February 14, 20179 yr My 2 gamers 1 PC build is essentially finished. Big thanks to everyone that's helped me with it in past posts. But I have an issue with memory. The machine has 48GB. I've assigned each VM only 16GB. I had assigned more but the dashboard kept telling me the memory was maxing out and my VMs would crash. But I've got them down to 16 now, for a total of 32 out of 48 being requested, and the dashboards memory usage monitor keeps creeping up to 100% again. Any ideas why this is happening? Side note: Device manager in each VM has "PCI Device" and "PCI Simple Communicaions". Those drivers in the e drive where the ethernet driver was? Which one's are they?
February 14, 20179 yr This is in the wrong forum, you might want to request the moderators move it. I assume you are familiar with http://www.linuxatemyram.com/ ?
February 14, 20179 yr The dashboard no longer displays ram usage including the cache Sent from my LG-D852 using Tapatalk
February 14, 20179 yr Author OK. The ate my ram site is useful. I'll try running that command as it looks like the ram is running out. Are you sure it doesn't display it? There's a big memory usage bar right under cpu usage. Is it normal for the VMs to fail booting without any errors in the log if I start them at the same time? It seems I need to let one finish booting before I can boot the other. This sort of makes sense as I assume memory allocation probably needs to finish for the first VM before it can start for the second, but if that's the case, if I let them auto start, I'm guessing they would fail.
February 14, 20179 yr attached images of dashboard ram usage and dynamix system stats usage. Dashboard does not show cached usage, otherwise it would show 99%.
February 14, 20179 yr Author attached images of dashboard ram usage and dynamix system stats usage. Dashboard does not show cached usage, otherwise it would show 99%. OK. So I'm referring to that memory usage bar that yours is at like 46. Mies starts at 85 and creeps to 98+ even when the 2 VMs are only assigned a total of 32 out of 48 in the machine.
February 14, 20179 yr attached images of dashboard ram usage and dynamix system stats usage. Dashboard does not show cached usage, otherwise it would show 99%. OK. So I'm referring to that memory usage bar that yours is at like 46. Mies starts at 85 and creeps to 98+ even when the 2 VMs are only assigned a total of 32 out of 48 in the machine. Have you tried to stop your VMs and see if memory usage stays low?
February 14, 20179 yr Author attached images of dashboard ram usage and dynamix system stats usage. Dashboard does not show cached usage, otherwise it would show 99%. OK. So I'm referring to that memory usage bar that yours is at like 46. Mies starts at 85 and creeps to 98+ even when the 2 VMs are only assigned a total of 32 out of 48 in the machine. Have you tried to stop your VMs and see if memory usage stays low? Not for a long test....but for a good 15 minutes. It just sits at 2%.
February 14, 20179 yr There were reports in the past about memory leakage when assigning high memory values. Can you try 8GB or 4GB assignments and see if that makes a difference?
February 14, 20179 yr Author There were reports in the past about memory leakage when assigning high memory values. Can you try 8GB or 4GB assignments and see if that makes a difference? Will try that tonight.
February 14, 20179 yr There were reports in the past about memory leakage when assigning high memory values. I'm on 6.2.4 and assign 32GB of ram to a single vm and don't see this. But also not on win10 vm. Also the screenshots I posted earlier are of 2 vm's running at the same time, and have been on for 5 days and don't show the same symptoms of the OP. Also not windows vm's. Not saying it's not a possibility here, just sharing personal experience. I'm also using ECC ram (though not sure that it matters?)
February 14, 20179 yr Author There were reports in the past about memory leakage when assigning high memory values. I'm on 6.2.4 and assign 32GB of ram to a single vm and don't see this. But also not on win10 vm. Also the screenshots I posted earlier are of 2 vm's running at the same time, and have been on for 5 days and don't show the same symptoms of the OP. Also not windows vm's. Not saying it's not a possibility here, just sharing personal experience. I'm also using ECC ram (though not sure that it matters?) It could definitely be something unique to me and my setup. I'm using non ecc ram on a board that does support ecc. But I have ecc turned off in bios. I will continue to tinker with bios adjustments and other things on my own to try to fix it. Hopefully don't break the unraid in the process. But I figured posting here could give me ideas I wouldn't think of otherwise. I'll grab screenshots and maybe link to a time-lapse video of what's happening if that helps.
February 14, 20179 yr Author Buddy of mine was just telling me he was having similar weird issues as me with 6.2.4. I don't know how I didn't see that there is now a version 6.3.1 when I did this. Think I'm just used to sourceforge which has latest download at the bottom of the list. I made other mistakes last night with setup. So I'm going to go ahead and start from scratch tonight with 6.3.1 to see what happens.
February 15, 20179 yr Author Rebuilt my setup completely from scratchwith 6.3.1 and everything seems to be normal. Both VMs set to 20GB RAM and dashboard sits at a steady 89% used. Thanks for your help, everyone.
Archived
This topic is now archived and is closed to further replies.