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

v1.15 backports 2024-02-14 #30767

Merged

Conversation

michi-covalent
Copy link
Contributor

Once this PR is merged, a GitHub action will update the labels of these PRs:

30151 30561 30759 30757 30762

pchaigno and others added 7 commits February 14, 2024 16:59
[ upstream commit c19a84e ]

This error can happen if a state is being destroyed while packets are in
flight. It should be rare as the window in the kernel where it can
happen is very short.

Signed-off-by: Paul Chaignon <paul.chaignon@gmail.com>
Signed-off-by: Michi Mutsuzaki <michi@isovalent.com>
[ upstream commit 51b4308 ]

Moved BGP rest APIs from daemon package to bgpv2 sub-system. This allows
us to remove references of BGP control plane from daemon.

Signed-off-by: harsimran pabla <hpabla@isovalent.com>
Signed-off-by: Michi Mutsuzaki <michi@isovalent.com>
[ upstream commit 3c479d4 ]

The test output are riddled with logs such as:

    Defaulted container "cilium-agent" out of: cilium-agent, config
    (init), mount-cgroup (init), apply-sysctl-overwrites (init),
    mount-bpf-fs (init), clean-cilium-state (init),
    install-cni-binaries (init)

This gets particularly noisy when waiting for the key rotation to
complete, during which time we run kubectl exec repeatedly. This commit
fixes it.

Signed-off-by: Paul Chaignon <paul.chaignon@gmail.com>
Signed-off-by: Michi Mutsuzaki <michi@isovalent.com>
[ upstream commit afb3eb6 ]

If we construct a `nodeAddressing` object with the IPv4 flag, and then
call `getDirectRouting` while the directRoutingDevice doesn't contain
an IPv4 address, it would return a nil address instead of the failure
condition. This can cause a panic in the caller who assumes that the
IP is never nil.

Signed-off-by: Dylan Reimerink <dylan.reimerink@isovalent.com>
Signed-off-by: Michi Mutsuzaki <michi@isovalent.com>
[ upstream commit b8f6976 ]

Issue cilium#30563 showed that veth devices were picked as selected
native devices even though they did not have a route. This was
due to a bug in the route StateDB table indexing. This commit
expands the devices controller tests to catch this wrong
behavior.

Signed-off-by: Jussi Maki <jussi@isovalent.com>
Signed-off-by: Michi Mutsuzaki <michi@isovalent.com>
[ upstream commit 0844200 ]

The RouteID.Key() was supposed to make a composite key from
'table + index + destination', but due to a copy-paste
brainfart the composite index was 'table + table + destination'.

Fix Key() to correctly build the key. While at it, make it more
efficient by pre-allocating the correct fixed-size byte slice.

Fixes: cilium#30563
Fixes: 8abf620 ("devices_controller: Switched devices and routes from StateDB to StateDB2")
Signed-off-by: Jussi Maki <jussi@isovalent.com>
Signed-off-by: Michi Mutsuzaki <michi@isovalent.com>
[ upstream commit 0e67360 ]

We are failing a e2e test quite consistantly, while the root cause is
not fully clear, it seems that setting devices explicitly might help.

Signed-off-by: Dylan Reimerink <dylan.reimerink@isovalent.com>
Signed-off-by: Michi Mutsuzaki <michi@isovalent.com>
@michi-covalent michi-covalent requested review from a team as code owners February 14, 2024 17:01
@michi-covalent michi-covalent added kind/backports This PR provides functionality previously merged into master. backport/1.15 This PR represents a backport for Cilium 1.15.x of a PR that was merged to main. labels Feb 14, 2024
@michi-covalent
Copy link
Contributor Author

/test-backport1.15

@michi-covalent
Copy link
Contributor Author

/test-backport-1.15

Copy link
Contributor

@brlbil brlbil left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

CI changes LGTM

@michi-covalent
Copy link
Contributor Author

running ci-e2e in #30771

@michi-covalent
Copy link
Contributor Author

✅ ran ci-e2e from a non-fork PR to test the workflow file change: https://github.com/cilium/cilium/actions/runs/7907721456. time to 🎸

@michi-covalent michi-covalent merged commit c9e2a4d into cilium:v1.15 Feb 14, 2024
115 of 117 checks passed
@michi-covalent michi-covalent deleted the pr/v1.15-backport-2024-02-14 branch February 14, 2024 22:11
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
backport/1.15 This PR represents a backport for Cilium 1.15.x of a PR that was merged to main. kind/backports This PR provides functionality previously merged into master.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

7 participants