-
Notifications
You must be signed in to change notification settings - Fork 290
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Vmmem eating up CPU after last update even without running any docker image #8742
Comments
I can confirm that I am also experiencing this problem. I am running Windows 10 Pro and using Docker Desktop 2.4.0.0 (48506) For me, the high Vmmem CPU usage seems to be tied to running compilation of a large C++ code base using parallel make/ninja. |
I did some more digging into this and I have come up with an acceptable workaround. I wonder if the root causes it that Docker Desktop 2.4.0 uses slightly more memory than before, causing the system to start paging to this earlier than it did for 2.3. These are my observations:
According to https://docs.microsoft.com/en-us/windows/wsl/wsl-config the default available memory to wsl2 is 80% of physical RAM. On my 16 GB machine this amounts to 12GB, which was enough for the system to start swapping with quite light usage of application on the host system (e.g. Visual Studio Code, Edge, some terminal windows). Limiting the amount to 10GB removed the problem for me altogether yesterday. I could successfully run the large compilation mentioned above without vmmem using all CPU. |
Can confirm that editing wsl-config to enforce global resource limits helped stabilise memory and CPU usage. Worth keeping in mind just how much memory all this stuff takes up though. On my system, after a fresh reboot and starting up Docker Desktop, vmmem usage sits around 1.5GB. +1GB for K8S. +1GB if I open up a new distro (Ubuntu 20.04). If the limits are set too low, the VM will start swapping again. |
Disabling the Kubernetes cluster fixed this for me btw, so there gotta be an issue with that part(?). |
Same issue here. Config: windows 10 20H2 (latest) and docker 3.1 (latest) |
https://medium.com/@lewwybogus/how-to-stop-wsl2-from-hogging-all-your-ram-with-docker-d7846b9c5b37 helped me |
In the past we could configure limits with .wslconfig but now my computer ignore this seetings. Microsoft Windows 10 Enterprise (10.0.19041 N/A Build 19041) |
Same issue. Windows Version: 20H2 (OS Build 19042.804) |
Same issue here. WSL already consumes like 12 out of my 16 GB RAM after startup, although not a single container is running. Windows Version: 20H2 (OS Build 19042.928) |
I have the same issue. 100% CPUT and 80% RAM is used by Vmmem. |
for me it is suddenly fixed, I am not sure why. I think it might be because
I updated ubuntu running in WSL which was not updated for a long time.
Maybe it helps
…On Thu, 22 Apr 2021 at 08:42, Marcin-TA ***@***.***> wrote:
I have the same issue. 100% CPUT and 80% RAM is used by Vmmem.
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
<#8742 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ABRZQ2GTJEBUOFE2B5JFPIDTJ7AONANCNFSM4SAFWJSQ>
.
|
Same here, 98% RAM (15.4GB) and 100%CPU used by Vmmem, with no containers running. |
@Gabriellavoura You can specify the max RAM Docker WSL2 can use thanks to the .wslconfig file.
|
Thanks @onomatopellan, I have already done that, just posted here to say that this problem wasn't fixed yet. |
Same issue here. VMMem process is using 2-3 GB of RAM, despite no running containers. Kubernetes is not enabled. Docker Desktop: 3.3.3 (64133) Setting memory limits for WSL is not a solution in my opinion, as it will limit the number of containers I'm able to run concurrently. |
Same for me. Even after quitting docker-desktop (Right click - quit docker desktop) vmmem does not exit. Also, even tho Docker is not in boot, vmmem still spins up. Manually doing |
Same for me. Edition Windows 10 Pro |
even with this, vmmem sometimes uses ~50%+ cpu and a couple of gigs of ram. |
I noticed this issue happens right after I create a docker image with |
"docker will simplify the development" ... they said. Now my pc sounds like a jet engine. Also, spent ton of time figuring why that happens, to also find that i can't stop it (not at least without much overhead). |
This work-around fixed it for me as well. Would be nice if we could get a fix for this. |
An alternative workaround that worked for me was to disable WSL 2 and use Hyper V. Docker may not be as fast but at least resources are much more reasonable. |
Same issue, even though non of the containers are working, vmmem eats 16.5gb of ram. |
Same issue here. Not a single container ever created, it will still eat all of the RAM. |
This comment has been minimized.
This comment has been minimized.
This is a good solution but unfortunately not available to users with Windows 10 Home. |
@danmaina the Hyper-V solution is the only one that worked for me as well (tried wslconfig and all that.) It is really maddening that this issue seems to get largely ignored. |
Relevant (but maybe not exactly the same) issue over at wsl microsoft/WSL#6982 (comment) |
same problem, docker is consuming a tremendous amount of CPU =( |
same :( |
I noticed that version 4 of docker is super slower on win10 :( |
This is happening to me running docker in a nested vm on a 5950x on windows 11, the VM is windows 10 |
Hi same here, Docker is pretty useless on Windows (run fine on Ubuntu though) |
Hi Im having the same high CPU usage problem. |
Any update on this? I'm experiencing the same issue... |
Same here: after Windows startup, I start Docker Desktop, "Vmmem" memory usage goes to 2GB without any container running. I'm sitting here looking at Task Manager and "Vmmem" memory usage keeps going up all by itself. It has to be a memory leak. Windows 10 Pro 21H2 build 19044.1466 .wslconfig |
Same issue for me, high memory consumptions even when no containers are running |
Same here in windows 11. Docker not running at all but CPU 100% + RAM consuming. I opened docker and closed it and fixed. |
I have the same problem. I don't have WSL set up but instead use the Hyper-V version. Docker Desktop is not running. Launching it shuts down the Hyper-V machine and then starts it again, resolving the issue. There's also tons of error messages in the console but I wasn't able to capture those yet. I only remember that it was something about |
after fresh reboot (docker desktop autostarting, kubernetes active, nothing else active on the machine) -> CPU Fan goes wild try #1: stopping Kuberntes, docker desktop still active -> no change --> wsl -- shutdown and restarting docker-desktop is a workaround (and only this). |
Issues go stale after 90 days of inactivity. Prevent issues from auto-closing with an If this issue is safe to close now please do so. Send feedback to Docker Community Slack channels #docker-for-mac or #docker-for-windows. |
Sad to see this issue go 'stale' with no resolution. |
/remove-lifecycle stale |
Still very much an issue! |
For me, after updates it is ok. |
Hi all! sorry to hear you're having problems. From the history of this thread, it sounds like there are several different issues related to CPU/RAM usage being reported here -- some are fixed and some might not be. The original issue has a workaround that is described further up-thread (editing the wsl-config - #8742 (comment)). Some of the issues are tracked elsewhere in this issue tracker. Because the original issue has a workaround, I'm going to close this. It's hard for us to inform people when your individual issue is fixed if we're lumping all resource utilization problems into a single issue. If you believe your problem is still there, please file a new issue with specific repro steps and a diagnostic ID. |
Closed issues are locked after 30 days of inactivity. If you have found a problem that seems similar to this, please open a new issue. Send feedback to Docker Community Slack channels #docker-for-mac or #docker-for-windows. |
Expected behavior
No Vmmem high CPU usage
Actual behavior
High CPU usage for Vmmem
Information
Steps to reproduce the behavior
The text was updated successfully, but these errors were encountered: