## Changes by Kind

### API Change

- Fixed a bug in the NestedNumberAsFloat64 Unstructured field accessor that could cause it to return rounded float64 values instead of errors when accessing very large int64 values. ([#128099](https://github.com/kubernetes/kubernetes/pull/128099), [@benluddy](https://github.com/benluddy)) [SIG API Machinery]

		actual := LastIndexByte([]byte(test.a), test.b[0])
		if actual != test.i {
			t.Errorf("LastIndexByte(%q,%c) = %v; want %v", test.a, test.b[0], actual, test.i)
		}
	}
}

// test a larger buffer with different sizes and alignments
func TestIndexByteBig(t *testing.T) {
	var n = 1024
	if testing.Short() {
		n = 128
	}
	b := make([]byte, n)
	for i := 0; i < n; i++ {

  // allocatedResources is the storage resource within AllocatedResources tracks the capacity allocated to a PVC. It may
  // be larger than the actual capacity when a volume expansion operation is requested.
  // For storage quota, the larger value from allocatedResources and PVC.spec.resources is used.
  // If allocatedResources is not set, PVC.spec.resources alone is used for quota calculation.

	flags := os.O_CREATE | os.O_WRONLY | os.O_TRUNC

	var w *os.File
	if sync {
		// Perform DirectIO along with fdatasync for larger xl.meta, mostly when
		// xl.meta has "inlined data" we prefer writing O_DIRECT and then doing
		// fdatasync() at the end instead of opening the file with O_DSYNC.
		//

  <mime-type type="application/vemmi"/>
  <mime-type type="application/vividence.scriptfile"/>
  <mime-type type="application/vnd.3gpp.bsf+xml"/>
  <mime-type type="application/vnd.3gpp.pic-bw-large">
    <glob pattern="*.plb"/>
  </mime-type>
  <mime-type type="application/vnd.3gpp.pic-bw-small">
    <glob pattern="*.psb"/>
  </mime-type>
  <mime-type type="application/vnd.3gpp.pic-bw-var">

- KCCM: fixed transient node addition and removal caused by #121090 while syncing load balancers on large clusters with a lot of churn. ([#121091](https://github.com/kubernetes/kubernetes/pull/121091), [@alexanderConstantinescu](https://github.com/alexanderConstantinescu))

- Fixed bug in reflector that couldn't recover from "Too large resource version" errors with API servers 1.17.0-1.18.5 ([#94316](https://github.com/kubernetes/kubernetes/pull/94316), [@janeczku](https://github.com/janeczku)) [SIG API Machinery]

			"-fno-builtin",
		)
	}

	c = append(c, p.GccOptions...)
	c = append(c, gccMachine()...)
	if goos == "aix" {
		c = append(c, "-maix64")
		c = append(c, "-mcmodel=large")
	}
	// disable LTO so we get an object whose symbols we can read
	c = append(c, "-fno-lto")
	c = append(c, "-") //read input from standard input
	return c
}

* Never prevent deletion of resources as part of namespace lifecycle ([#48733](https://github.com/kubernetes/kubernetes/pull/48733), [@liggitt](https://github.com/liggitt))
* AWS cloudprovider plugin: Fix for large clusters (200+ nodes). Also fix bug with volumes not getting detached from a node after reboot. ([#48312](https://github.com/kubernetes/kubernetes/pull/48312), [@gnufied](https://github.com/gnufied))

* If a service of type node port declares multiple ports, quota on "services.nodeports" will charge for each port in the service. ([#29457](https://github.com/kubernetes/kubernetes/pull/29457), [@derekwaynecarr](https://github.com/derekwaynecarr))