Downstream Sync 04.13#289
Conversation
Signed-off-by: timflannagan <timflannagan@gmail.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: c5ef8b448cef4c83e3a5f542aad449e1e60b1857
|
/test e2e-gcp-olm-flaky |
|
We can use this as a placeholder for the SD Epic work. /hold |
Signed-off-by: Catherine Chan-Tse <cchantse@redhat.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: f8d7f78b2f8db6f01c0651d5072742752c69663d
Signed-off-by: Per Goncalves da Silva <pegoncal@redhat.com> Co-authored-by: Per Goncalves da Silva <pegoncal@redhat.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: 8abe078086bba81e2765a244b5a16fe1196f0f7c
|
/test e2e-gcp-olm-flaky |
This commit introduces the upgradeStrategy field to the operatorGroup type. The upgradeStrategy field defines OLM behavior when upgrading operators in the namespace. Two upgrade strategies are currently supported, "Default" and "TechPreviewUnsafeFailForward". The "Default" strategy mimics existing behavior. The "TechPreviewUnsafeFailForward" strategy allows users to recover from failed installPlans or failed clusterServiceVersions, it should be noted that these upgrades are unsafe and can lead to unpredictable behavior and potentially dataloss. Upstream-repository: api Upstream-commit: 33310d6154f344da5494c788451bd34cdf54711c
Signed-off-by: Alexander Greene <greene.al1991@gmail.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: 63c09814cbb9cb6c904492e92c4a906167c5e8de
This commit introduces OLM support for fail forward upgrades. Existing upgrade behavior is now defined as "Default" upgrade behavior. A new upgrade strategy referred to as UnsafeFailForward has been introduced in a tech preview state. When fail forward upgrades are enabled, CSVs will now be able to move from the failed state to the replacing state, effectively allowing operators in a namespace to fail forward from failed installs or upgrades. Additionally, it will be possible to recover from failed installPlans by updating the upgrade graph defined in the catalog. Fail forward upgrades are generally unsafe and unsupported as they could result in unexpected behavior or lost data due to the fact that critical version of an operator may not be visited. Signed-off-by: Alexander Greene <greene.al1991@gmail.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: 9d6e88c418c627afde7e32d4b7df456c55a56450
Co-authored-by: Ankita Thomas <ankithom@redhat.com> Co-authored-by: timflannagan <timflannagan@gmail.com> Signed-off-by: timflannagan <timflannagan@gmail.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: 918a4d1d40b53605e3889fb6ddeff60c455c248c
Signed-off-by: timflannagan <timflannagan@gmail.com> Upstream-repository: operator-lifecycle-manager Upstream-commit: 73a2b072172e34b036ebe9fa83b1a6a5bb3259a3
|
/hold cancel - let's get this merged! |
|
/hold cancel |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: perdasilva, timflannagan The full list of commands accepted by this bot can be found here. The pull request process is described here DetailsNeeds approval from an approver in each of these files:
Approvers can indicate their approval by writing |
|
/test e2e-gcp-console-olm |
|
/retest-required |
|
/test e2e-gcp-olm-flaky |
1 similar comment
|
/test e2e-gcp-olm-flaky |
|
/test e2e-upgrade |
|
@perdasilva: all tests passed! Full PR test history. Your PR dashboard. DetailsInstructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. I understand the commands that are listed here. |
We want to use a partial object metadata query for CSVs to reduce resource utilization. We still want to ask questions about whether these CSVs are copied, so we need to refactor this method to take only object metadata. While it's not a perfect port of the previous logic, it defies explanation how an object would have all the other markings of being copied but the wrong status. Signed-off-by: Steve Kuznetsov <skuznets@redhat.com> Upstream-repository: api Upstream-commit: 409ec70c6c889945acb0447397305aa5c2150ea2 Signed-off-by: Steve Kuznetsov <skuznets@redhat.com>
This is a downstream sync PR taking the latest and greatest from the upstream back down, including the fail forward feature for 4.11