-
Notifications
You must be signed in to change notification settings - Fork 939
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
QEMU: Occupy vsock Context ID through syscall #11993
QEMU: Occupy vsock Context ID through syscall #11993
Conversation
c5072cf
to
d5401c3
Compare
Signed-off-by: Julian Pelizäus <[email protected]>
d5401c3
to
769c415
Compare
769c415
to
e01abe9
Compare
780b4d4
to
e30f707
Compare
e30f707
to
5b7a53b
Compare
5b7a53b
to
b666277
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM, thanks!
Please can you do a full suite of VM tests, so VMs running inside different nested containers, VMs running inside nested VMs etc to check for any regressions.
I have created a small script for testing those scenarios, the outcome is what we expect. Gist: https://gist.github.com/roosterfish/ea62fe1f7fff2838eb786d4444901c1c |
@roosterfish you could open a PR on https://github.com/canonical/lxd-ci put it in a new "tests" directory please |
b666277
to
5d81e22
Compare
@roosterfish did your manual tests pass OK ? |
When trying to acquire a new vsock Context ID, we want to occupy it right away to prevent collisions. This helps for scenarios where two VMs have the same identical volatile.vsock_id (VM export/import) and in case LXD is used in a nested environment. If either the acquisition or the syscall fails, the next one gets selected. Signed-off-by: Julian Pelizäus <[email protected]>
5d81e22
to
293f36f
Compare
Yes, all clear. |
@roosterfish thanks for your thoroughness on picking up these issues and fixing them. |
When trying to acquire a new vsock Context ID, we want to occupy it right away to prevent collisions.
This helps for scenarios where two VMs have the same identical volatile.vsock_id (VM export/import) and
in case LXD is used in a nested environment.
If either the acquisition or the syscall fails, the next one gets selected.
Follow up on #11896 and observed during investigation of #11907