[main][Automation] Update elastic/beats to 1dc4d7a6380e #9952
Conversation
|
Pinging @elastic/elastic-agent-control-plane (Team:Elastic-Agent-Control-Plane) |
|
This pull request does not have a backport label. Could you fix it @github-actions[bot]? 🙏
|
|
I didn't investigate each failure exhaustingly, but in quick review the test I see fail are related to liveness probing/monitoring: === FAIL: testing/integration/ess TestMonitoringPreserveTextConfig/TestMonitoringLiveness (2.14s)
...
=== FAIL: testing/integration/ess TestMonitoringPreserveTextConfig (57.85s)
...
=== FAIL: testing/integration/ess TestMonitoringLivenessReloadable/TestMonitoringLiveness (2.12s)
...
=== FAIL: testing/integration/ess TestMonitoringLivenessReloadable (53.82s)@nkvoll could these failures relate to the recent liveness changes made in #9673? |
0fa4321 to
0cd3684
Compare
|
These failures are related to the Looking at the JSON test logs, the test fails at {"Time":"2025-09-16T12:59:32.168290271Z","Action":"output","Package":"github.com/elastic/elastic-agent/testing/integration/ess","Test":"TestMonitoringLivenessReloadable/TestMonitoringLiveness","Output":" \t \tGet \"http://localhost:6792/liveness\": dial tcp 127.0.0.1:6792: connect: connection refused\n"}Looking at the agent diagnostics the monitoring server changes 3 times with the change to {"log.level":"info","@timestamp":"2025-09-16T12:59:09.248Z","log.logger":"api","log.origin":{"function":"github.com/elastic/elastic-agent-libs/api.(*Server).Start.func1","file.name":"api/server.go","file.line":87},"message":"Metrics endpoint listening on: 127.0.0.1:6791 (configured: http://localhost:6791)","log":{"source":"elastic-agent"},"ecs.version":"1.6.0"}
{"log.level":"info","@timestamp":"2025-09-16T12:59:31.394Z","log.logger":"api","log.origin":{"function":"github.com/elastic/elastic-agent-libs/api.(*Server).Start.func1","file.name":"api/server.go","file.line":87},"message":"Metrics endpoint listening on: 127.0.0.1:6791 (configured: http://localhost:6791)","log":{"source":"elastic-agent"},"ecs.version":"1.6.0"}
{"log.level":"info","@timestamp":"2025-09-16T12:59:33.669Z","log.logger":"api","log.origin":{"function":"github.com/elastic/elastic-agent-libs/api.(*Server).Start.func1","file.name":"api/server.go","file.line":87},"message":"Metrics endpoint listening on: 127.0.0.1:6792 (configured: http://localhost:6792)","log":{"source":"elastic-agent"},"ecs.version":"1.6.0"}The way the test is written seems inherently racy, it makes a policy change via the override API, then checks for components to be healthy (which is not an indicator the policy change happened), then makes a single HTTP request an expects it to succeed. The simplest fix is probably just to add some retries, a slightly better fix would be to poll for the policy revision to change or even the log line indicating the metrics server port change happened. |
|
I linked to the elastic-agent/testing/integration/ess/monitoring_probe_reload_test.go Lines 131 to 140 in 3e1fc2b |
|
I've narrowed down the cause of the [UPDATE 1] The change I made in 8525a53 should get this test to pass again. [UPDATE 2] I didn't take a closer look but I suspect something in stretchr/testify#1427, which was included in |
ce94411 to
8525a53
Compare
Made with ❤️️ by updatecli
Made with ❤️️ by updatecli
Made with ❤️️ by updatecli
7d2dd09 to
2845ec9
Compare
Made with ❤️️ by updatecli
* Add rollback field to UpgradeRequest * Introduce rollback parameter to upgrade * Concurrently retry taking over watcher * Gracefully shutdown agent watcher * Add rollbacks available to upgrade marker * disable rollback window by default * Add formal checks to manual rollback arguments * Add minimum version check for creating rollbacks entries in update marker * Gracefully terminate watcher process on windows * Allow watcher to listen to signals only during watch loop * make watcher rollback only if the agent has not been already rolled back * Remove parent death signal for watcher on linux * Distinguish between upgrade and rollback operations in upgrade subcommand * remove DESIRED_OUTCOME in favor of watch --rollback * Add version agent rollbacks to in manual rollback reason * Check upgrade details state before allowing a manual rollback
…tion CFT environment (#10007) * Have FIPS integration tests spin up deployments in Production CFT environment * Add explanatory comment * Run extended tests if FIPS integration tests pipeline changes * Revert "Run extended tests if FIPS integration tests pipeline changes" This reverts commit ae0e89d.
* Fix monitoring reloading tests Ensure we verify the policy update was actually applied before querying the monitoring endpoint. Add a retry on said request as well. * Use the explicit policy revision returned by Kibana
129623d to
183b202
Compare
|
Closing in favor of #10019, there were several conflicts in this branch with the fixes on main and my attempt to force rebase the fixes out failed. |
|
What
Update
elastic/beatsto the latest version on branchmain.Changeset
Bump beats
Update to elastic/beats@1dc4d7a6380e
ran shell command ".ci/scripts/update-beats.sh 1dc4d7a6380e"
Created automatically by Updatecli
Options:
Most of Updatecli configuration is done via its manifest(s).
Feel free to report any issues at github.com/updatecli/updatecli.
If you find this tool useful, do not hesitate to star our GitHub repository as a sign of appreciation, and/or to tell us directly on our chat!