Open elfosardo opened 4 days ago
[APPROVALNOTIFIER] This PR is NOT APPROVED
This pull-request has been approved by: Once this PR has been reviewed and has the lgtm label, please ask for approval from elfosardo. For more information see the Kubernetes Code Review Process.
The full list of commands accepted by this bot can be found here.
/cc @dtantsur @Rozzii
a partial test has been successfully done in ironic CI https://review.opendev.org/c/openstack/ironic/+/935895 there we switched to quay.io as docker.io was too often having issues due to rate limits
/test metal3-centos-e2e-integration-test-release-1-8 metal3-dev-env-integration-test-ubuntu-main
DOCKER_HUB_PROXY
is configured also in project-infra and CAPM3. One is pointing to docker.io the other is pointing via our Nordix registry proxy. So not 100% sure how this actually changes where we pull stuff during tests.
/cc @Sunnatillo PTAL
@elfosardo: The following tests failed, say /retest
to rerun all failed tests or /retest-required
to rerun all mandatory failed tests:
Test name | Commit | Details | Required | Rerun command |
---|---|---|---|---|
metal3-centos-e2e-integration-test-release-1-8 | 14debc32d71b4960a17a03495151742a1a12c716 | link | true | /test metal3-centos-e2e-integration-test-release-1-8 |
metal3-dev-env-integration-test-ubuntu-main | 14debc32d71b4960a17a03495151742a1a12c716 | link | true | /test metal3-dev-env-integration-test-ubuntu-main |
@elfosardo: The following tests failed, say /retest
to rerun all failed tests or /retest-required
to rerun all mandatory failed tests:
Test name | Commit | Details | Required | Rerun command |
---|---|---|---|---|
metal3-centos-e2e-integration-test-release-1-8 | 14debc32d71b4960a17a03495151742a1a12c716 | link | true | /test metal3-centos-e2e-integration-test-release-1-8 |
metal3-dev-env-integration-test-ubuntu-main | 14debc32d71b4960a17a03495151742a1a12c716 | link | true | /test metal3-dev-env-integration-test-ubuntu-main |
Full PR test history. Your PR dashboard.
This is also not working as-is:
[2024-11-22T08:42:23.067Z] ++ sudo docker pull quay.io/kindest/node:v1.31.0
[2024-11-22T08:42:28.015Z] Error response from daemon: unauthorized: access to the requested resource is not authorized
Yes we need to sync this with https://github.com/metal3-io/project-infra/blob/00a638ebab1a7dff1a87fcb6a6e7c5871425271d/jenkins/scripts/integration_test_env.sh#L65-L67 I was under the impression that we already used the Nordix proxies in all dev-env tests. Where are we seeing issues? In BMO e2e I know it is not configured yet, but it hasn't been an issue there yet as far as I know
@lentzi90 @tuminoid we could assign the value of DOCKER_HUB_PROXY in project-infra to CONTAINER_HUB_PROXY and then change the logic once this merges something like export CONTAINER_HUB_PROXY=$DOCKER_HUB_PROXY right after https://github.com/metal3-io/project-infra/blob/00a638ebab1a7dff1a87fcb6a6e7c5871425271d/jenkins/scripts/integration_test_env.sh#L65-L67
This is also not working as-is:
[2024-11-22T08:42:23.067Z] ++ sudo docker pull quay.io/kindest/node:v1.31.0 [2024-11-22T08:42:28.015Z] Error response from daemon: unauthorized: access to the requested resource is not authorized
@tuminoid right so that is not migrated yet to non rate limited repo, they are actually taking advantage of the docker open source project participation to get unlimited rate, so we could maybe hardcode docker.io there
This is also not working as-is:
[2024-11-22T08:42:23.067Z] ++ sudo docker pull quay.io/kindest/node:v1.31.0 [2024-11-22T08:42:28.015Z] Error response from daemon: unauthorized: access to the requested resource is not authorized
@tuminoid right so that is not migrated yet to non rate limited repo, they are actually taking advantage of the docker open source project participation to get unlimited rate, so we could maybe hardcode docker.io there
Not a fan of hardcoding as we try to utilize the proxy to not stress any registry. (We do have a lot of work to do for that, I give you that.)
This is also not working as-is:
[2024-11-22T08:42:23.067Z] ++ sudo docker pull quay.io/kindest/node:v1.31.0 [2024-11-22T08:42:28.015Z] Error response from daemon: unauthorized: access to the requested resource is not authorized
@tuminoid right so that is not migrated yet to non rate limited repo, they are actually taking advantage of the docker open source project participation to get unlimited rate, so we could maybe hardcode docker.io there
Not a fan of hardcoding as we try to utilize the proxy to not stress any registry. (We do have a lot of work to do for that, I give you that.)
not a fan of hardcoded things either, but I don't see other solution if not dropping this, and I strongly believe that docker rate limits is way worse in terms of "stress" :)
This is also not working as-is:
[2024-11-22T08:42:23.067Z] ++ sudo docker pull quay.io/kindest/node:v1.31.0 [2024-11-22T08:42:28.015Z] Error response from daemon: unauthorized: access to the requested resource is not authorized
@tuminoid right so that is not migrated yet to non rate limited repo, they are actually taking advantage of the docker open source project participation to get unlimited rate, so we could maybe hardcode docker.io there
Not a fan of hardcoding as we try to utilize the proxy to not stress any registry. (We do have a lot of work to do for that, I give you that.)
not a fan of hardcoded things either, but I don't see other solution if not dropping this, and I strongly believe that docker rate limits is way worse in terms of "stress" :)
Point being, Is there is a reason not use the proxy env instead, so it would go via Harbor? Can't dig into the code right now myself.
@tuminoid ok, I think I skipped a step here, yes, we can use the proxy and we should wouldn't that mean keeping DOCKER_HUB_PROXY variable too? the kindest/node image needs docker.io
@tuminoid ok, I think I skipped a step here, yes, we can use the proxy and we should wouldn't that mean keeping DOCKER_HUB_PROXY variable too? the kindest/node image needs docker.io
Yes. With proxy used in more places than it is now, technically we wouldn't need to move to quay.io either. We're depending on both anyways, so it doesn't matter which one we use more. We should just use proxy more than we are now.
This should help avoid issues related to dockerhub rate limits.