Skip to content
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

networking: Add hack to kick new bonds etc to be fully awake #8769

Conversation

mvollmer
Copy link
Member

@mvollmer mvollmer commented Mar 5, 2018

This is expected to fix the failures uncovered by the new testBondActive test in #8736. Those tests are included and extended here.

@mvollmer
Copy link
Member Author

mvollmer commented Mar 5, 2018

This is a hackish workaround, and we should hear from the NM people what they think of this.

@mvollmer mvollmer force-pushed the networking-explicitly-activate-new-bonds branch from 973dd51 to 93f987f Compare March 5, 2018 09:30
@mvollmer
Copy link
Member Author

mvollmer commented Mar 7, 2018

This is a hackish workaround, and we should hear from the NM people what they think of this.

There is no clear message yet: https://bugzilla.redhat.com/show_bug.cgi?id=1548265

I'll see if I can tie up the loose ends here and make this merge-able.

@mvollmer mvollmer force-pushed the networking-explicitly-activate-new-bonds branch 2 times, most recently from 4b73c33 to dbbabf0 Compare March 7, 2018 13:04
@mvollmer mvollmer changed the title networking: Add hack to explicitly activate new connections that have autoconnect set networking: Add hack to kick new bonds etc to be fully active Mar 7, 2018
@mvollmer mvollmer changed the title networking: Add hack to kick new bonds etc to be fully active networking: Add hack to kick new bonds etc to be fully awake Mar 7, 2018
@mvollmer
Copy link
Member Author

mvollmer commented Mar 7, 2018

Okay, I added tests also for bridge and teams, which suffer from the same regression, and (hopefully) made the workaround work and not break anything.

@mvollmer mvollmer removed the blocked label Mar 7, 2018
NetworkManager 1.8.2 has a regression here, so let's test this
explicitly.

testBondingMain also hits the regression, but because the active
interface has "manual" ipv4 settings, and Cockpit copies those to the
bond, the bond ends up looking okay although it doesn't really work.

testBond also hits the regression, but because it has one inactive
interface, the bond ends up being functional (with just one slave).

Fixes cockpit-project#8735
@mvollmer mvollmer force-pushed the networking-explicitly-activate-new-bonds branch from dbbabf0 to 51d2429 Compare March 7, 2018 14:21
@mvollmer
Copy link
Member Author

Let's not merge this just yet, it's possible that NM wants to revert to the old behavior.

@mvollmer
Copy link
Member Author

There is a proposed test for NM, so I assume we don't need the workaround.

NetworkManager/NetworkManager-ci#158

@mvollmer mvollmer closed this Mar 23, 2018
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

1 participant