-
Notifications
You must be signed in to change notification settings - Fork 660
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[launch] add support for --bridged
network
#2074
Conversation
b4a5a39
to
0f727ec
Compare
Codecov Report
@@ Coverage Diff @@
## main #2074 +/- ##
==========================================
+ Coverage 81.45% 81.48% +0.02%
==========================================
Files 184 184
Lines 9442 9456 +14
==========================================
+ Hits 7691 7705 +14
Misses 1751 1751
Continue to review full report at Codecov.
|
0f727ec
to
f695d08
Compare
Thanks for this @Saviq! The name The code looks good, I only have a minor concern with respect to the functionality: what happens if the host has an interface named 'bridged'? I think we should provide a way to bridge it. I know this is unlikely, but it's still possible. One possibility I see to address this corner case is to allow |
That would require us to change the RPC and CLI ( |
Well, I don't think this is strictly necessary at that cost. It would be nice to have if easier to implement. Let's leave it like it is now and if some user complains at some point, we can think again about it. |
@LuisP I added the tweak to only restart in the daemon if the driver changed. Unfortunately I was unable to add a reliable test, but didn't want to block on it. |
3e6759e
to
1e5ff65
Compare
Looks good, thanks! |
4c17e28
to
47ebbdb
Compare
@luis4a0 I reverted the recent changes, went back to just adding the setting and |
bors r+ |
Build failed: |
bors retry |
Canceled. |
bors r=luis4a0,Saviq |
bors cancel |
Canceled. |
bors retry |
bors r=luis4a0,Saviq |
Build failed: |
bors retry |
Build failed: |
One last time. bors retry |
Build succeeded: |
For context: the other options we were considering (
--public
,--exposed
) would not be intuitive to anyone. Even if "bridging" isn't strictly correct, e.g. on Hyper-V, that is the term other hypervisors use and people that know what it is will likely know from the get go what it's about. Even ourselves we refer to the feature as "bridging", because that's the prevalent term.I decided against validating on
multipass set local.bridged-interface
as the given interface may not exist at that point in time for whatever reason, and onlaunch
we'll communicate well what the problem is, if any.