AnnabellaRenee87

So did exactly what @limetech said to do then rebooted. I went into my XML file for the VM I'm working with and added this to my XML after the cputune section. <numatune> <memory mode='preferred' nodeset='1'/> </numatune> After doing that I get this now when running numastat -c qemu. I was gonna us the "strict" option but it didn't seem to really work or anything. After a little playing around I decided to move my 32GB or RAM Plex Media Server VM to my Node 0 which is GPU Accelerated. I moved my 16GB of RAM Gaming VM to Node 1. I moved my pfSense 8GB of RAM to Node 0. My Ubuntu Server VM which is 16GB of RAM which isn't mission critical to have high performance I just have it set the normal without the "numatune" section added to the XML file. Here it is now. Some of the VM is getting put in to the other node but I think that's the RAM being used for the QEMU service and the RAM in the Node is equal or larger than the allocated memory. You can read more about how to use this in the following sites/threads. https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux/6/html-single/virtualization_tuning_and_optimization_guide/index#chap-Virtualization_Tuning_Optimization_Guide-NUMA https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux/7/html/virtualization_tuning_and_optimization_guide/sect-virtualization_tuning_optimization_guide-numa-numa_and_libvirt So far everything looks to be good but I need to do some testing on the VMs to see if its fixed any performance issues.

I'm sorry for being so annoying about this but I've been having bad GPU performance and lag in VMs and from the looks of it my VMs RAM is getting split between NUMA Nodes even though I only pin cores that belong to one or the other CPU. Here is the original discussion about it that I know of. https://forums.unraid.net/bug-reports/prereleases/660-rc2-vm-memory-allocation-across-numa-boundaries-r151/ Sent from my SM-G955U using Tapatalk

From looking around it looks like it's here https://github.com/K1773R/numad/blob/master/numad.c

Can you please add NUMAD?

Can NUMAD please be added?

So when I hot plug/added new SAS drives to my server they would show up fine in Unassigned Devices, the power went out and the UPS didn't have enough power to keep the server up (I'm working in getting a better UPS). After booting back up the SAS drives are not showing up in UD. I pulled some of them and put them back in as their not assigned anywhere and found that only one of them are showing up now. Am I doing anything wrong? server-diagnostics-20190121-1838.zip

NUMAd needs to be added, Here is what I was talking about. I'm sorry to make you wait so long but when I said it should help I meant I think this will help. Here is the main thread where we were talking about it.

I have similair problems with my 1080Ti in my HP DL370. I'm thinking it's because there's no way to force VMs to stay in their own NUMA pools yet. When I get to a computer I'll update this with the link to what should fix it. Sent from my SM-G955U using Tapatalk

I just figured it out, after turning it off and hitting Apply I have to go back to the top and toggle on the "Uninstall" switch and it removes everything that isn't on, so when I did that it pulled atop out and left the others. Thanks everyone

Okay, I feel dumb, how do I remove things in Nerd Pack? I turned the switch for it off but it didn't remove it........ I'm sure I'm missing something.

Looks like I do have atop installed. Should I rip it out? I can't remember why I installed it come to think of it, lol

Can someone look over my logs and see why my logs are getting full? server-diagnostics-20190117-1715.zip

Yeah, like a small hint of what changes are coming would be amazing, all I know is icons are apparently gonna change. lol Sent from my SM-G955U using Tapatalk

Yeah, a few of the plugins lately have been mentioning 6.7 for last few weeks, seen how this one requires 6.7 and was like "hopefully the forums will start talking about it now" lol Sent from my SM-G955U using Tapatalk

I've done lots of research on this subject. Depending on the video being transcoded it can take all the load off the CPU. I have my PMS in a VM with 32GB of RAM with most of that allocated to a RAM drive for transcoding. My GPU I'm using is a old GTX 760. One day I will get a Quadro P2000 but for now since most of my stuff is H.264 it's working fine. I have noticed that depending on the video some h.265/HEVC videos will go through the GPU too. That VM also has 6 threads (3 cores) assigned to it. If you do live TV just note that currently it won't transcode anything recorded in the GPU. You can turn the option on in the settings to have it transcode recordings but it only does it after recording and even then it's done on the CPU. If watching live TV then it will use the GPU for the client. If you got one of those tuners from Hdhomerun that does transcoding in the tuner then you don't need to worry about any of this because that box will send out a h.264 stream to your PMS. Make sure you're using Windows for you PMS VM as it's the only OS that allows decoding and encoding at the same time on a Nvidia GPU. Finally consider cost of resources. Compair the cost of running a GPU 24/7 vs. buying an Nvidia Shield TV and running the built in PMS in it (which does GPU encoding for everything up to h.265). It may be easier for your situation to do that vs. worrying about your server, having an extra slot taken up by something, etc. So in closing let's me give you the breakdown of my PMS Server/VM. OS: Windows 10 Pro build 1809 Used group policy to disable the auto reboots for updates, disabled sleep settings. Installed ImDisk Tools for the RAM drive, I have mine at 28 GB (saves on wear and tear of SSDs/Cache arrays.). Windows doesn't have to be activated, mine isn't. RAM: 32 GB Maybe overkill but my server has 96GB of RAM and I never used half of it so why not? CPU: 3 Cores I just gave mine 3 cores which is 6 threads. Video Card: GTX 760 It transcodes most of my library so for now it's good but in the future when prices come down I'll get a Quadro P2000 Just map a network drive to your media share and setup Plex as normal after that. Any questions after that just ask. I gave you all the information I had because a lot of this isn't documented or documented very well, hope this helps! Sent from my SM-G955U using Tapatalk

Strange, I'm using the same patch on my server with 6.6.6 on my DL370 G6. Sent from my SM-G955U using Tapatalk

Yes, please add this! Being on a dual CPU system, my VMs memory is being devided between the two nodes. @limetech

Yeah, having access to NUMAD would be great, I have a Dual CPU system and I have horrible lag in a lot of things. I can see now that my VMs memory is being split between the memory pools/NUMA Nodes. @limetech? If I go by the instructions at https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux/6/html-single/virtualization_tuning_and_optimization_guide/index and add <numatune> <memory mode='strict' placement='auto'/> </numatune> or <numatune> <memory mode='strict' nodeset='1'/> </numatune> to my XML file I get this error;

Looks to be working now!! Seeing these error on the server's screen tho. Cant remember if vnstatd is a Plugin or not, don't know what cat is.

Sorry for posting so much, I was able to downgrade with iLO. I used Midnight Commander and moved all the files from the "Previous" folder back in to the /boot folder and was able to get back to 6.6.0 My Web GUI wouldn't load when I attempted 6.6.2

White rebooting I seen this screen for a few moments before getting back to the previous screenshot too. Sent from my SM-G955U using Tapatalk

Looks like we have a problem [emoji22] Sent from my SM-G955U using Tapatalk

@limetech Any way this could be integrated in to the main release or would it cause havoc on non ProLiant peoples?

@limetech don't suppose there is a newer version of my Network Drivers is there?

So I did this on my ProLiant DL370 G6 and got this back, whats up with these thread numberings?