On Mon, Oct 5, 2015 at 5:20 PM, Miroslav Lichvar <[email protected]> wrote: > But can we safely assume noone needs to keep time in the past, for > instance to workaround a bug in some unsupported software?
This would only limit being able to CHANGE the time to the past OFFLINE (initiated by the RTC). I don't see a actual scenario that would be affected. Ok, you want to run the time in the past. 1- You want several computers running with a "island" time. That is, you in fact want NTP. So you just change the time at the NTP server, and so the change is not initiated by the RTC. Not affected. 2- You just want your computer running a "custom" time in the past. What is the role of Chrony on that? You'll need to disable chrony or use a special config. Custom case where the default setting will never apply anyway. > And if we could add such check for the RTC, could we check also time > received from NTP servers and refuse to go in the past? I don't think so. At least that would be another option, not related to this issue. I mean, you could also have a configuration related to makestep to say "if the estimate is over 1 hour, do a step, if the estimate is over 1 day, do not try to correct because NTP time is unreliable.". But that looks to be that should be user-configurable, if done, while the RTC can be solved by default without side effects. Thanks, Nuno -- To unsubscribe email [email protected] with "unsubscribe" in the subject. For help email [email protected] with "help" in the subject. Trouble? Email [email protected].
