Recovery behavior when gateway node is completely unavailable - Githubissues

westbywest / robin-mesh-v2

Derivative of the ROBIN Mesh Firmware http://wiki-robin.meshroot.com/

4 stars 19 forks source link

Recovery behavior when gateway node is completely unavailable #10

Open westbywest opened 11 years ago

westbywest commented 11 years ago

Options for when a gateway node is completely unavailable:

If no other stations detected on mesh, periodic restart wireless interface and/or /etc/init.d/networking. (Did have to restart wifi on TP-Link MR3020 after 5days uptime due to silent wifi driver failure.) This option does at least preserve active DHCP leases.
If other stations detected on mesh but no gateway node, periodic restart of OLSRd?
Or, periodic reboot of node?

Period for any such restart process should be quite long to avoid oscillation, e.g. 30min or 1hr.