So as I’ve come to understand, without a battery-backed up RTC, there is a period on startup where the BB’s time reflects whatever is in the file /etc/timestamp before an NTP process can update it. This is reflected in timestamps in “/var/log/syslog” and elsewhere.
Several people have posted frustrations with inaccurate syslog times for unexpected reboots and other situations where this behavior makes life difficult. Several people have suggested various strategies for updating the /etc/timestamp file periodically and/or as part of a shutdown script, but the one is only as good as the update frequency and the other is only good if there is a normal shutdown.
Has anyone created a robust solution for this problem that has been proven out in practice that they can share? Is there any chance something like this could be made into a service or some kind of shared solution in GitHub or elsewhere?
Thanks in advance!