* Immediately close the other side of the connection when proxying. ([#67288](https://github.com/kubernetes/kubernetes/pull/67288), [@MHBauer](https://github.com/MHBauer))
* Fix potential panic when getting azure load balancer status ([#68609](https://github.com/kubernetes/kubernetes/pull/68609), [@feiskyer](https://github.com/feiskyer))

# CL 158950043 database/sql: add Drivers, returning list of registered drivers, Russ Cox <******@****.***>
pkg database/sql, func Drivers() []string

# CL 117280043 debug/dwarf: fix Reader panic on DW_TAG_unspecified_type, Derek Parker <******@****.***>
pkg debug/dwarf, method (*UnspecifiedType) Basic() *BasicType
pkg debug/dwarf, method (*UnspecifiedType) Common() *CommonType

	if !s.IsValid(w, r) {
		return
	}
	rw := streamHTTPResponse(w)
	defer func() {
		if r := recover(); r != nil {
			debug.PrintStack()
			rw.CloseWithError(fmt.Errorf("panic: %v", r))
		}
	}()

	var req ReadMultipleReq
	mr := msgpNewReader(r.Body)
	defer readMsgpReaderPoolPut(mr)
	err := req.DecodeMsg(mr)
	if err != nil {
		rw.CloseWithError(err)
		return

		shards[i+k] = nil
		if parityOK[i+k] {
			shards[i+k] = splitData[i+k][startOffset:offset]
		}
	}
	// Reconstruct with current config.
	if err := rs.ReconstructData(shards); err != nil {
		panic(fmt.Sprintln("Internal error, could NOT partially reconstruct:", err))
	}
	// Copy reconstructed data back.
	verified := 0
	reconstructed := 0
	for shardsIdx, ok := range shardConfig {
		if ok == 0 {

- Fixed node lifecycle controller panic when conditionType ready is been patch `nil` by mistake. ([#122874](https://github.com/kubernetes/kubernetes/pull/122874), [@fusida](https://github.com/fusida))
- Fixed panic of Evented `PLEG` during kubelet start-up. ([#122475](https://github.com/kubernetes/kubernetes/pull/122475), [@pacoxu](https://github.com/pacoxu))

- Kubeadm: fix panic in the command "kubeadm certs check-expiration" when "/etc/kubernetes/pki" exists but cannot be read. ([#124124](https://github.com/kubernetes/kubernetes/pull/124124), [@carlory](https://github.com/carlory)) [SIG Cluster Lifecycle]

- Fix: GetLabelsForVolume panic issue for azure disk PV ([#92166](https://github.com/kubernetes/kubernetes/pull/92166), [@andyzhangx](https://github.com/andyzhangx)) [SIG Cloud Provider]

// NewGetObjectReader creates a new GetObjectReader. The cleanUpFns
// are called on Close() in FIFO order as passed in ObjReadFn(). NOTE: It is
// assumed that clean up functions do not panic (otherwise, they may
// not all run!).
func NewGetObjectReader(rs *HTTPRangeSpec, oi ObjectInfo, opts ObjectOptions, h http.Header) (
	fn ObjReaderFn, off, length int64, err error,
) {

- Ensure getPrimaryInterfaceID not panic when network interfaces for Azure VMSS are null ([#94802](https://github.com/kubernetes/kubernetes/pull/94802), [@nilo19](https://github.com/nilo19)) [SIG Cloud Provider]
- Fix azure file migration panic ([#94853](https://github.com/kubernetes/kubernetes/pull/94853), [@andyzhangx](https://github.com/andyzhangx)) [SIG Cloud Provider]

* Allow kubernetes components to react to SIGTERM signal and shutdown gracefully. ([#57756](https://github.com/kubernetes/kubernetes/pull/57756), [@mborsz](https://github.com/mborsz))
* Fix scheduler cache panic when updating pod conditions. ([#56733](https://github.com/kubernetes/kubernetes/pull/56733), [@bsalamat](https://github.com/bsalamat))