You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Docker containers hosted on Windows have their clock drift by exactly the amount of time the Windows host spends in suspend mode. Upon resuming, all current and future docker containers created (until reboot) will have their clock stuck permanently in the past.
Steps to reproduce the issue:
Steps to reproduce:
1 Have Docker on Windows.
2 Check that containers are reading the same time as the system:
> date
Tue, Oct 25, 2016 1:06:05 AM
> docker run -it centos:centos7 date
Tue Oct 25 01:06:08 UTC 2016
3 Walk away. After 30 minutes, windows goes into suspend, and stays that way.
4 After N > 30 minutes, return and resume the machine.
5 Notice that the containers think that only 30 minutes have passed, for all values of N.
> date
Tue, Oct 25, 2016 3:21:42 AM
> docker run -it centos:centos7 date
Tue Oct 25 01:36:41 UTC 2016
Describe the results you received:
Container clocks have drifted by exactly the amount of time spent suspended.
Describe the results you expected:
Container clocks should not be significantly skewed by suspend/resume.
Output of docker version:
Client:
Version: 1.12.1
API version: 1.24
Go version: go1.6.3
Git commit: 23cf638
Built: Thu Aug 18 17:52:38 2016
OS/Arch: windows/amd64
Server:
Version: 1.12.1
API version: 1.24
Go version: go1.6.3
Git commit: 23cf638
Built: Thu Aug 18 17:52:38 2016
OS/Arch: linux/amd64
Description
Docker containers hosted on Windows have their clock drift by exactly the amount of time the Windows host spends in suspend mode. Upon resuming, all current and future docker containers created (until reboot) will have their clock stuck permanently in the past.
Steps to reproduce the issue:
Steps to reproduce:
1 Have Docker on Windows.
2 Check that containers are reading the same time as the system:
3 Walk away. After 30 minutes, windows goes into suspend, and stays that way.
4 After N > 30 minutes, return and resume the machine.
5 Notice that the containers think that only 30 minutes have passed, for all values of N.
Describe the results you received:
Container clocks have drifted by exactly the amount of time spent suspended.
Describe the results you expected:
Container clocks should not be significantly skewed by suspend/resume.
Output of
docker version
:Output of
docker info
:The System Time, in the above, is behind the true system time by about 2 hours.
Additional environment details (AWS, VirtualBox, physical, etc.):
Windows 10 Pro 64-bit Version 1511 OS Build 10586.633
The text was updated successfully, but these errors were encountered: