1812

yeah so, if your first pair are 0,14, I would end up probably using 14 (also because most of my other threads are otherwise occupied.) But everyone is going to have a different way they prefer. it depends on many factors: is the thread isolated from unraid? how active and what type of vm is it? not a straightforward question. in the end you'll have to probably do trial and error unless you just want to use 1 thread per isolated pin which will be the "best" to limit IOwait and some stuttering, but also waste 1 thread in the process for anything else. it's all a trade off.

stacking vm's on cpu threads can introduce latency and audio issues. It's fine to share an emulator pin/thread/core among a few vm's as long as it doesn't get overloaded and max out managing them. if you want the "emulator pin to be on the same cores" then don't specify it and it will be there automatically. linux systems have a preference for core 0, so best to avoid. I tend to run an emulator pin on the hyper threaded pair without issue.

I'm experiencing an either inability to access unassigned devices or extremely long timeout when an smb device is set to mount but no longer active on the network. is there a way to lower what I assume is the timeout?

I ran cat 6a into all my rooms a couple months ago. Just waiting on the rest of the networking components prices to drop!

sounds like a plan! in the future, if you ever go 10gbe, you can setup cache disks in raid 1 or 10 for even faster throughput. 10gbe direct connections can be done pretty cheaply now and in some ways easer than a dual nic setup.

1: I don't run my vm's on my system cache disk as they can get interrupted by dockers and their activity and large file transfers to the drive. Yes you could put your 4tb disk as a cache disk, but it's only going to have the performance of that single spinning disk. Depending on your dockers, they may be just fine residing on a spinning disk. What many people have been doing is have 1 ssd (or a mirrored pair) for cache, utilized as array write cache and docker usage, and a second (or in my case 2nd, 3rd, and 4th) ssd mounted via unassigned devices holding vm images. 2. ?

it is advisable to leave a thread or two for unraid, meaning both vm's can have 6. aso, your "ssd cache for raid" should be "ssd cache for array" as unRaid is not really raid. You will not have as fast read/write to the array as you would with traditional raid configurations. you can have fast writes to a cache drive setup on user specified shares. but any data read from the array will be listed to the single drive speed of the disk it resides on. You need to consider whether or not you need dedicated USB cards for each vm. I believe the hot plug plugin is still available, but it means that anytime you want to attach a device you have to go into the unraid GUI and do it IIRC. I run a similar setup (as in 2 vm's with GPU output) but on different hardware and it was a pain dealing with needing usb access until I gave each of them a usb card. Additionally, if you add a device (like a wireless mouse and keyboard) and it becomes disconnected, then it is no longer available in the vm until reboot.

It's not that it's un-askable. It's that it has been covered, and simple search would reveal that. No one expects the Spanish Inquisition. If unraid cost the same as windows server, you and many others like you may have a point. But it's ridiculously cheap for what it does. You're essentially gong to McDonalds and asking why they don't just give you a 10% off coupon for no other reason than you want it and just because you bought a happy meal from them once before. Stop being cheap and recognize that you're already getting a great deal.

not to mention when you have a license that supports "Unlimited attached storage devices" multiple pools/arrays makes more sense. (this coming from someone how has 4 unraid licenses)

bump

I own 4 licenses. Worth every penny spent.

read this thread first:

I don't have any gen 8's and am unable to replicate these problems on older hardware. But looking at your original thread, you had multiple call traces, which are probably not helping. Unfortunately I am unable to help diagnose those. But I'd be curious if they persisted after you modified the tunables.

awesomesauce!

DL 580 G7 . Idled at 600 watts.

I've run it on 4 processors and 88 threads. so, yeah. it's fine.

it's delivered to the same email address (normally)

I have everything working flawlessly now routed through a site url with 1 exception: It's giving the browser an unsafe website waring, saying the "security certificate is from XXXXX.duckdns.org. Shouldn't lets encrypt or the docker page itself be sending the certificate and not that warning? Forgive me as I'm having one of those types of days. I forgot to change the lets encrypt over to that subdomain...

heres a funny thing: so I set it up and theme'd it up. then once I route to it via the public web address/url, all theming changes are lost.... edit-- think I figured out my issue... messed up the config files

I tired it when I was having major issues with an unstable internet connection and switched from pfsense to rule out some sort of error. It ended up being a flaky cable modem that was causing problems but not showing them diagnostically. But I found the UI to be much cleaner and nicer This is probably the biggest reason I stayed on it. Some things/settings are easier to find in terms of navigation, sort of where you expect them to be vs pfsense. Others seem way different if you've learned where they are in pfsense. OPNsense says they do more updates vs pfsense and I believe them. It seems like something is updated/patched every week. There is less documentation than pfsense, but most issues are cross-resoveable if you can find the settings.

so even if it's enabled on the ports that are forwarded, I'm looking for a general "allow nat reflection" or similar, correct? Found the setting finally :Firewall: Settings: Advanced--- Automatic outbound NAT for Reflection thanks!

that's what I was guessing and digging around in for the past hour. I'm using opnsense and can't find anything about it.

FINALLY I managed to get sonar working with duck dns, but when I setup next cloud, it works only outside of my network. On the network when you try to access the address, it forwards to the duck dns which never resolves... any idea what I messed up? I feel like I've triple checked all the settings.

smb over gigabit networking. you can install the Krusader docker and manage files that way easily on the server. You would would first ant to mount the remote server in unassigned devices. you can do parity before or after, just depends on if you want protection immediately with the transfered data or later. It would a little bit faster if you wait on parity.

You could try dumping the vbios, or modifying one from tech powerup and passing that to the vm. https://www.techpowerup.com/vgabios/?architecture=NVIDIA&manufacturer=NVIDIA&model=Quadro+4000&interface=&memType=&memSize=&since= https://www.youtube.com/watch?v=1IP-h9IKof0 but also