[00:10] folks - trying to use `netplan set` to set a default route for an interface. I can't make heads or tails out of how I'd translate the -to: and via: lines of the yaml into the format that set wants. I don't see any examples in the `test_cli_get_set.py` that cover routes or similar structures (or ANY `set` examples in the "examples"...). can someone point me to a clue? [08:35] keen: reading the parsing code, I *think* something along those lines should work: `netplan set 'ethernets.eth0.routes=[{"to":"1.2.3.4/24", "via":"56.78.90.12"}]'`. Basically, add your routes as a JSON array. [08:36] keen: however, there's a caveat: if you already have routes configured, the code will ignore your new routes if the JSON array size has the same number of elements than the one in the current definition. [08:36] I'm guessing that's a bug :) [13:32] appreciated! apparently the lack of single quotes was killing me. that's exactly where I landed as "this must work" but apparently ignored the fact that the quotes inside the flat json would be ignored. [13:32] and...fun bug. [13:33] . o O ( well, consumed, and not make it to the parser ) [13:33] `netplan set 'ethernets.eth1.routes=[{"to":"default", "via":"10.200.1.1"}]' [13:33] ` [13:33] works! [13:45] Great to hear :) [13:52] bridge.interfaces, otoh...it seems to validate the interfaces existance inside bridges, not inside ethernets. [13:52] at least for this one I can just create the yaml raw (I dont need to update it) [13:53] bridges..interfaces anyway. and there seem to be a few random conditions that dont work there [14:01] keen: that seems odd. Do you have a reproducer for your bridge interface existence issue? [14:30] yeah I'll put a pastebin together when I get a chance (probably after my next call) [16:28] schopin: https://gist.github.com/keen99/fd740872c84a8ba35b7f1c7c716e80f7