* wire now (time) to the availability checks in the StatefulSet controller
- this helps to make the controller reconcilliation consistent
* schedule pod availability checks at the correct time in StatefulSets
* replace "k8s.io/klog/v2/ktesting" with "k8s.io/kubernetes/test/utils/ktesting"
for advanced features (e.g. Eventually)
* add StatefulSetAvailabilityCheck test
Add ownersLock to protect concurrent access to node.owners between
GraphBuilder.processGraphChanges() (writer) and GC worker goroutines
reading in blockingDependents() and unblockOwnerReferences() methods.
Also fix concurrent reads in the HTTP debug handler (/graph endpoint)
for owners, dependents, beingDeleted, deletingDependents, and virtual
fields by using their respective thread-safe accessor methods.
Using `t` instead of `tCtx` is subtly wrong: the failure is attributed to the
parent test, not the sub-test. Using a separate function with tCtx as
parameter ensures that t is not in scope of the code and thus this mistake
cannot happen. The number of lines is the same, it's just a bit more code.
For TestRetry another advantage is the reduced indention.
It's worth calling out that the same cannot be done for benchmarks:
- They need methods (Loop) or fields (N) which are not exposed by TContext.
- The `for b.Loop()` pattern only works if the for loop is written exactly
like that.
This avoids having to call the rule lister (which theoretically, but not in
practice) fail and having to iterate over rules which can be ignored (might be
a small performance boost).
Support for DeviceTaintRules depends on a significant amount of
additional code:
- ResourceSlice tracker is a NOP without it.
- Additional informers and corresponding permissions in scheduler and controller.
- Controller code for handling status.
Not all users necessarily need DeviceTaintRules, so adding a second feature
gate for that code makes it possible to limit the blast radius of bugs in that
code without having to turn off device taints and tolerations entirely.
To update the right statuses, the controller must collect more information
about why a pod is being evicted. Updating the DeviceTaintRule statuses then is
handled by the same work queue as evicting pods.
Both operations already share the same client instance and thus QPS+server-side
throttling, so they might as well share the same work queue. Deleting pods is
not necessarily more important than informing users or vice-versa, so there is
no strong argument for having different queues.
While at it, switching the unit tests to usage of the same mock work queue as
in staging/src/k8s.io/dynamic-resource-allocation/internal/workqueue. Because
there is no time to add it properly to a staging repo, the implementation gets
copied.
This was broken since 666a41c2ea when the label value became non-integer encoded
The chance of one controller revision hash label being int-parsable: 7/27 ^ 8 = 0.00002041 = ~0
The chance of both being int-parsable: 0.00002041^2 = ~0
Hash comparison locks in differences in content failing EqualRevision
even when the semantic content is normalized to be equal.
