Add OKD featureset enhancement #1899
Conversation
openshift-ci
bot
added
the
do-not-merge/work-in-progress
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: The full list of commands accepted by this bot can be found here. DetailsNeeds approval from an approver in each of these files:Approvers can indicate their approval by writing |
jatinsu
force-pushed
the
OKD-featuregates
branch
4 times, most recently
from
a0218a1 to
8808148
Compare
enhancements/okd/okd-featureset.md
Outdated
| - Resource consumption impact should be minimal as the feature set itself only controls which features are enabled, not the features themselves | ||
| - Individual features enabled by the OKD feature set may have their own resource implications | ||
|
|
||
| **MicroShift:** |
There was a problem hiding this comment.
OKD featureset shouldn't affect microshift right?
There was a problem hiding this comment.
i don't think so, would be good to check with someone in the Microshift team
openshift-ci
bot
removed
the
do-not-merge/work-in-progress
Prashanth684
changed the title
openshift-ci
bot
added
the
do-not-merge/work-in-progress
enhancements/okd/okd-featureset.md
Outdated
| - Resource consumption impact should be minimal as the feature set itself only controls which features are enabled, not the features themselves | ||
| - Individual features enabled by the OKD feature set may have their own resource implications | ||
|
|
||
| **MicroShift:** |
There was a problem hiding this comment.
i don't think so, would be good to check with someone in the Microshift team
jatinsu
force-pushed
the
OKD-featuregates
branch
from
e35a6c7 to
5f9509a
Compare
jatinsu
force-pushed
the
OKD-featuregates
branch
from
52ec4b8 to
ee8db1a
Compare
jatinsu
force-pushed
the
OKD-featuregates
branch
2 times, most recently
from
79df581 to
f9249fc
Compare
jatinsu
force-pushed
the
OKD-featuregates
branch
from
f9249fc to
3c03c28
Compare
jatinsu
changed the title
openshift-ci
bot
removed
the
do-not-merge/work-in-progress
jatinsu
force-pushed
the
OKD-featuregates
branch
from
3c03c28 to
f2098f5
Compare
|
/retest |
- Condensed enhancement to make it more concise - reworded some features and addressed comments - made points into paragraphs and fixed long lines - Removed version.IsOKD() and explained the logic
jatinsu
force-pushed
the
OKD-featuregates
branch
from
f2098f5 to
a42e502
Compare
| As an OKD cluster administrator, I want to enable TechPreview features on my cluster so that I can adopt upcoming | ||
| OpenShift functionality and provide feedback, without permanently blocking my ability to upgrade. I also want to | ||
| adopt new features earlier in the lifecycle compared to OCP. |
There was a problem hiding this comment.
If we were to promote a feature to the OKD feature set, and then decide we no longer wanted to pursue the feature, would we be ok dropping the feature from these OKD clusters?
How much thought would need to go into these kinds of conversations/decisions?
There was a problem hiding this comment.
How much thought would need to go into these kinds of conversations/decisions?
Selecting the features that need to go into the OKD featureset needs to be done carefully - ideally, when a feature is at a point where it is close to reaching stable and BU has agreed that this is a feature that will make it into OCP, it should be enabled in OKD - not sure how we quantify this other than the fact that TPNU tests succeeed?
There was a problem hiding this comment.
Perhaps defining a process/criteria in this EP for when a feature is suitable for the OKD featureset would be useful?
enhancements/okd/okd-featureset.md
Outdated
|
|
||
| #### Hypershift / Hosted Control Planes | ||
|
|
||
| The OKD feature set applies equally to Hypershift with minimal resource impact |
There was a problem hiding this comment.
Does HCP use the same CVO logic to know to correctly apply the right manifests for OKD?
There was a problem hiding this comment.
if CVO is built for OKD, then it will use the OKD featureset. Will that imply that CVO picks the right manifest?
There was a problem hiding this comment.
Worth asking HCP folks I think, I don't know quite how CVO works in HCP to be able to give you an answer
There was a problem hiding this comment.
After looking a bit into it, it seems like featuregates and featuresets as a whole are a bit broken in HCP. HyperShift applies all CVO manifests found in the payload regardless of the actual feature set that is enabled which and there is a band-aid fix but nothing concrete as of now. I'll fix this section to point out that there is a difference in how hypershift uses CVO. Thanks for pointing this out!
Some resources/tickets for hypershifts behavior with CVO in case you wanted to read:
https://issues.redhat.com/browse/CNTRLPLANE-619
https://issues.redhat.com/browse/OTA-1397
jatinsu
force-pushed
the
OKD-featuregates
branch
from
44f18dd to
b32d0df
Compare
jatinsu
force-pushed
the
OKD-featuregates
branch
from
b32d0df to
4102435
Compare
|
@jatinsu: 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-sigs/prow repository. I understand the commands that are listed here. |
3 participants
This PR is to create an enhancement for the OKD featureset.
For more context: openshift/api#2451