This PR adds back reboot detection in the poller, allowing it to check whether one of the devices rebooted and in this case it will rewrite all the records, this allows the user to see changes whenever a device reboots.
Additionally the PR tries to reduce the time drift of the bootup timestamp due to the fact that some of the devices return the uptime instead of a epoch timestamp, therefore we need to work out the timestamp. To mitigate this issue we:
Always use the response timestamp to work out the bootup timestamps
Always use seconds for the bootup timestamps
Get the epoch timestamp instead of uptime in junos, linux, cumulus and sonic
Type of change
Please delete options that are not relevant.
Bug fix (non-breaking change which fixes an issue)
New feature (non-breaking change which adds functionality)
Description
This PR adds back reboot detection in the poller, allowing it to check whether one of the devices rebooted and in this case it will rewrite all the records, this allows the user to see changes whenever a device reboots.
Additionally the PR tries to reduce the time drift of the bootup timestamp due to the fact that some of the devices return the uptime instead of a epoch timestamp, therefore we need to work out the timestamp. To mitigate this issue we:
Type of change
Please delete options that are not relevant.