feat: support building integrations in arm by mdelapenya · Pull Request #25195 · elastic/beats

mdelapenya · 2021-04-21T09:52:00Z

What does this PR do?

It extracts the logic to build&push integrations' docker images using mage to a method in the pipeline, that will be used to be run in parallel: one branch for amd and one for arm. The result is having the integrations built in ARM format.

Besides that, we are adding a new mage command manifesttool:pushSupportedVersions for pushing the integration images with multi-platform support. We are using a tool provided by Infra to perform that task, which we wrapped into the new mage command. This mage command:

will be run only on linux machines, because is coupled with the CI workers. That could be changed for new images.
it will calculate the Docker tag used on the compose:buildSupportedVersions reading the values from the supported-versions.yml file for each integration module. It will add logic to build the proper tag based in the ${MODULE}_VARIANT and ${MODULE}_VERSION, and for the specific use case of ceph we are adding an ${MODULE}_INDEX to control the custom image number we use internally for tracking changes.

Why is it important?

The integrations are built for AMD only, and the e2e testing framework needs to consume them in ARM to provide a full experience testing in ARM.

Checklist

My code follows the style guidelines of this project
I have commented my code, particularly in hard-to-understand areas
I have made corresponding changes to the documentation
I have made corresponding change to the default configuration files
I have added tests that prove my fix is effective or that my feature works
I have added an entry in CHANGELOG.next.asciidoc or CHANGELOG-developer.next.asciidoc.

Author's Checklist

Do we need to use an agent in the first stage? Instead, I think we can move that agent to the checkout stage

Related issues

Relates [Integrations] Docker images are only built for AMD, not for ARM #25194

Other considerations

Maybe this approach could be discarded if we realise that maintaining ARM/AMD images for the integrations is overkill. It could be the case where we have to build and maintain the binaries for each integration in both platforms, very likely because there is no ARM flavour of the image. Because the existence of ARM images for the integrations is blocking elastic/e2e-testing#707, maybe we can get to the point of not running e2e tests for the integrations.

elasticmachine · 2021-04-21T10:05:22Z

💚 Build Succeeded

the below badges are clickable and redirect to their specific view in the CI or DOCS

Expand to view the summary

Build stats

Reason: null
Start Time: 2022-02-03T10:19:02.712+0000
Duration: 151 min 10 sec
Commit: 62518f5

Test stats 🧪

Test	Results
Failed	0
Passed	47607
Skipped	4281
Total	51888

💚 Flaky test report

Tests succeeded.

🤖 GitHub comments

To re-run your PR in the CI, just comment with:

/test : Re-trigger the build.
/package : Generate the packages and run the E2E tests.
/beats-tester : Run the installation tests with beats-tester.
run elasticsearch-ci/docs : Re-trigger the docs validation. (use unformatted text in the comment!)

This index will be used by the manifest-tool to generate the proper tag

This target is intended to be run on the CI, only

See elastic/beats#25195

dev-tools/mage/target/manifesttool/manifesttool.go

jsoriano · 2021-04-27T11:16:06Z

.ci/build-docker-images.groovy

+            parallelTasks["arm_metricbeat"] = generateBuildStep(platform: "arm", beatPath: 'metricbeat')
+            parallelTasks["ubuntu_20_metricbeat"] = generateBuildStep(platform: "ubuntu-20", beatPath: 'metricbeat')


Are arm and ubuntu-20 jenkins labels? It looks weird to me to see an OS and an architecture used for the same platform parameter.

Yes, they are Jenkins labels

jsoriano · 2021-04-27T11:56:00Z

dev-tools/mage/target/manifesttool/manifesttool.go

+			case moduleUppercase + "_CODENAME":
+			case moduleUppercase + "_VARIANT":
+				codename = v
+			case moduleUppercase + "_INDEX":


This is a significative change, not sure if we want to do it this way.

So far, for good or bad, the variables set in each variant don't have any special meaning, they are just applied when running docker-compose subcommands. Currently we know that when we want to do something with all the variants, we have to call docker-compose once for each variant, passing the variables defined there.

With this change we are forcing a convention that may affect what images are used and which ones are cached. For example we have images with index different to 1 (postgresql, nats, prometheus...). Should we also define a _INDEX variable apart of defining it in the docker-compose.yml file? This can be error-prone.

If this is done to cover the ceph images, I would prefer to rethink how ceph is tested, or even reduce its coverage, before introducing this logic here.

In see though that this system complicates the obtention of the tags. Maybe another approach for that is to parse the docker-compose.yml files to get the image fields for each service, and replace the variables there for each one of the variants. Imitating this logic of docker-compose can be a bit hacky, but it is a well known format, so it can be still clean. And this way we could continue having any variable in these files without expecting any special meaning.

* feat: add support for running the tests in ARM * fix: arm workers are not immutable yet * chore: calculate artifact from the GOARCH variable At this moment we support amd64 and arm64 * chore: run elasticsearch in the right platform * chore: add log for architecture * chore: print env variable for GOARCH * chore: set GOARCH based in the worker * fix: wrong copy&paste * chore: install k8s dependencies on ARM * chore: install godog using GOARCH * fix: proper values for GO env in Makefile * chore: add log when requesting a node * fix: define Groovy variable * chore: use proper platform for kibana * chore: simplify platforms selection logic * fix: use proper CI label for node selector * chore: remove running containers in clean-docker script * chore: clean up docker state on mutable workers * chore: use parallel branch name for filebeat logs * fix: add forgotten double quotes * chore: print Docker compose version On ARM * chore: use aarch64 architecture for RPM packages * chore: support ARM in tar installer * fix: proper arch for TAR installer * chore: log when the artifact is found * chore: be defensive about no running containers * chore: add tests for RPM artifact name * chore: add tests for x86_64 TAR artifact * chore: remove docker label from CI workers * chore: separate ARM tests in a second parallel stage This pretends to avoid using too many workers, adding a second layer for ARM. * chore: update stage name * fix: update arm target platform * chore: use GOARCH in the junit file * chore: use GOARCH in filebeat logs * chore: run all stages in parallel * chore: add platform to log when skipping tests * fix: use a filebeat version that supports multiplatform * chore: do not run integrations on ARM See elastic/beats#25195 * chore: move test resources for ARM * chore: use staging environment for package-registry See https://github.com/elastic/package-storage * fix: remove clean because arm workers are now immutable * chore: use new multiplatform images * chore: use new multiplatform base images for fleet-server * chore: add platform to elastic-agent cloud flavour * fix: use proper variable for platform * chore: do not clean up installer's workspace Because we are using ephemeral workers, we do not want to remove previous files * Revert "chore: do not clean up installer's workspace" This reverts commit 9104501. * chore(ci): skip pull of images from code * Revert "chore(ci): skip pull of images from code" This reverts commit 01cc441. * chore: set platform when pulling images * chore: update packer cache with our systemd images * fix: request proper platform image for the agent * feat: exclude scenarios per platform * chore: remove tags for endpoint, as they are not used at all They are also blocking the addition of a parent tag at the feature level, as maximun number of tags is 3 * fix: skip endpoint integration on ARM64 * feat: provide supported platforms at the scenario level on CI The scenario will be able to override its parent suite's platforms, making possible to select different platforms at the scenario level. * fix: use proper length method for groovy arrays * fix: groovy arrays use size * fix: override suite platforms per scenario * chore: update Jenkins label for arm workers Co-authored-by: Victor Martinez <victormartinezrubio@gmail.com> * chore: simplify platforms, reusing stack * chore: explicitly pass GOARCH to CI script * Revert "chore: update Jenkins label for arm workers" This reverts commit 9ef81ca. Co-authored-by: Victor Martinez <victormartinezrubio@gmail.com>

* feat: add support for running the tests in ARM * fix: arm workers are not immutable yet * chore: calculate artifact from the GOARCH variable At this moment we support amd64 and arm64 * chore: run elasticsearch in the right platform * chore: add log for architecture * chore: print env variable for GOARCH * chore: set GOARCH based in the worker * fix: wrong copy&paste * chore: install k8s dependencies on ARM * chore: install godog using GOARCH * fix: proper values for GO env in Makefile * chore: add log when requesting a node * fix: define Groovy variable * chore: use proper platform for kibana * chore: simplify platforms selection logic * fix: use proper CI label for node selector * chore: remove running containers in clean-docker script * chore: clean up docker state on mutable workers * chore: use parallel branch name for filebeat logs * fix: add forgotten double quotes * chore: print Docker compose version On ARM * chore: use aarch64 architecture for RPM packages * chore: support ARM in tar installer * fix: proper arch for TAR installer * chore: log when the artifact is found * chore: be defensive about no running containers * chore: add tests for RPM artifact name * chore: add tests for x86_64 TAR artifact * chore: remove docker label from CI workers * chore: separate ARM tests in a second parallel stage This pretends to avoid using too many workers, adding a second layer for ARM. * chore: update stage name * fix: update arm target platform * chore: use GOARCH in the junit file * chore: use GOARCH in filebeat logs * chore: run all stages in parallel * chore: add platform to log when skipping tests * fix: use a filebeat version that supports multiplatform * chore: do not run integrations on ARM See elastic/beats#25195 * chore: move test resources for ARM * chore: use staging environment for package-registry See https://github.com/elastic/package-storage * fix: remove clean because arm workers are now immutable * chore: use new multiplatform images * chore: use new multiplatform base images for fleet-server * chore: add platform to elastic-agent cloud flavour * fix: use proper variable for platform * chore: do not clean up installer's workspace Because we are using ephemeral workers, we do not want to remove previous files * Revert "chore: do not clean up installer's workspace" This reverts commit 9104501. * chore(ci): skip pull of images from code * Revert "chore(ci): skip pull of images from code" This reverts commit 01cc441. * chore: set platform when pulling images * chore: update packer cache with our systemd images * fix: request proper platform image for the agent * feat: exclude scenarios per platform * chore: remove tags for endpoint, as they are not used at all They are also blocking the addition of a parent tag at the feature level, as maximun number of tags is 3 * fix: skip endpoint integration on ARM64 * feat: provide supported platforms at the scenario level on CI The scenario will be able to override its parent suite's platforms, making possible to select different platforms at the scenario level. * fix: use proper length method for groovy arrays * fix: groovy arrays use size * fix: override suite platforms per scenario * chore: update Jenkins label for arm workers Co-authored-by: Victor Martinez <victormartinezrubio@gmail.com> * chore: simplify platforms, reusing stack * chore: explicitly pass GOARCH to CI script * Revert "chore: update Jenkins label for arm workers" This reverts commit 9ef81ca. Co-authored-by: Victor Martinez <victormartinezrubio@gmail.com> (cherry picked from commit e1a97f3) # Conflicts: # .ci/packer_cache.sh # cli/config/compose/profiles/fleet/docker-compose.yml # cli/config/compose/profiles/metricbeat/docker-compose.yml # e2e/_suites/fleet/stand-alone.go # internal/deploy/docker_client.go # internal/installer/elasticagent_deb.go # internal/installer/elasticagent_docker.go # internal/installer/elasticagent_rpm.go # internal/installer/elasticagent_tar.go # internal/utils/utils_test.go

* feat: add support for running the tests in ARM * fix: arm workers are not immutable yet * chore: calculate artifact from the GOARCH variable At this moment we support amd64 and arm64 * chore: run elasticsearch in the right platform * chore: add log for architecture * chore: print env variable for GOARCH * chore: set GOARCH based in the worker * fix: wrong copy&paste * chore: install k8s dependencies on ARM * chore: install godog using GOARCH * fix: proper values for GO env in Makefile * chore: add log when requesting a node * fix: define Groovy variable * chore: use proper platform for kibana * chore: simplify platforms selection logic * fix: use proper CI label for node selector * chore: remove running containers in clean-docker script * chore: clean up docker state on mutable workers * chore: use parallel branch name for filebeat logs * fix: add forgotten double quotes * chore: print Docker compose version On ARM * chore: use aarch64 architecture for RPM packages * chore: support ARM in tar installer * fix: proper arch for TAR installer * chore: log when the artifact is found * chore: be defensive about no running containers * chore: add tests for RPM artifact name * chore: add tests for x86_64 TAR artifact * chore: remove docker label from CI workers * chore: separate ARM tests in a second parallel stage This pretends to avoid using too many workers, adding a second layer for ARM. * chore: update stage name * fix: update arm target platform * chore: use GOARCH in the junit file * chore: use GOARCH in filebeat logs * chore: run all stages in parallel * chore: add platform to log when skipping tests * fix: use a filebeat version that supports multiplatform * chore: do not run integrations on ARM See elastic/beats#25195 * chore: move test resources for ARM * chore: use staging environment for package-registry See https://github.com/elastic/package-storage * fix: remove clean because arm workers are now immutable * chore: use new multiplatform images * chore: use new multiplatform base images for fleet-server * chore: add platform to elastic-agent cloud flavour * fix: use proper variable for platform * chore: do not clean up installer's workspace Because we are using ephemeral workers, we do not want to remove previous files * Revert "chore: do not clean up installer's workspace" This reverts commit 9104501. * chore(ci): skip pull of images from code * Revert "chore(ci): skip pull of images from code" This reverts commit 01cc441. * chore: set platform when pulling images * chore: update packer cache with our systemd images * fix: request proper platform image for the agent * feat: exclude scenarios per platform * chore: remove tags for endpoint, as they are not used at all They are also blocking the addition of a parent tag at the feature level, as maximun number of tags is 3 * fix: skip endpoint integration on ARM64 * feat: provide supported platforms at the scenario level on CI The scenario will be able to override its parent suite's platforms, making possible to select different platforms at the scenario level. * fix: use proper length method for groovy arrays * fix: groovy arrays use size * fix: override suite platforms per scenario * chore: update Jenkins label for arm workers Co-authored-by: Victor Martinez <victormartinezrubio@gmail.com> * chore: simplify platforms, reusing stack * chore: explicitly pass GOARCH to CI script * Revert "chore: update Jenkins label for arm workers" This reverts commit 9ef81ca. Co-authored-by: Victor Martinez <victormartinezrubio@gmail.com> (cherry picked from commit e1a97f3) # Conflicts: # cli/config/compose/profiles/fleet/docker-compose.yml # cli/config/compose/profiles/metricbeat/docker-compose.yml # e2e/_suites/fleet/stand-alone.go # internal/installer/elasticagent_deb.go # internal/installer/elasticagent_docker.go # internal/installer/elasticagent_rpm.go # internal/installer/elasticagent_tar.go # internal/utils/utils_test.go

* feat: add support for running the tests in ARM * fix: arm workers are not immutable yet * chore: calculate artifact from the GOARCH variable At this moment we support amd64 and arm64 * chore: run elasticsearch in the right platform * chore: add log for architecture * chore: print env variable for GOARCH * chore: set GOARCH based in the worker * fix: wrong copy&paste * chore: install k8s dependencies on ARM * chore: install godog using GOARCH * fix: proper values for GO env in Makefile * chore: add log when requesting a node * fix: define Groovy variable * chore: use proper platform for kibana * chore: simplify platforms selection logic * fix: use proper CI label for node selector * chore: remove running containers in clean-docker script * chore: clean up docker state on mutable workers * chore: use parallel branch name for filebeat logs * fix: add forgotten double quotes * chore: print Docker compose version On ARM * chore: use aarch64 architecture for RPM packages * chore: support ARM in tar installer * fix: proper arch for TAR installer * chore: log when the artifact is found * chore: be defensive about no running containers * chore: add tests for RPM artifact name * chore: add tests for x86_64 TAR artifact * chore: remove docker label from CI workers * chore: separate ARM tests in a second parallel stage This pretends to avoid using too many workers, adding a second layer for ARM. * chore: update stage name * fix: update arm target platform * chore: use GOARCH in the junit file * chore: use GOARCH in filebeat logs * chore: run all stages in parallel * chore: add platform to log when skipping tests * fix: use a filebeat version that supports multiplatform * chore: do not run integrations on ARM See elastic/beats#25195 * chore: move test resources for ARM * chore: use staging environment for package-registry See https://github.com/elastic/package-storage * fix: remove clean because arm workers are now immutable * chore: use new multiplatform images * chore: use new multiplatform base images for fleet-server * chore: add platform to elastic-agent cloud flavour * fix: use proper variable for platform * chore: do not clean up installer's workspace Because we are using ephemeral workers, we do not want to remove previous files * Revert "chore: do not clean up installer's workspace" This reverts commit 9104501. * chore(ci): skip pull of images from code * Revert "chore(ci): skip pull of images from code" This reverts commit 01cc441. * chore: set platform when pulling images * chore: update packer cache with our systemd images * fix: request proper platform image for the agent * feat: exclude scenarios per platform * chore: remove tags for endpoint, as they are not used at all They are also blocking the addition of a parent tag at the feature level, as maximun number of tags is 3 * fix: skip endpoint integration on ARM64 * feat: provide supported platforms at the scenario level on CI The scenario will be able to override its parent suite's platforms, making possible to select different platforms at the scenario level. * fix: use proper length method for groovy arrays * fix: groovy arrays use size * fix: override suite platforms per scenario * chore: update Jenkins label for arm workers Co-authored-by: Victor Martinez <victormartinezrubio@gmail.com> * chore: simplify platforms, reusing stack * chore: explicitly pass GOARCH to CI script * Revert "chore: update Jenkins label for arm workers" This reverts commit 9ef81ca. Co-authored-by: Victor Martinez <victormartinezrubio@gmail.com>

…1223) * feat: add support for running the tests in ARM (#707) * feat: add support for running the tests in ARM * fix: arm workers are not immutable yet * chore: calculate artifact from the GOARCH variable At this moment we support amd64 and arm64 * chore: run elasticsearch in the right platform * chore: add log for architecture * chore: print env variable for GOARCH * chore: set GOARCH based in the worker * fix: wrong copy&paste * chore: install k8s dependencies on ARM * chore: install godog using GOARCH * fix: proper values for GO env in Makefile * chore: add log when requesting a node * fix: define Groovy variable * chore: use proper platform for kibana * chore: simplify platforms selection logic * fix: use proper CI label for node selector * chore: remove running containers in clean-docker script * chore: clean up docker state on mutable workers * chore: use parallel branch name for filebeat logs * fix: add forgotten double quotes * chore: print Docker compose version On ARM * chore: use aarch64 architecture for RPM packages * chore: support ARM in tar installer * fix: proper arch for TAR installer * chore: log when the artifact is found * chore: be defensive about no running containers * chore: add tests for RPM artifact name * chore: add tests for x86_64 TAR artifact * chore: remove docker label from CI workers * chore: separate ARM tests in a second parallel stage This pretends to avoid using too many workers, adding a second layer for ARM. * chore: update stage name * fix: update arm target platform * chore: use GOARCH in the junit file * chore: use GOARCH in filebeat logs * chore: run all stages in parallel * chore: add platform to log when skipping tests * fix: use a filebeat version that supports multiplatform * chore: do not run integrations on ARM See elastic/beats#25195 * chore: move test resources for ARM * chore: use staging environment for package-registry See https://github.com/elastic/package-storage * fix: remove clean because arm workers are now immutable * chore: use new multiplatform images * chore: use new multiplatform base images for fleet-server * chore: add platform to elastic-agent cloud flavour * fix: use proper variable for platform * chore: do not clean up installer's workspace Because we are using ephemeral workers, we do not want to remove previous files * Revert "chore: do not clean up installer's workspace" This reverts commit 9104501. * chore(ci): skip pull of images from code * Revert "chore(ci): skip pull of images from code" This reverts commit 01cc441. * chore: set platform when pulling images * chore: update packer cache with our systemd images * fix: request proper platform image for the agent * feat: exclude scenarios per platform * chore: remove tags for endpoint, as they are not used at all They are also blocking the addition of a parent tag at the feature level, as maximun number of tags is 3 * fix: skip endpoint integration on ARM64 * feat: provide supported platforms at the scenario level on CI The scenario will be able to override its parent suite's platforms, making possible to select different platforms at the scenario level. * fix: use proper length method for groovy arrays * fix: groovy arrays use size * fix: override suite platforms per scenario * chore: update Jenkins label for arm workers Co-authored-by: Victor Martinez <victormartinezrubio@gmail.com> * chore: simplify platforms, reusing stack * chore: explicitly pass GOARCH to CI script * Revert "chore: update Jenkins label for arm workers" This reverts commit 9ef81ca. Co-authored-by: Victor Martinez <victormartinezrubio@gmail.com> * fix: remove packer_cache from maintenance branches Co-authored-by: Victor Martinez <victormartinezrubio@gmail.com>

#1224) * feat: add support for running the tests in ARM (#707) * feat: add support for running the tests in ARM * fix: arm workers are not immutable yet * chore: calculate artifact from the GOARCH variable At this moment we support amd64 and arm64 * chore: run elasticsearch in the right platform * chore: add log for architecture * chore: print env variable for GOARCH * chore: set GOARCH based in the worker * fix: wrong copy&paste * chore: install k8s dependencies on ARM * chore: install godog using GOARCH * fix: proper values for GO env in Makefile * chore: add log when requesting a node * fix: define Groovy variable * chore: use proper platform for kibana * chore: simplify platforms selection logic * fix: use proper CI label for node selector * chore: remove running containers in clean-docker script * chore: clean up docker state on mutable workers * chore: use parallel branch name for filebeat logs * fix: add forgotten double quotes * chore: print Docker compose version On ARM * chore: use aarch64 architecture for RPM packages * chore: support ARM in tar installer * fix: proper arch for TAR installer * chore: log when the artifact is found * chore: be defensive about no running containers * chore: add tests for RPM artifact name * chore: add tests for x86_64 TAR artifact * chore: remove docker label from CI workers * chore: separate ARM tests in a second parallel stage This pretends to avoid using too many workers, adding a second layer for ARM. * chore: update stage name * fix: update arm target platform * chore: use GOARCH in the junit file * chore: use GOARCH in filebeat logs * chore: run all stages in parallel * chore: add platform to log when skipping tests * fix: use a filebeat version that supports multiplatform * chore: do not run integrations on ARM See elastic/beats#25195 * chore: move test resources for ARM * chore: use staging environment for package-registry See https://github.com/elastic/package-storage * fix: remove clean because arm workers are now immutable * chore: use new multiplatform images * chore: use new multiplatform base images for fleet-server * chore: add platform to elastic-agent cloud flavour * fix: use proper variable for platform * chore: do not clean up installer's workspace Because we are using ephemeral workers, we do not want to remove previous files * Revert "chore: do not clean up installer's workspace" This reverts commit 9104501. * chore(ci): skip pull of images from code * Revert "chore(ci): skip pull of images from code" This reverts commit 01cc441. * chore: set platform when pulling images * chore: update packer cache with our systemd images * fix: request proper platform image for the agent * feat: exclude scenarios per platform * chore: remove tags for endpoint, as they are not used at all They are also blocking the addition of a parent tag at the feature level, as maximun number of tags is 3 * fix: skip endpoint integration on ARM64 * feat: provide supported platforms at the scenario level on CI The scenario will be able to override its parent suite's platforms, making possible to select different platforms at the scenario level. * fix: use proper length method for groovy arrays * fix: groovy arrays use size * fix: override suite platforms per scenario * chore: update Jenkins label for arm workers Co-authored-by: Victor Martinez <victormartinezrubio@gmail.com> * chore: simplify platforms, reusing stack * chore: explicitly pass GOARCH to CI script * Revert "chore: update Jenkins label for arm workers" This reverts commit 9ef81ca. Co-authored-by: Victor Martinez <victormartinezrubio@gmail.com> * chore: remove packer_cache from maintenance branch Co-authored-by: Victor Martinez <victormartinezrubio@gmail.com>

mergify · 2021-09-22T12:51:56Z

This pull request does not have a backport label. Could you fix it @mdelapenya? 🙏
To fixup this pull request, you need to add the backport labels for the needed
branches, such as:

backport-v./d./d./d is the label to automatically backport to the 7./d branch. /d is the digit

NOTE: backport-skip has been added to this pull request.

mergify · 2022-03-16T01:12:13Z

This pull request is now in conflicts. Could you fix it? 🙏
To fixup this pull request, you can check out it locally. See documentation: https://help.github.com/articles/checking-out-pull-requests-locally/

git fetch upstream
git checkout -b build-integrations-multi-platform upstream/build-integrations-multi-platform
git merge upstream/main
git push upstream build-integrations-multi-platform

mdelapenya · 2022-03-16T06:38:38Z

With the move to integration packages, I'd say this PR is not needed anymore, so we can close it

feat: support building integrations in arm

2cc16d8

mdelapenya self-assigned this Apr 21, 2021

mdelapenya added the Team:Automation Label for the Observability productivity team label Apr 21, 2021

mdelapenya requested a review from a team April 21, 2021 09:52

botelastic bot added needs_team Indicates that the issue/PR needs a Team:* label and removed needs_team Indicates that the issue/PR needs a Team:* label labels Apr 21, 2021

cachedout approved these changes Apr 21, 2021

View reviewed changes

mdelapenya added 5 commits April 23, 2021 20:20

feat: add a mage target that pushes images using multi-platform manifest

e2def0d

chore: import new mage namespace into filebeat and metricbeat

b01b030

chore: treat ceph differently about custom patch version

abb5a8e

This index will be used by the manifest-tool to generate the proper tag

chore: push images only on linux

9df4f13

This target is intended to be run on the CI, only

chore: use new mage goal in the CI

5970349

mdelapenya requested a review from jsoriano April 23, 2021 18:23

fix: use proper format after lint

d2efca0

mdelapenya added a commit to mdelapenya/e2e-testing that referenced this pull request Apr 23, 2021

chore: do not run integrations on ARM

7142ab8

See elastic/beats#25195

cachedout approved these changes Apr 26, 2021

View reviewed changes

v1v approved these changes Apr 26, 2021

View reviewed changes

mdelapenya commented Apr 26, 2021

View reviewed changes

dev-tools/mage/target/manifesttool/manifesttool.go Outdated Show resolved Hide resolved

fix: mount the entire docker config dir

62518f5

jsoriano reviewed Apr 27, 2021

View reviewed changes

mergify bot added the backport-skip Skip notification from the automated backport with mergify label Sep 22, 2021

jlind23 closed this Mar 31, 2022

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

feat: support building integrations in arm#25195

feat: support building integrations in arm#25195
mdelapenya wants to merge 8 commits intoelastic:mainfrom
mdelapenya:build-integrations-multi-platform

mdelapenya commented Apr 21, 2021 •

edited

Loading

Uh oh!

elasticmachine commented Apr 21, 2021 •

edited by jenkins-beats-ci bot

Loading

Build stats

Test stats 🧪

Uh oh!

Uh oh!

jsoriano Apr 27, 2021

Uh oh!

mdelapenya Apr 28, 2021

Uh oh!

jsoriano Apr 27, 2021

Uh oh!

mergify bot commented Sep 22, 2021

Uh oh!

mergify bot commented Mar 16, 2022

Uh oh!

mdelapenya commented Mar 16, 2022

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

		parallelTasks["arm_metricbeat"] = generateBuildStep(platform: "arm", beatPath: 'metricbeat')
		parallelTasks["ubuntu_20_metricbeat"] = generateBuildStep(platform: "ubuntu-20", beatPath: 'metricbeat')

Conversation

mdelapenya commented Apr 21, 2021 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

What does this PR do?

Why is it important?

Checklist

Author's Checklist

Related issues

Other considerations

Uh oh!

elasticmachine commented Apr 21, 2021 • edited by jenkins-beats-ci bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

💚 Build Succeeded

Build stats

Test stats 🧪

💚 Flaky test report

🤖 GitHub comments

Uh oh!

Uh oh!

jsoriano Apr 27, 2021

Choose a reason for hiding this comment

Uh oh!

mdelapenya Apr 28, 2021

Choose a reason for hiding this comment

Uh oh!

jsoriano Apr 27, 2021

Choose a reason for hiding this comment

Uh oh!

mergify bot commented Sep 22, 2021

Uh oh!

mergify bot commented Mar 16, 2022

Uh oh!

mdelapenya commented Mar 16, 2022

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

mdelapenya commented Apr 21, 2021 •

edited

Loading

elasticmachine commented Apr 21, 2021 •

edited by jenkins-beats-ci bot

Loading