```

</details>

## 什麼是容器 { #what-is-a-container }

容器（主要是 Linux 容器）是一種非常輕量的方式，用來封裝應用及其所有相依與必要檔案，並讓其與同一系統中的其他容器（其他應用或元件）隔離。

Linux 容器使用與主機（機器、虛擬機、雲端伺服器等）相同的 Linux kernel。這意味著它們非常輕量（相較於完整模擬整個作業系統的虛擬機）。

因此，容器只消耗很少的資源，與直接執行行程相當（而虛擬機會消耗更多）。

容器也有其各自隔離的執行行程（通常只有一個行程）、檔案系統與網路，簡化部署、安全性與開發等。

## 什麼是容器映像 { #what-is-a-container-image }

容器是由容器映像啟動執行的。

  KernelList kernel_list;
  kernel_list.ParseFromArray(kernel_list_buf->data, kernel_list_buf->length);
  ASSERT_EQ(kernel_list.kernel_size(), 1);
  EXPECT_EQ(kernel_list.kernel(0).op(), "TestOpWithSingleKernel");
  EXPECT_EQ(kernel_list.kernel(0).device_type(), "CPU");
  TF_DeleteBuffer(kernel_list_buf);
  TF_DeleteStatus(status);
}

TEST(TestKernel, TestGetRegisteredKernelsForOpNoKernels) {

		Name:      readTotal,
		Help:      "Total read SysCalls to the kernel. /proc/[pid]/io syscr",
		Type:      counterMetric,
	}
}

func getMinIOProcessSysCallWMD() MetricDescription {
	return MetricDescription{
		Namespace: nodeMetricNamespace,
		Subsystem: sysCallSubsystem,
		Name:      writeTotal,
		Help:      "Total write SysCalls to the kernel. /proc/[pid]/io syscw",
		Type:      counterMetric,
	}
}

- Faster mount detection for linux kernel 5.10+ using openat2 speeding up pod churn rates. On Kernel versions less 5.10, it will fallback to using the original way of detecting mount points i.e by parsing /proc/mounts. ([#109217](https://github.com/kubernetes/kubernetes/pull/109217), [@manugupt1](https://github.com/manugupt1))...

                                                 size_t name_len,
                                                 TF_Status* status);

// --------------------------------------------------------------------------
// Kernel definition information.

// Returns a serialized KernelList protocol buffer containing KernelDefs for all
// registered kernels.
TF_CAPI_EXPORT extern TF_Buffer* TF_GetAllRegisteredKernels(TF_Status* status);

    * Add hugepages monitoring support
    * Fix incorrect CPU usage metrics with 4.7 kernel
    * Add tmpfs monitoring support
* Support for Huge pages in empty_dir volume plugin ([#50072](https://github.com/kubernetes/kubernetes/pull/50072), [@squall0gd](https://github.com/squall0gd))
    * [Huge pages](https://www.kernel.org/doc/Documentation/vm/hugetlbpage.txt) can now be used with empty dir volume plugin.

### Bug or Regression

    * Add hugepages monitoring support
    * Fix incorrect CPU usage metrics with 4.7 kernel
    * Add tmpfs monitoring support
* Support for Huge pages in empty_dir volume plugin ([#50072](https://github.com/kubernetes/kubernetes/pull/50072), [@squall0gd](https://github.com/squall0gd))
    * [Huge pages](https://www.kernel.org/doc/Documentation/vm/hugetlbpage.txt) can now be used with empty dir volume plugin.

- Add --permit-address-sharing flag to `kube-apiserver` to listen with `SO_REUSEADDR`. While allowing to listen on wildcard IPs like 0.0.0.0 and specific IPs in parallel, it avoids waiting for the kernel to release socket in `TIME_WAIT` state, and hence, considerably reducing `kube-apiserver` restart times under certain conditions. ([#93861](https://github.com/kubernetes/kubernetes/pull/93861), [@sttts](https://github.com/sttts))

- Add `/sys/devices/virtual/powercap` to default masked paths. It avoids the potential security risk that the ability to read these files may offer a power-based sidechannel attack against any workloads running on the same kernel. ([#125970](https://github.com/kubernetes/kubernetes/pull/125970), [@carlory](https://github.com/carlory)) [SIG Node]
- Fix a bug that Pods could stuck in the unschedulable pod pool