planetwilson

I am building a new server and moving unRAID over to it. I was planning to make a note if exactly which drives are acting as data vs parity then just plugging it all in and hoping for the best. I know my VMs will likely need a bit of re-configuring but is there anything else I need to bear in mind. I am moving from a quad core Xeon based Lenovo TS140 to a 14 core Xeon/X99 mobo/64GB ECC custom machine. Anything else I need to keep in mind?

Just out of interest - what does this do? I am attempting a new build myself at the weekend and initially will have a single GTX 950 until I can source another 750 for the Mac VM.

Great video (as usual) Anyone know when kernel 4.10 might end up in unRAID? I could really do with the HyperV stuff working for some DSC stuff in Windows....

I have 4K from displayport at 60Hz working on a GTX 950. I am not surprised though that it isn't working on HDMI. It requires HDMI 2.0 supported at both ends and I thought the graphics cards only started doing that quite recently. I don't think either of those supports HDMI 2.0

I am wondering about that exact point. I just bought a Xeon E5-2683 v3 in the UK from eBay, retail version. However I am thinking I might be better off with a Ryzen build, I know it is 8 cores as opposed to 14 but the clock speed is higher and can be overclocked too. Just hard getting the boards for them right now. Decisions decisions. Put this one back on eBay and go Ryzen or stick with it?

So if you aren't planning on doing crossfire with gfx cards - is there any point in going for X370 over B350? large price difference in the boards. I am still not 100% clear on the PCI lane thing, will it be an issue running M.2 SSD cards and a couple of graphics cards in there (not crossfire, just hooked up to different VMs)

Is there any sort of API available to remotely manage VMs? It would be useful to be able to stop/start them through a REST API as I could do with an app on my phone to quickly start my main desktop one for instance. Sent from my iPhone using Tapatalk

Ha ha, never thought of it like that. It's more the clock speeds...although tbh at an all core boost of 2.6 or so and peak of 3.0 I'll probably be fine. can't grumble at the cost.

Damn! starting to regret my hasty E5-2683 v3 purchase this morning for £250

Did you build this in the end? I have just bought that CPU and was also thinking about the same motherboard as it looks good value...

I am in the same boat, Ryzen or a Xeon E5 build? The Xeon's just don't clock as high in general but wondering if I should be too worried. I am not a massive gamer so as long as they boost up to say 3GHz then together with my GTX 950 I think it would be one for moderate gaming. Then I could still have 12 or 14 cores for VMs...those X99 mobs are a bit pricey as well. The Ryzen equivalents look cheaper.

So I have been slowly thinking about replacing my Lenovo TS140 32GB quad core Xeon server with something a little beefier. I was thinking along the lines of what Gridrunner has, a nice Engineering sample Xeon with say 14 cores and an X99 mono to take plenty of RAM. I might end up going with a tower though like the Phanteks Evolve or similar as it has a smaller area foot print. But I am still in two minds about the Xeon, it is hard getting a decent ES one where the clock speeds are not massively downgraded from stock. You also take a risk on whether they fully support VT and so on. With the appearance of Ryzens though it looks like I can have decent clock speeds *and* a higher number of cores. I know 8 isn't quite as many as 14 but it would probably be fine for the VM work I want to do. Anyone done anything similar or thinking about it (not sure if the Ryzens are widely available yetin the UK, looks like I can find 1700 ones) Would I hit any issues with running VMs with these, would it need KVM to be updated?

My fault. I had replaced ngingx.conf wholesale from the previous Docker. I did it again just transferring my server entries and all working now. Can't quite spot the difference though!

I have switched from Aptalca's docker which was working perfectly to the linuxserver one, kept my old nginx config and now getting this error:- nginx: [emerg] bind() to 0.0.0.0:443 failed (98: Address in use) I have the https port mapped to 4443. The reason for this is that I also run an OpenVPN-AS docker which intercepts all 443 traffic coming into the NAS (so I can run a VPN to home on TCP/443 and get around firewall restrictions in a lot of locations). You can configure OpenVPN to pass non-VPN traffic on to another server, in this case my letsencrypt/nginx docker which I have set to 4443 normally (with it appearing as 443 inside the docker) Now this worked perfectly before but doesn't with the newer linuxserver docker. Is this new docker hard-wired to only work with 443 somehow? doesn't seem like that should be the case?

Aah no I didn't actually! That is the example I copied from somewhere, my actual images reside on /mnt/cache/VM

Installing the drivers did it! many thanks, not sure I managed to miss that. I'll also make the xvga change too.

So I have had a working OSX Sierra VM thanks to gridrunner's videos but I am keen to get it working with GPU passthrough. I tried copying the <hostdev> entries from my fully working Win10 GPU passthrough VM but that doesn't seem to work. I mean it will boot into OSX but the screen is massively corrupted. I see 4 small stretched versions of the screen on the monitor. I can log into Splashtop fine. I am not sure if I need to also change something in Clover config somewhere? VM XML:- <domain type='kvm' xmlns:qemu='http://libvirt.org/schemas/domain/qemu/1.0'> <name>macOS Sierra with GPU</name> <uuid>f2ebef5b-1639-a5fe-8b1f-929e4028b9b3</uuid> <metadata> <vmtemplate xmlns="unraid" name="Ubuntu" icon="OSX-10.12.png" os="ubuntu"/> </metadata> <memory unit='KiB'>4194304</memory> <currentMemory unit='KiB'>4194304</currentMemory> <memoryBacking> <nosharepages/> </memoryBacking> <vcpu placement='static'>4</vcpu> <cputune> <vcpupin vcpu='0' cpuset='0'/> <vcpupin vcpu='1' cpuset='1'/> <vcpupin vcpu='2' cpuset='2'/> <vcpupin vcpu='3' cpuset='3'/> </cputune> <os> <type arch='x86_64' machine='pc-q35-2.5'>hvm</type> <loader readonly='yes' type='pflash'>/usr/share/qemu/ovmf-x64/OVMF_CODE-pure-efi.fd</loader> <nvram>/etc/libvirt/qemu/nvram/f2ebef5b-1639-a5fe-8b1f-929e4028b9b3_VARS-pure-efi.fd</nvram> <boot dev='hd'/> </os> <features> <acpi/> </features> <cpu mode='custom' match='exact'> <model fallback='allow'>Penryn</model> </cpu> <clock offset='utc'/> <on_poweroff>destroy</on_poweroff> <on_reboot>restart</on_reboot> <on_crash>destroy</on_crash> <devices> <emulator>/usr/local/sbin/qemu</emulator> <disk type='file' device='disk'> <driver name='qemu' type='qcow2' cache='writeback'/> <source file='/mnt/user/VM/macOS Sierra/vdisk1.img'/> <target dev='hda' bus='sata'/> <address type='drive' controller='0' bus='0' target='0' unit='0'/> </disk> <controller type='usb' index='0'> <address type='pci' domain='0x0000' bus='0x02' slot='0x02' function='0x0'/> </controller> <controller type='sata' index='0'> <address type='pci' domain='0x0000' bus='0x00' slot='0x1f' function='0x2'/> </controller> <controller type='pci' index='0' model='pcie-root'/> <controller type='pci' index='1' model='dmi-to-pci-bridge'> <model name='i82801b11-bridge'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x1e' function='0x0'/> </controller> <controller type='pci' index='2' model='pci-bridge'> <model name='pci-bridge'/> <target chassisNr='2'/> <address type='pci' domain='0x0000' bus='0x01' slot='0x01' function='0x0'/> </controller> <interface type='bridge'> <mac address='52:54:00:51:66:48'/> <source bridge='br0'/> <model type='e1000-82545em'/> <address type='pci' domain='0x0000' bus='0x02' slot='0x03' function='0x0'/> </interface> <memballoon model='none'/> <hostdev mode='subsystem' type='pci' managed='yes'> <driver name='vfio'/> <source> <address domain='0x0000' bus='0x01' slot='0x00' function='0x0'/> </source> <rom file='/mnt/user/VM/vbiosbound.rom'/> <address type='pci' domain='0x0000' bus='0x02' slot='0x04' function='0x0'/> </hostdev> <hostdev mode='subsystem' type='pci' managed='yes'> <driver name='vfio'/> <source> <address domain='0x0000' bus='0x01' slot='0x00' function='0x1'/> </source> <address type='pci' domain='0x0000' bus='0x02' slot='0x05' function='0x0'/> </hostdev> </devices> <seclabel type='none' model='none'/> <qemu:commandline> <qemu:arg value='-device'/> <qemu:arg value='ioh3420,bus=pcie.0,addr=1c.0,multifunction=on,port=2,chassis=1,id=root.1'/> <qemu:arg value='-device'/> <qemu:arg value='usb-kbd'/> <qemu:arg value='-device'/> <qemu:arg value='usb-mouse'/> <qemu:arg value='-device'/> <qemu:arg value='isa-applesmc,osk=blahblahblah'/> <qemu:arg value='-smbios'/> <qemu:arg value='type=2'/> <qemu:arg value='-cpu'/> <qemu:arg value='Penryn,vendor=GenuineIntel'/> </qemu:commandline> </domain>

Sorry if this has been answered, I did find a mention this error but it sounded like a problem with the Docker container? When I attempt to update my nginx + let's encrypt docker I get:- "Error: layers from manifest don't match image configuration" Do I copy off my config somewhere, make a note of the settings and delete / re-download? If. do that will it need to regenerate the certificates again? Currently got t sitting behind an OpenVPN docker which intercepts all 443 traffic. I can get that out of the way I guess for a new setup...

I think there is very little real world difference between those processors. Xeons are designed for servers so you often hear that they come from the "best silicon" etc but not sure it is true or not. The Xeon will support ECC RAM but if you are not planning on having ECC RAM then it doesn't matter. Would the choice mean a different motherboard for each one? If so perhaps the motherboard features could sway it for you?

Before I try gong down the GPU passthrough route. Does anyone know how to set the resolution for VNC or splashtop? I tried updating the config.plist file in the EFI partition for Clover but it seemed to make no difference. <key>ScreenResolution</key> <string>1280x1024</string>

I just rewatched the video - I have no idea how I missed that!! Think I got distracted half way through. Doh! Will sort that out now.

I get to clover yes and choose the install option. At some point during the Apple logo showing a bunch of text appears on the screen, I don't get a lot of chance to look at that before it reboots. Then there is a no entry sign. Penryn is specified in the XML, is that what you mean? <qemu:arg value='Penryn,vendor=GenuineIntel'/> update: added XML file to this post sierra.txt

Download is working for me. I did however have an issue in creating the ISO image from the USB key. Not sure if this is a Sierra problem or not. In the end I have gone with a hybrid approach and created a blank file image using the instructions from danofun here:- https://lime-technology.com/forum/index.php?topic=43816.420 but then used the EFI folder provided by grid runner on his video example. Although I am obviously doing something wrong as I never get to the installer. get the Apple logo, then a loa of text, a reboot and then a no entry sign.

I am doing what you have there but on a much smaller budget than comfox I picked up a Lenovo TS140 server, one of the E3 Xeon models, fitted 32GB ECC RAM to it. This model comes with a beefier power supply than the non-Xeon models so I could get away with a graphics card which requires the 6 pin extra line (again the lower TS140s don't have this extra cable in them) so I went with a Geforce GTX 950. Got 3 x 3TB drives in for storage, a 512GB SSD for cache / VM duties. Whole thing runs really quietly as the E3 Xeons sip power when not under load. Total price around £1000 I suppose but then there was a cashback deal on the server which gave me back 100 or so. Runs VMs without any issue. I have a Win10 VM with GPU pass through for gaming. Currently getting OSX up and running. Also use various Windows Server VMs for development work and have 3 running at once (domain controller, SQL, other stuff) etc all without breaking into a sweat. I also have dockers running for things like SAB, Plex, OpenVPN, nginx etc.

Thanks that helps quite a bit. I still have to use that plugin to re-attach if I swap but it doesn't happen often.