bastl
-
Posts
1,267 -
Joined
-
Last visited
-
Days Won
3
Content Type
Profiles
Forums
Downloads
Store
Gallery
Bug Reports
Documentation
Landing
Posts posted by bastl
-
-
This isn't a specific Ryzen issue. It's more like a Win7 issue thats around for ages. It usually only happens during the first setup of Win7. After the install and the first boot you should be able to shutdown the VM and add more cores without issues. What you can try is to try to use the machine type Q35 instead. If you are planing to passthrough devices like GPU audio mouse keyboard etc, first set it up without any devices.
-
@danyg I noticed something strange in my docker config. I had 2 paths configured pointed to the same source. The config path (called host path in my old docker) and the override path both pointed to "/mnt/user/appdata/netdata". I guess this caused the issue overriding the files. I completly removed the docker and the config and redownloaded the docker image. Only one path left with the new config and the files i create stay in place. I can't remember ever touch the config before. The container I use for almost 2 years now. Maybe with an update the paths changed and in my config instead of replacing the old declaration the update added a second path.
BUT at first I couldn't find any sensor informations after adding your tweaks. After double and triple checking the config files I think I found the issue. A single space in front of the "sensors: yes" in the python.d.conf caused the issue to not showing up the sensor informations in NetData. 😣
On 9/16/2019 at 9:16 AM, bastl said:(remove the # at the beginning of that line)
AND the space between '#' and 'sensors' 😂
-
@Jaster Try to restart the whole server and try again. In some rare cases if the driver in the VM crashes the GPU won't reset properly unless you restart the whole system.
-
@Rick Sanchez Check the xml of the VM at the bottom are all the passed through USB devices listed. It should be looking like this
<hostdev mode='subsystem' type='usb' managed='no'> <source> <vendor id='0x04fc'/> <product id='0x0003'/> </source> <address type='usb' bus='0' port='1'/> </hostdev> <hostdev mode='subsystem' type='usb' managed='no'> <source> <vendor id='0x0a12'/> <product id='0x0001'/> </source> <address type='usb' bus='0' port='2'/> </hostdev>
In my case one is the usb mouse and the the second one is the bluetooth dongle for my keyboard. Compare it to the normal config view and check if all the devices also listed there.
If you for some reason disconnect the passed through USB device before starting up the VM you get the exact message you posted above.
-
@Jaster What GPU are you using? AMD?
-
From all that I've read so far about the x570 boards is, that in most cases the IOMMU groups are horrible. Tons of stuff are grouped together and can't be split and a couple things under linux aren't stable yet. Even Wendel reported it in one of his videos that this platform isn't suggested yet for virtualisation. Not sure if newer BIOS/Agesa versions released the last weeks already fixed passthrough issues. I remember the first gen Threadrippers had nearly the same problems. 3 months or so after release the plattform became stable. Until that I had the same issues, random crashes, freezes, GPU passthrough not fully functional, terrible IOMMU groupings. Fingers crossed that your issues get solved soon.
-
@frodr I use the following user script running once a day someone posted here on the forum. Can't find the source right now. It saves the xml and the NVRAM files in a zip file in the path you have to set at the top. Backups older than 30 days gets cleaned out with every run.
#!/bin/bash #backs up #change the location below to your backup location backuplocation="/mnt/user/Backup/vm_settings/" # do not alter below this line timestamp=`date '+%Y-%m-%d__%H-%M-%S'` dir="$backuplocation"vm_settings/"$timestamp" # dont change anything below here if [ ! -d $dir ] ; then echo "making folder for todays date $timestamp" # make the directory as it doesnt exist mkdir -vp $dir else echo "As $dir exists continuing." fi echo "Saving vm xml files" rsync -a --no-o /etc/libvirt/qemu/*xml $dir/xml/ echo "Saving ovmf nvram" rsync -a --no-o /etc/libvirt/qemu/nvram/* $dir/nvram/ chmod -R 777 $dir echo "information: creating ZIP-archive and cleaning up old data" zip -r -9 "$backuplocation"vm_settings/"$timestamp".zip "$backuplocation"vm_settings/"$timestamp" && rm -r "$backuplocation"vm_settings/"$timestamp" # Amendmend to the original script of danioj by Deeks 2017 # Delete all backups older than 2 days ONLY if newer files exist # Two remarks for quick and easy testing # 1. Line 148 : Set parameter --> actually_copy_files="0" # 2. Lines 1068 & line 1070 : Use parameter -mmin in stead of -mtime to test with minutes rather than days # Credits for this code to JBRELAND on https://tinyurl.com/ycrrbobv echo "information: cleaning out backups older than 30 days in location ONLY if newer files exist" $backuplocation/ for j in $backuplocation/ do if [[ -n $(find "$j" -type f -mtime -29) ]]; then #echo "debug: valid criteria to enter loop" find "$j" -type f -mtime +30 -exec rm -f {} \; fi #echo "debug: ended if sequence after loop" done sleep 5 exit -
Your favorite movie or series of all time??? 😂
-
1
-
-
@Sleepyme What you can try, use another slot for the card and also have a display plugged in while setting it up. For BIOS updates AMD released the newest AGESA version couple days ago, which should fix a couple things. Not all motherboard vendors released a newer version based on this yet. You might have to check again.
-
@drawde Same for me. Did you tried what danyg posted above? For me the created files mentioned in the tutorial are emptied as soon as I startup the container and no effect. No stats for fans or temps. 🤨
-
@Sleepyme Did you played around with different BIOS versions for your motherboard?
-
Quick question, is the model of the drive the same as the one you RMAed? Same day of purchase, same order?
-
Just a question, are you even allowed to use the internet for private stuff on your work machine? Depending on which industry you're working, I can understand why the admins wanna prevent their user to accessing personal stuff with portable browsers. One single click on the wrong page or an personal email and the desaster is there.
-
Ok. If it's a machine connected to a domain your profil with all it's settings might sync in the backround from a company server. Usually with every logoff or reboot this profil should sync back to the server and save any changes you made. Not actual sure if it's possible to setup by the admin in a way that login credentials are only allowed to add and change and disallowed to remove. I know for fileshares it's possible to set it up this way. The user is allowed to ad files, change existing files but aren't allowed to remove them. Sounds stupid, i know but there might be usecases for this. Not actual sure if it's possible for the credential store. Another possibility is that your companies firewall/proxy configuration caches all sorts of cookies and session information and kinda Man-In-the-Middle checks all the traffic from the clients. Lots of enterprise equipment still intercepts the connections to analyse for bad traffic and often breaks logins and webservices on non company networks.
Have you tried to use a portable version of a browser like firefox? No need to install, everything is stored in a single folder, let's say on the desktop. If you don't need it anymore, delete it, an everything is gone. Sure this isn't the solution for your main problem to delete the stored credentials but maybe helps to prevent future issues.
-
Are you using a Windows user account or a local account? The Microsoft account syncs all kind of stuff. I wouldn't be surprised if it syncs all the login credentials you maybe have saved on an old installation. Just an idea.
Best solution might be to stop using the Internet Explorer 😂 Btw. does it still get's updates? I think I've read something that MS will stop the support for it with the ongoin development and implementation of Edge.
-
If the card has UEFI support this might work.
-
Change the boot order in this part from 1 to 2
<disk type='block' device='disk'> <driver name='qemu' type='raw' cache='writeback'/> <source dev='/dev/sdd1'/> <backingStore/> <target dev='hdc' bus='sata'/> <boot order='1'/> <alias name='sata0-0-2'/> <address type='drive' controller='0' bus='0' target='0' unit='2'/> </disk>
and than search for the right entry for your NVME controller. Under tools > system devices search for your NVME controller and which bus/slot/function is shown.
41:00.0 in my example:
IOMMU group 45: [144d:a804] 41:00.0 Non-Volatile memory controller: Samsung Electronics Co Ltd NVMe SSD Controller SM961/PM961Now search your xml file for that number (source adress)
<source> <address domain='0x0000' bus='0x41' slot='0x00' function='0x0'/> </source>
and add the boot order tag as shown below from my setting.
<hostdev mode='subsystem' type='pci' managed='yes'> <driver name='vfio'/> <source> <address domain='0x0000' bus='0x41' slot='0x00' function='0x0'/> </source> <boot order='1'/> <alias name='hostdev3'/> <address type='pci' domain='0x0000' bus='0x07' slot='0x00' function='0x0'/> </hostdev>
The bus, slot and function number will be different in your case. Make sure you get the correct one from your listed system devices. If you change the VM settings in the UI you might loose this custom edit. Keep that in mind.
Another hint, try to avoid using core 0 and it's HT core for a VM. Unraid will always use this cores for it's own to handle stuff in the background. This can cause some hickups inside the VM if Unraid is doin stuff.
-
1
-
1
-
-
21 hours ago, danyg said:
HOW TO TEMPERATURES VOLTAGES FANS
- Go to http://unraid:19999/netdata.conf (change unraid for your unraid servername or ip, my case is called unraid)
- Paste the contents in a text editor
- Find the [plugins] section
- Change the line where says # python.d = yes for python.d = yes (so remove the # at the beginning of that line)
- Copy the content of your text editor and paste the content in a new file called netdata.conf in /mnt/user/appdata/netdata (or wherever you have configured the data of your unraid)
- Now copy and paste the content of this: https://raw.githubusercontent.com/netdata/netdata/master/collectors/python.d.plugin/python.d.conf in a text editor
- Find the line # sensors: yes and change it for sensors: yes (remove the # at the beginning of that line)
- Save this file as python.d.conf in /mnt/user/appdata/netdata
- Create a folder python.d in /mnt/user/appdata/netdata
- Copy the contents of https://raw.githubusercontent.com/netdata/netdata/master/collectors/python.d.plugin/sensors/sensors.conf in a text editor
- Check the lines that starts with types: and add a # at the beginning of each sensor that you DON'T want to have in netdata
- Save this file as sensors.conf in /mnt/user/appdata/netdata/python.d
- Restart the Docker of netdata, and you will see a new section called Sensors and there all your temperatures and voltages and more
If after all of this, still not works, I recommend to install Dynamix System Temperature and configure it properly, you should be able to see temperatures in the footer of unraid admin web, as that tool configures the command sensors to work as expected and therefore the plugin for netdata can read those sensors too.
Hope this Helps
First of all, thanks for the tutorial. I've noticed today that the sensors section is gone for me in netdata. It's the first time this happened since I use that docker. That section was there as default since I can remember using this docker, without doin anything special with adding plugins.
I followed your tutorial created all the files, with the changes you mentioned and as soon as I start the container the files become empty. Do I have to change something else? Any idea?
-
Maybe your system specs and what you already tried to make it work would be a good starting point. Without that people can't really help you.
-
15 hours ago, fluisterben said:
What do you mean by that?
With "sitting on the cache drive" I mean depending on how the share is setup, the vdisk file is stored on ether the cache, the array or an unassigned device. It only can sit on one underlying source and not mixed/split like file shares. It's a single file. With "Prefer cache" as soon as the cache gets full or reaches the a certain threshold new writes will go directly to the array. BUT this is only true for single files writen to the share, not files writen inside a vdisk file. Let's say your cache has 10GB free and you transfer a file with 20GB to a share, it will be directly written to the array. If let's say you write that file inside the VM to your vdisk, the vdisk get's bigger until the cache drive get's full and the copy process will either pause, abort or the VM will crash completly. There is no way for unraid to transfer a part of the vdisk to the array to keep the cache free for a couple GB. Unraid sees the vdisk as a single file and can't split it like "oh thats some files on that vdisk the user not often use, let's put that on the array and keep the rest on the fast cache". That's not how it works.
15 hours ago, fluisterben said:Isn't each write within the VM routed to cache either way?
No it's not. If the vdisk is on an unassigned device or on the array, the writes within that file will directly writen to the underlying storage, where you basically see the raw or close to raw performance of that storage device.
15 hours ago, fluisterben said:Why would it perform better for the entire image to be stored there? Only at start/boot and at shutdown that would save you a second or something, not?
Depending on the usecase of a VM you should store the vdisk in most scenarios on the fastest storage possible. Not even starting up the VM benefits from it, also starting programs and working with them heavily see a positive affect of the underlying storage.
Setup a VM with a vdisk on the array or a spinning drive and compare the performance with a VM vdisk on a ssd/nvme device and you will see huge huge performance differences in starting/using and working within that VM.
-
I checked my setup where I never used IE before. The default path for the cache is "C:\Users\username\AppData\Local\Microsoft\Windows\INetCache\". Did you tried the option in the settings to remove the cache and history by closing the IE? There is an extra Option for "allow caches and databases for websites" which is ticked by default. Sry, german UI but it looks like the following:
Maybe there are some other places storing logins and cookies.
-
@Fireball3 I know from Firefox that a stored cookie is enough for the unraid forums to be auto logged in. Have you tried to clean the cache and cookie stuff in IE? Long time ago that I've used the IE to tell you where you can find it.
-
First google search came up with the following:
QuoteAdd, remove, edit, backup, restore Stored User Names and Passwords in Windows 10
The Stored User Names and Passwords in Windows lets you securely manage user names and passwords as a part of your profile. It lets you automatically enter saved user names and passwords for various network resources, servers, websites and applications, to authenticate yourself. In this post we will see how to add, remove, edit, backup, restore Stored User Names and Passwords & Credentials in Windows 10/8/7.
rundll32.exe keymgr.dll,KRShowKeyMgrhttps://www.thewindowsclub.com/stored-user-names-and-passwords-windows
-
"/mnt/user/domains/vmname/vdisk1.img" is the default and I don't know why you should change that. What is your main problem? Full cache drive? How big is the cache? Having a VM with an vdisk size bigger than the cache size sooner or later will end up with problems and propably data loss. You know that, right?
For performance reasons it's prefered to have the VM vdisks sit on the cache drive if it's a SSD/NVME or on an unassigned drive. On default the "domain" share is sitting on the cache drive. So fine so far. It's on the user and the specific configuration how he set's up the VM. If you have a small cache SSD, let's say 128GB and you use it for caching your shares you should know how many data you put on your cache before the mover kicks in to transfer data to the array or of the share is cache only. You can overprovision a vdisk so for the VM itself it looks like a 128GB disk, where it's actually maybe only uses 30GB space. As soon as you fill up the VM's vdisk from inside the VM you more and more use space on the cache drive up to the point where used space by shares on cache + docker + vmdisk reaches the max amount possible even if Windows thinks 50 still free inside the VM and your VM will pause or crash. You should be aware of this.
Terrible Perfomance on threadripper
in VM Engine (KVM)
Posted
If you passthrough an NVME the best choice is to passthrough the entire controller and not the by-id. Check my config at the bottom you can find an entry "boot order 1". This is the controller for the NVME. Also using SCSI for the other SSD can give you slightly lower latency and less overhead. I think there is also a SpaceInvader video about how to passthrough an NVME controller. Also you can see in my configs a couple tweaks especialy for TR4 where the CPU is reported as Epyc to make the OS aware of the real amount of L2 and L3 cache of the CPU. Another note, you have set the emulatorpin to cores from node0. I've got better results having it set to the node the cores i'am giving to the VM.