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

When shutting down VMs, they often get stuck at 50% CPU usage #4057

Open
andrewdavidwong opened this Issue Jul 5, 2018 · 3 comments

Comments

Projects
None yet
3 participants
@andrewdavidwong
Member

andrewdavidwong commented Jul 5, 2018

Qubes OS version:

R3.2

Affected component(s):

Using qvm-shutdown from the command-line and shutting down VMs using Qubes Manager


Steps to reproduce the behavior:

  1. Select a running VM in Qubes Manager.
  2. Press the shut down button.

or

  1. Use qvm-shutdown --wait to shut down the VM.

Expected behavior:

The VM shuts down normally.

Actual behavior:

Around 30-40% of the time, the VM's CPU usage (as shown in Qubes Manager) goes up to 50%. The laptop's fan speeds up (ThinkPad T450s), and the VM does not shut down for a long time. If you wait, it will eventually either shut down or get killed due to the timeout.

@donob4n

This comment has been minimized.

Show comment
Hide comment
@donob4n

donob4n Jul 5, 2018

Hi, I think this also happens in Qubes 4.0, I have seen it many of times but I didn't report since I didn't found a way for reproduce it.

In my case I tailed '/var/log/xen/console/guest-'vm-name'.log' and I saw systemd waiting for some services to stop but unfortunately I could not read the affected services (the output was pretty ugly).

donob4n commented Jul 5, 2018

Hi, I think this also happens in Qubes 4.0, I have seen it many of times but I didn't report since I didn't found a way for reproduce it.

In my case I tailed '/var/log/xen/console/guest-'vm-name'.log' and I saw systemd waiting for some services to stop but unfortunately I could not read the affected services (the output was pretty ugly).

@andrewdavidwong

This comment has been minimized.

Show comment
Hide comment
@andrewdavidwong

andrewdavidwong Jul 6, 2018

Member

@marmarek, let us know if there are any logs we should collect on 3.2 and 4.0 next time this happens.

Member

andrewdavidwong commented Jul 6, 2018

@marmarek, let us know if there are any logs we should collect on 3.2 and 4.0 next time this happens.

@mig5

This comment has been minimized.

Show comment
Hide comment
@mig5

mig5 Jul 16, 2018

I see this happening too pretty frequently.

Another way to connect to the shutting-down VM is using Xen's xl tool via dom0:

sudo xl console VMNAME

When I console to the VM that is taking a long time to shutdown, I see this shutdown job is taking a long time to finish:

[ **    ] A stop job is running for Session c3 of user user (1min 2/ 1min 30s)

As you can see there is a kill-switch timer there of 1.5 minutes. After this the machine shuts down. I don't know why this particular stop job takes so long, may not be Qubes-specific. Feels SystemD specific.

http://forums.debian.net/viewtopic.php?f=10&t=134311

I too am on a Thinkpad T450s. I think it has affected both Fedora and Debian VMs for me.

mig5 commented Jul 16, 2018

I see this happening too pretty frequently.

Another way to connect to the shutting-down VM is using Xen's xl tool via dom0:

sudo xl console VMNAME

When I console to the VM that is taking a long time to shutdown, I see this shutdown job is taking a long time to finish:

[ **    ] A stop job is running for Session c3 of user user (1min 2/ 1min 30s)

As you can see there is a kill-switch timer there of 1.5 minutes. After this the machine shuts down. I don't know why this particular stop job takes so long, may not be Qubes-specific. Feels SystemD specific.

http://forums.debian.net/viewtopic.php?f=10&t=134311

I too am on a Thinkpad T450s. I think it has affected both Fedora and Debian VMs for me.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment