Skip to content

v1.19 Backports 2026-01-26#44003

Merged
joestringer merged 12 commits intov1.19from
pr/v1.19-backport-2026-01-26-09-17
Jan 26, 2026
Merged

v1.19 Backports 2026-01-26#44003
joestringer merged 12 commits intov1.19from
pr/v1.19-backport-2026-01-26-09-17

Conversation

aanm and others added 8 commits January 26, 2026 09:17
[ upstream commit 0f58e5c ]

conformance-ipsec.yaml needs to be tested on a scheduled basis on stable
branches, therefore it should be added to this list since GitHub doesn't
trigger schedule events for non-default branches.

Signed-off-by: André Martins <andre@cilium.io>
Signed-off-by: Marco Hofstetter <marco.hofstetter@isovalent.com>
[ upstream commit 2a3bb63 ]

This shorter timeout seems to be too short for the tests to be
successful. We should increase them and see if the flakiness decreases.

Signed-off-by: André Martins <andre@cilium.io>
Signed-off-by: Marco Hofstetter <marco.hofstetter@isovalent.com>
[ upstream commit e014ca0 ]

Since introduction of base-SHA as required dispatch argument, triggering
that workflow started failing on branch 1.19 as we were not providing
it.

Signed-off-by: Marcel Zieba <marcel.zieba@isovalent.com>
Signed-off-by: Marco Hofstetter <marco.hofstetter@isovalent.com>
[ upstream commit 0c24f26 ]

Signed-off-by: Joe Stringer <joe@cilium.io>
Signed-off-by: Marco Hofstetter <marco.hofstetter@isovalent.com>
[ upstream commit 2c72678 ]

The CiliumEndpointSlice migration workflow is currently run on schedule
on the main branch. However, it is not being triggered for stable branches,
because it is not part of the test suite triggered via /test. Let's get
this divergence fixed adding it to the list of nightly workflows, so that
it gets periodically run on stable branches as well.

Reported-by: André Martins <andre@cilium.io>
Signed-off-by: Marco Iorio <marco.iorio@isovalent.com>
Signed-off-by: Marco Hofstetter <marco.hofstetter@isovalent.com>
….TCP)

[ upstream commit 3b0defb ]

When multiple oneof variants are specified in a field mask (e.g., both
TCP and UDP), the previous implementation would copy all variants,
creating spurious structures for inactive ones. This caused identical
flows to generate different aggregation keys.

Added oneof-aware copying using WhichOneof() to only copy the active
variant, preventing spurious structures and ensuring consistent keys.

Signed-off-by: mereta <mereta.degutyte@hotmail.co.uk>
Signed-off-by: Marco Hofstetter <marco.hofstetter@isovalent.com>
…ontext

[ upstream commit 4549ee8 ]

Aggregation only keeps and aggregates on the fields specified in the fieldAggregate.
There is no way to preserve a timestamp, and specifying time in fieldAggregate defeats aggregation.
The solution is to preserve the 1st occurring timestamp in the processedFlow after the aggregation key is generated.
The aggregation logic is not affected and temporal context is preserved.

Signed-off-by: mereta <mereta.degutyte@hotmail.co.uk>
Signed-off-by: Marco Hofstetter <marco.hofstetter@isovalent.com>
[ upstream commit 063e40c ]

Add support for configuring Hubble Relay log format and level through Helm values, similar to the existing logOptions support in cilium-agent and cilium-operator.

This allows users to enable JSON logging for Hubble Relay, which is useful for log aggregation tools in environments like AWS EKS.

New Helm values:
- hubble.relay.logOptions.format: text, text-ts, json, json-ts
- hubble.relay.logOptions.level: debug, info, warn, error

Fixes: #43009
Signed-off-by: Pavan More <pavansmore05@gmail.com>
Signed-off-by: Marco Hofstetter <marco.hofstetter@isovalent.com>
@mhofstetter mhofstetter added kind/backports This PR provides functionality previously merged into master. backport/1.19 This PR represents a backport for Cilium 1.19.x of a PR that was merged to main. labels Jan 26, 2026
@mhofstetter
Copy link
Copy Markdown
Member Author

/test

Copy link
Copy Markdown
Member

@giorio94 giorio94 left a comment

Choose a reason for hiding this comment

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

Thanks, my commit looks good.

@mhofstetter mhofstetter marked this pull request as ready for review January 26, 2026 09:02
@mhofstetter mhofstetter requested review from a team as code owners January 26, 2026 09:02
@mhofstetter mhofstetter force-pushed the pr/v1.19-backport-2026-01-26-09-17 branch from 3775810 to 3adf8c5 Compare January 26, 2026 10:38
@mhofstetter
Copy link
Copy Markdown
Member Author

mhofstetter commented Jan 26, 2026

@christarazi ci-l7 & ci-l3-l4 fail for patch releases in the drift checker

  🟥 check-log-errors/no-errors-in-logs:pkg/driftchecker:kind-kind/kube-system/cilium-7qtz9 (cilium-agent): Found 1 logs in kind-kind/kube-system/cilium-7qtz9 (cilium-agent) matching list of errors that must be investigated:
time=2026-01-26T09:58:44.756603282Z level=warn source=/go/src/github.com/cilium/cilium/pkg/driftchecker/checker.go:102 msg="Mismatch found" module=agent.controlplane.config-drift-checker key=mesh-auth-enabled actual=false expectedValue=true expectedSource=mesh-auth-enabled/cilium-config (1 occurrences)
    ⛑️ The following owners are responsible for reliability of the testsuite: 
        - @cilium/sig-foundations (no-errors-in-logs:pkg/driftchecker)
        - @cilium/ci-structure (.github/workflows/conformance-l3-l4.yaml)

i changed the underlying tests-e2e-upgrade.yaml to exclude this field in patch downgrades in the drift checker (--set-string=extraConfig.ignore-flags-drift-checker=mesh-auth-enabled) 🤞 (instead of excluding the check from the connectivity tests in main and having to await a CLI release)

@mhofstetter
Copy link
Copy Markdown
Member Author

/test

@mhofstetter mhofstetter force-pushed the pr/v1.19-backport-2026-01-26-09-17 branch from 3adf8c5 to 302bdbd Compare January 26, 2026 11:15
@mhofstetter
Copy link
Copy Markdown
Member Author

/test

[ upstream commit a4bcb54 ]

Due to the way the pkg/signal manager is implemented, it requires at
least one handler to be registered. Because the authmanager used to be a
default handler that was always registered, the code assumes there is
always one handler. Instead of reworking the signalmanager, set up a
no-op handler when auth is disabled. When the auth code is eventually
removed, the expected auth handler can be removed from the signalmanager
side.

Signed-off-by: Chris Tarazi <chris@isovalent.com>
Signed-off-by: Marco Hofstetter <marco.hofstetter@isovalent.com>
[ upstream commit c3b621c ]

This commit will update the status of a policy if it contains auth rules
when the feature is disabled, so that the user is warned about the fact
that the auth rule is ineffective.

Signed-off-by: Chris Tarazi <chris@isovalent.com>
Signed-off-by: Marco Hofstetter <marco.hofstetter@isovalent.com>
christarazi and others added 2 commits January 26, 2026 12:43
[ upstream commit 8ea193f ]

This feature is still in beta and has seen limited adoption. Its control
plane implementation has reliability problems. Although the
authentication feature is only active when explicitly set in a network
policy, the feature's control plane is enabled by default. This
default-on state causes it to interact with and disrupt other core
Cilium subsystems.

Therefore, disabling it by default for now.

Signed-off-by: Chris Tarazi <chris@isovalent.com>
Signed-off-by: Marco Hofstetter <marco.hofstetter@isovalent.com>
[ upstream commit 9927bda ]

The grep output being printed here added noise to the build process
without any actionable outcomes. Hide it by using "grep -q".

Fixes: 0b0cfc7 ("chore: Add a linter for non-idempotent secrets")
Signed-off-by: Joe Stringer <joe@cilium.io>
Signed-off-by: Marco Hofstetter <marco.hofstetter@isovalent.com>
@mhofstetter mhofstetter force-pushed the pr/v1.19-backport-2026-01-26-09-17 branch from 302bdbd to 5b2ee5d Compare January 26, 2026 11:43
@mhofstetter
Copy link
Copy Markdown
Member Author

/test

@mhofstetter
Copy link
Copy Markdown
Member Author

requesting review by @devodev & @Artyop for #43644 & #43924 - thanks for taking a quick look!

@mhofstetter mhofstetter requested review from Artyop and devodev January 26, 2026 12:27
Copy link
Copy Markdown
Contributor

@devodev devodev left a comment

Choose a reason for hiding this comment

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

Thank you! :)

Copy link
Copy Markdown
Contributor

@Artyop Artyop left a comment

Choose a reason for hiding this comment

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

✔️

Copy link
Copy Markdown
Member

@joestringer joestringer left a comment

Choose a reason for hiding this comment

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

I looked over backports for #43890, #42665, #43960. LGTM thanks!

@joestringer joestringer added this pull request to the merge queue Jan 26, 2026
Merged via the queue into v1.19 with commit 6cdf61c Jan 26, 2026
547 of 550 checks passed
@joestringer joestringer deleted the pr/v1.19-backport-2026-01-26-09-17 branch January 26, 2026 18:33
@cilium-release-bot cilium-release-bot bot moved this to Released in cilium v1.19.0 Feb 3, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

backport/1.19 This PR represents a backport for Cilium 1.19.x of a PR that was merged to main. kind/backports This PR provides functionality previously merged into master.

Projects

No open projects
Status: Released

Development

Successfully merging this pull request may close these issues.