how are people handling clock drift?

[nlf]

just wondering how other people are handling clock drift in their vms when the host machine is put to sleep and wakes up later. currently i'm just restarting ntpd in a cron job, but it's pretty hacky. any better solutions?

[jakajancar]

Well, if you're using ntpd, you don't have to restart it, just add the following to /etc/ntpd.conf:

tinker panic 0

But I was hoping using ntpd would not even be needed when I reported this.

[xez]

Not a workaround but the reason this happens is that xhyve's timers stop when the host goes to sleep. When it wakes up again it doesn't account for the time sleeping and the timers keep ticking from their old value. To fix this xhyve would have to subscribe to sleep/wake notifications. I'll take a look when I find some time.

[bfreis]

Any updates on this issue? It's very annoying for Docker on Mac!

Thanks!