It is an error to have multiple caches accessing the same cache directory simultaneously. Most applications should call `new OkHttpClient()` exactly once, configure it with their cache, and use that same instance everywhere. Otherwise the two cache instances will stomp on each other, corrupt the response cache, and possibly crash your program.

## Changelog since v1.15.10

## Changes by Kind

### Other (Bug, Cleanup or Flake)

- Add delays between goroutines for vm instance update ([#88094](https://github.com/kubernetes/kubernetes/pull/88094), [@aramase](https://github.com/aramase)) [SIG Cloud Provider]

   * doesn't need to be a view, copy the returned multimap into a new multimap of your choosing.
   *
   * <p><b>Warning:</b> This method assumes that for any instance {@code k} of {@code
   * EntryTransformer} key type {@code K}, {@code k.equals(k2)} implies that {@code k2} is also of
   * type {@code K}. Using an {@code EntryTransformer} key type for which this may not hold, such as

   * doesn't need to be a view, copy the returned multimap into a new multimap of your choosing.
   *
   * <p><b>Warning:</b> This method assumes that for any instance {@code k} of {@code
   * EntryTransformer} key type {@code K}, {@code k.equals(k2)} implies that {@code k2} is also of
   * type {@code K}. Using an {@code EntryTransformer} key type for which this may not hold, such as

    private final Inet4Address server;
    private final Inet4Address client;
    private final int port;
    private final int flags;

    /**
     * Constructs a TeredoInfo instance.
     *
     * <p>Both server and client can be {@code null}, in which case the value {@code "0.0.0.0"} will
     * be assumed.
     *

## Changes by Kind

### API Change

- We have added a new Priority & Fairness rule that exempts all probes (/readyz, /healthz, /livez) to prevent 
  restarting of "healthy" kube-apiserver instance(s) by kubelet. ([#101114](https://github.com/kubernetes/kubernetes/pull/101114), [@tkashem](https://github.com/tkashem)) [SIG API Machinery]

### Bug or Regression

## Changelog since v1.8.11

### Other notable changes

* GCE: Fix for internal load balancer management resulting in backend services with outdated instance group links. ([#62887](https://github.com/kubernetes/kubernetes/pull/62887), [@nicksardo](https://github.com/nicksardo))

- Fix: skip instance not found when decoupling vmss from lb ([#105836](https://github.com/kubernetes/kubernetes/pull/105836), [@nilo19](https://github.com/nilo19)) [SIG Cloud Provider]

<p>
On machines where <code>SP</code> and <code>PC</code> are
traditionally aliases for a physical, numbered register,
in the Go assembler the names <code>SP</code> and <code>PC</code>
are still treated specially;
for instance, references to <code>SP</code> require a symbol,
much like <code>FP</code>.
To access the actual hardware register use the true <code>R</code> name.
For example, on the ARM architecture the hardware

### Bug or Regression

- Fix: skip instance not found when decoupling vmss from lb ([#105835](https://github.com/kubernetes/kubernetes/pull/105835), [@nilo19](https://github.com/nilo19)) [SIG Cloud Provider]