Good day,
I don't know yet if this belongs here, on Github as a feature request, or if it should be discussed on Slack.
Babel is the future rather than OLSR, ok. But I've found Babel is extremely slow to allow a new node onto the mesh, and many changes to features on a node including LAN address changes disturbs an existing connection for some time including raising the load average for a while on the node it is connected to. Once a good connection is established I understand Babel works well, but when a connection is marginal Babel drops the connection, then if you reboot tries again, drops it, re-connects etc. Each time Babel makes any decision the result down the line is it seems to take a very long time to have the other node (and any associated services) come back and be available.
I'm wondering if there is any possibility of adding back some of the controls we had during OLSR with LQM? We could exclude a poor quality connection by putting in a minimum snr ... we could put in a maxiumum distance ... and we even could exclude a node from connecting if it was problematic by using the MAC address of the bad node. Now with current firmware, I have absolutely no control over the network and can only trust Babel to make all the decisions for me.
What I would like is to have the LQM controls back. I'd expect that with Babel routing I'd need them far less often than before, but just like having a self driving car, you still need a steering wheel and the ability to take over control.
Thanks, Ed K7EOK
I don't know yet if this belongs here, on Github as a feature request, or if it should be discussed on Slack.
Babel is the future rather than OLSR, ok. But I've found Babel is extremely slow to allow a new node onto the mesh, and many changes to features on a node including LAN address changes disturbs an existing connection for some time including raising the load average for a while on the node it is connected to. Once a good connection is established I understand Babel works well, but when a connection is marginal Babel drops the connection, then if you reboot tries again, drops it, re-connects etc. Each time Babel makes any decision the result down the line is it seems to take a very long time to have the other node (and any associated services) come back and be available.
I'm wondering if there is any possibility of adding back some of the controls we had during OLSR with LQM? We could exclude a poor quality connection by putting in a minimum snr ... we could put in a maxiumum distance ... and we even could exclude a node from connecting if it was problematic by using the MAC address of the bad node. Now with current firmware, I have absolutely no control over the network and can only trust Babel to make all the decisions for me.
What I would like is to have the LQM controls back. I'd expect that with Babel routing I'd need them far less often than before, but just like having a self driving car, you still need a steering wheel and the ability to take over control.
Thanks, Ed K7EOK
Orv W6BI
Orv