Join GitHub today
GitHub is home to over 28 million developers working together to host and review code, manage projects, and build software together.Sign up
Time synchronisation appears to fail silently on Alpha 1590.1.0 #2277
Container Linux Version
Cluster of 5 AWS EC2 m4.xlarge instances.
On Stable release
The timestamps of the last successful time synchronisation on
On Alpha release
However, the timestamps on
We also see strange time-related errors in our apps on the alpha cluster, where publish processes can complete before they've started with a negative duration, depending on what node of the cluster the apps are running on, and depending on that node's time drift.
If it's useful, I'm happy to share the app logs as well.
Please note that this seems to be case regardless of whether
Upgrading from Stable > Alpha seems to reproduce this issue. Beta doesn't appear to have this problem right now.
The 1590 releases are using systemd v235, which uses
Thanks @dm0- - that explains why the timestamp for last successful sync no longer updates on
I'll speak to our devs who raised the issue, and ask them to drop some more info here with exactly what's going wrong with time synchronisation.