[Bug]: distinguish unsupported sandbox in-place resize - #933
Conversation
|
[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 |
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## master #933 +/- ##
==========================================
+ Coverage 83.10% 83.15% +0.04%
==========================================
Files 257 259 +2
Lines 22536 22548 +12
==========================================
+ Hits 18728 18749 +21
+ Misses 3094 3079 -15
- Partials 714 720 +6
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
dafc1b7 to
151b434
Compare
| ) | ||
| } | ||
|
|
||
| func isTerminalResourceResizeFailure(message string) bool { |
There was a problem hiding this comment.
this is not good way to fix.
will disscus in issue
ca94a9b to
b58a9c7
Compare
b58a9c7 to
b8b2f2e
Compare
b8b2f2e to
15ac25a
Compare
Signed-off-by: CYJiang <googs1025@gmail.com>
15ac25a to
c04ef17
Compare
Ⅰ. Describe what this PR does
This PR refines Sandbox in-place resource resize failure reporting for clusters where the Pod resize path is not supported.
Before this change, a resize capability mismatch was reported with the same generic condition reason used by other in-place update failures:
That message is visible to operators, but the reason is not machine-readable enough to distinguish an unsupported cluster/runtime resize path from other generic in-place update failures.
This PR adds one structured
InplaceUpdatereason:UnsupportedResize: the cluster/runtime resize path is not supported.Other terminal resource resize failures, such as kubelet reporting an infeasible resize after the Pod has moved to the target revision, continue to use the existing
Failedreason.The claim wait-ready behavior intentionally remains compatibility-preserving.
NewSandboxWaitReadyTaskonly waits while the Sandbox reportsInplaceUpdating; terminal in-place update failure reasons, includingUnsupportedResize, fall through to the existing usable-Sandbox readiness check. As a result, this PR improves status classification without changing the current best-effort claim semantics.Broader API semantics, such as a future
spec.inplaceUpdate.failurePolicywithBestEffort,Fail, orCreateNew, are tracked in #934 and should be designed separately.Ⅱ. Does this pull request fix one issue?
Related to #934
Ⅲ. Describe how to verify it
Ⅳ. Special notes for reviews
This PR does not introduce a fallback policy and does not make
UnsupportedResizedirectly control SandboxClaim success. It only makes the unsupported resize path explicit while preserving the existing best-effort claim behavior.