva1622? I have a motherboard with a J4125 processor and an integrated Intel UHD600 graphics card. Thank you!