Skip to content

Containerd can become unusable after the process has been restarted on Windows #5621

New issue
New issue
Closed
#5974
Closed
Containerd can become unusable after the process has been restarted on Windows#5621
#5974
Assignees
kevpar
Labels
kind/bugplatform/windowsWindows
@claudiubelu

Description

@claudiubelu
claudiubelu
opened on Jun 18, 2021

Description

If the containerd process ended abruptly, and restarted afterwards, any previously running Containers / Pods can no longer de stopped / removed. Attempting to do so, on the next containerd restart, the following message can be seen by using crictl:

time="2021-06-18T15:51:17Z" level=fatal msg="listing pod sandboxes: rpc error: code = Unknown desc = server is not initialized yet"

ctr can still be used, and it can be used to kill the remaining tasks. Doing so, on the next containerd restart, it will become usable again.

Steps to reproduce the issue:

On windows:

# create pod / container.
crictl --runtime-endpoint=npipe://./pipe/run/containerd-test/containerd pull docker.io/claudiubelu/pause:3.4.1
$POD_ID=(crictl --runtime-endpoint=npipe://./pipe/run/containerd-test/containerd runp .\pod.json)
$CONTAINER_ID=(crictl --runtime-endpoint=npipe://./pipe/run/containerd-test/containerd create $POD_ID .\container.json .\pod.json)
crictl --runtime-endpoint=npipe://./pipe/run/containerd-test/containerd start $CONTAINER_ID

# before restart.
ctr.exe --address //./pipe//run/containerd-test/containerd --namespace k8s.io task list
crictl --runtime-endpoint=npipe://./pipe/run/containerd-test/containerd ps -a
tskill containerd

# after restart.
ctr.exe --address //./pipe//run/containerd-test/containerd --namespace k8s.io task list
crictl --runtime-endpoint=npipe://./pipe/run/containerd-test/containerd ps -a

# try to stopp and delete the container.
bash -c "timeout 10s crictl.exe --runtime-endpoint npipe://./pipe//run/containerd-test/containerd stopp $POD_ID"

crictl --runtime-endpoint=npipe://./pipe/run/containerd-test/containerd ps -a
ctr.exe --address //./pipe//run/containerd-test/containerd --namespace k8s.io task list

bash -c "timeout 10s crictl.exe --runtime-endpoint npipe://./pipe//run/containerd-test/containerd stop $CONTAINER_ID"

bash -c "timeout 10s crictl.exe --runtime-endpoint npipe://./pipe//run/containerd-test/containerd rmp --force $POD_ID"

crictl --runtime-endpoint=npipe://./pipe/run/containerd-test/containerd ps -a
ctr.exe --address //./pipe//run/containerd-test/containerd --namespace k8s.io task list

# restart containerd again.
tskill containerd

# crictl cannot be used now.
crictl.exe --runtime-endpoint npipe://./pipe//run/containerd-test/containerd pods
crictl --runtime-endpoint=npipe://./pipe/run/containerd-test/containerd ps -a
ctr.exe --address //./pipe//run/containerd-test/containerd --namespace k8s.io task list

ctr.exe --address //./pipe//run/containerd-test/containerd --namespace k8s.io task kill $POD_ID
ctr.exe --address //./pipe//run/containerd-test/containerd --namespace k8s.io task list

crictl --runtime-endpoint=npipe://./pipe/run/containerd-test/containerd ps -a

# final restart.
tskill containerd

crictl.exe --runtime-endpoint npipe://./pipe//run/containerd-test/containerd pods
crictl.exe --runtime-endpoint npipe://./pipe//run/containerd-test/containerd stopp $POD_ID
crictl.exe --runtime-endpoint npipe://./pipe//run/containerd-test/containerd rmp $POD_ID

Files:

pod.json:

{
  "metadata": {
    "name": "ready-sandbox",
    "namespace": "default",
    "attempt": 0,
    "uid": ""52fdfc072182654f163f5f0f9a621d729566c74d10037c4d7bbb0407d1e2c649""
  },
  "logDirectory": "/tmp"
}

container.json:

{
  "metadata": {
      "name": "busybox"
  },
  "image":{
      "image": "docker.io/claudiubelu/pause:3.4.1"
  },
  "command": [],
  "args": []
}

Describe the results you received:

Output on Windows:

PS C:\tmp> # create pod / container.
PS C:\tmp> crictl --runtime-endpoint=npipe://./pipe/run/containerd-test/containerd pull docker.io/claudiubelu/pause:3.4.1
Image is up to date for sha256:a233cb888fa372899fd8cd02bad651820809c11303d8537525e53c159d1a7e1e

PS C:\tmp> $POD_ID=(crictl --runtime-endpoint=npipe://./pipe/run/containerd-test/containerd runp .\pod.json)
PS C:\tmp> $CONTAINER_ID=(crictl --runtime-endpoint=npipe://./pipe/run/containerd-test/containerd create $POD_ID .\container.json .\pod.json)
PS C:\tmp> crictl --runtime-endpoint=npipe://./pipe/run/containerd-test/containerd start $CONTAINER_ID

PS C:\tmp> # before restart.
PS C:\tmp> ctr.exe --address //./pipe//run/containerd-test/containerd --namespace k8s.io task list
ce1f2fa22beed3f635dc464c9ddf449bbb261f3aaf1730e3c30bce03a864f405    6108    RUNNING
81fba9b46124f721665de02ae9535953b9a055aefb6181f1595131696c1fbe87    7136    RUNNING
PS C:\tmp> crictl --runtime-endpoint=npipe://./pipe/run/containerd-test/containerd ps -a
CONTAINER           IMAGE                               CREATED             STATE               NAME                ATTEMPT             POD ID
ce1f2fa22beed       docker.io/claudiubelu/pause:3.4.1   14 seconds ago      Running             busybox             0                   81fba9b46124f

PS C:\tmp> # try to stopp and delete the container.
PS C:\tmp> bash -c "timeout 10s crictl.exe --runtime-endpoint npipe://./pipe//run/containerd-test/containerd stopp $POD_ID"

PS C:\tmp> crictl --runtime-endpoint=npipe://./pipe/run/containerd-test/containerd ps -a
CONTAINER           IMAGE                               CREATED             STATE               NAME                ATTEMPT             POD ID
ce1f2fa22beed       docker.io/claudiubelu/pause:3.4.1   25 seconds ago      Running             busybox             0                   81fba9b46124f
PS C:\tmp> ctr.exe --address //./pipe//run/containerd-test/containerd --namespace k8s.io task list
TASK                                                                PID     STATUS
81fba9b46124f721665de02ae9535953b9a055aefb6181f1595131696c1fbe87    7136    RUNNING
ce1f2fa22beed3f635dc464c9ddf449bbb261f3aaf1730e3c30bce03a864f405    6108    STOPPED

PS C:\tmp> bash -c "timeout 10s crictl.exe --runtime-endpoint npipe://./pipe//run/containerd-test/containerd stop $CONTAINER_ID"
PS C:\tmp> bash -c "timeout 10s crictl.exe --runtime-endpoint npipe://./pipe//run/containerd-test/containerd rmp --force $POD_ID"

PS C:\tmp> crictl --runtime-endpoint=npipe://./pipe/run/containerd-test/containerd ps -a
CONTAINER           IMAGE                               CREATED             STATE               NAME                ATTEMPT             POD ID
ce1f2fa22beed       docker.io/claudiubelu/pause:3.4.1   45 seconds ago      Running             busybox             0                   81fba9b46124f
PS C:\tmp> ctr.exe --address //./pipe//run/containerd-test/containerd --namespace k8s.io task list
TASK                                                                PID     STATUS
81fba9b46124f721665de02ae9535953b9a055aefb6181f1595131696c1fbe87    7136    RUNNING
ce1f2fa22beed3f635dc464c9ddf449bbb261f3aaf1730e3c30bce03a864f405    6108    STOPPED

PS C:\tmp> # restart containerd again.
PS C:\tmp> tskill containerd

PS C:\tmp> # crictl cannot be used now.
PS C:\tmp> crictl.exe --runtime-endpoint npipe://./pipe//run/containerd-test/containerd pods
time="2021-06-18T15:51:17Z" level=fatal msg="listing pod sandboxes: rpc error: code = Unknown desc = server is not initialized yet"
PS C:\tmp> crictl --runtime-endpoint=npipe://./pipe/run/containerd-test/containerd ps -a
time="2021-06-18T15:51:17Z" level=fatal msg="listing containers: rpc error: code = Unknown desc = server is not initialized yet"

PS C:\tmp> ctr.exe --address //./pipe//run/containerd-test/containerd --namespace k8s.io task list
TASK                                                                PID     STATUS
81fba9b46124f721665de02ae9535953b9a055aefb6181f1595131696c1fbe87    7136    RUNNING
ce1f2fa22beed3f635dc464c9ddf449bbb261f3aaf1730e3c30bce03a864f405    6108    STOPPED

PS C:\tmp> ctr.exe --address //./pipe//run/containerd-test/containerd --namespace k8s.io task kill $POD_ID
PS C:\tmp> ctr.exe --address //./pipe//run/containerd-test/containerd --namespace k8s.io task list
TASK                                                                PID     STATUS
81fba9b46124f721665de02ae9535953b9a055aefb6181f1595131696c1fbe87    7136    STOPPED
ce1f2fa22beed3f635dc464c9ddf449bbb261f3aaf1730e3c30bce03a864f405    6108    STOPPED

PS C:\tmp> crictl --runtime-endpoint=npipe://./pipe/run/containerd-test/containerd ps -a
time="2021-06-18T15:51:17Z" level=fatal msg="listing containers: rpc error: code = Unknown desc = server is not initialized yet"

PS C:\tmp> # final restart.
PS C:\tmp> tskill containerd

PS C:\tmp> crictl.exe --runtime-endpoint npipe://./pipe//run/containerd-test/containerd pods
POD ID              CREATED              STATE               NAME                NAMESPACE           ATTEMPT             RUNTIME
81fba9b46124f       About a minute ago   NotReady            ready-sandbox       default             0                   (default)
PS C:\tmp> crictl.exe --runtime-endpoint npipe://./pipe//run/containerd-test/containerd stopp $POD_ID
Stopped sandbox 81fba9b46124f721665de02ae9535953b9a055aefb6181f1595131696c1fbe87
PS C:\tmp> crictl.exe --runtime-endpoint npipe://./pipe//run/containerd-test/containerd rmp $POD_ID
Removed sandbox 81fba9b46124f721665de02ae9535953b9a055aefb6181f1595131696c1fbe87

Describe the results you expected:

On Linux, this isn't an issue, the test TestContainerdRestart ensures that it works properly.

What version of containerd are you using:

$ containerd --version
containerd github.com/containerd/containerd v1.5.0-153-g0a3a77bc4 0a3a77bc445379851a5c737955411fc9e630c3ce

Any other relevant information (runC version, CRI configuration, OS/Kernel version, etc.):

runc --version
$ runc --version

crictl info
$ crictl info

{
  "status": {
    "conditions": [
      {
        "type": "RuntimeReady",
        "status": true,
        "reason": "",
        "message": ""
      },
      {
        "type": "NetworkReady",
        "status": true,
        "reason": "",
        "message": ""
      }
    ]
  },
  "cniconfig": {
    "PluginDirs": [
      "c:/k/azurecni/bin"
    ],
    "PluginConfDir": "c:/k/azurecni/netconf",
    "PluginMaxConfNum": 1,
    "Prefix": "eth",
    "Networks": [
      {
        "Config": {
          "Name": "azure",
          "CNIVersion": "0.3.0",
          "Plugins": [
            {
              "Network": {
                "type": "azure-vnet",
                "capabilities": {
                  "dns": true,
                  "portMappings": true
                },
                "ipam": {
                  "type": "azure-vnet-ipam"
                },
                "dns": {
                  "nameservers": [
                    "10.0.0.10",
                    "168.63.129.16"
                  ],
                  "search": [
                    "svc.cluster.local"
                  ]
                }
              },
              "Source": "{\"AdditionalArgs\":[{\"Name\":\"EndpointPolicy\",\"Value\":{\"ExceptionList\":[\"10.0.0.0
/8\",\"10.240.0.0/12\"],\"Type\":\"OutBoundNAT\"}},{\"Name\":\"EndpointPolicy\",\"Value\":{\"DestinationPrefix\":\"
10.0.0.0/16\",\"NeedEncap\":true,\"Type\":\"ROUTE\"}},{\"Name\":\"EndpointPolicy\",\"Value\":{\"Action\":\"Block\",
\"Direction\":\"Out\",\"Priority\":200,\"Protocols\":\"6\",\"RemoteAddresses\":\"168.63.129.16/32\",\"RemotePorts\"
:\"80\",\"RuleType\":\"Switch\",\"Type\":\"ACL\"}},{\"Name\":\"EndpointPolicy\",\"Value\":{\"Action\":\"Allow\",\"D
irection\":\"In\",\"Priority\":65500,\"Type\":\"ACL\"}},{\"Name\":\"EndpointPolicy\",\"Value\":{\"Action\":\"Allow\
",\"Direction\":\"Out\",\"Priority\":65500,\"Type\":\"ACL\"}}],\"bridge\":\"azure0\",\"capabilities\":{\"dns\":true
,\"portMappings\":true},\"dns\":{\"Nameservers\":[\"10.0.0.10\",\"168.63.129.16\"],\"Search\":[\"svc.cluster.local\
"]},\"ipam\":{\"type\":\"azure-vnet-ipam\"},\"mode\":\"bridge\",\"type\":\"azure-vnet\"}"
            }
          ],
          "Source": "{\r\n    \"cniVersion\":  \"0.3.0\",\r\n    \"name\":  \"azure\",\r\n    \"adapterName\":  \"\
",\r\n    \"plugins\":  [\r\n                    {\r\n                        \"type\":  \"azure-vnet\",\r\n
                 \"mode\":  \"bridge\",\r\n                        \"bridge\":  \"azure0\",\r\n
    \"capabilities\":  {\r\n                                             \"portMappings\":  true,\r\n
                               \"dns\":  true\r\n                                         },\r\n
     \"ipam\":  {\r\n                                     \"type\":  \"azure-vnet-ipam\"\r\n
          },\r\n                        \"dns\":  {\r\n                                    \"Nameservers\":  [\r\n
                                                       \"10.0.0.10\",\r\n
              \"168.63.129.16\"\r\n                                                    ],\r\n
              \"Search\":  [\r\n                                                   \"svc.cluster.local\"\r\n
                                        ]\r\n                                },\r\n                        \"Additi
onalArgs\":  [\r\n                                               {\r\n
      \"Name\":  \"EndpointPolicy\",\r\n                                                   \"Value\":  {\r\n
                                                          \"Type\":  \"OutBoundNAT\",\r\n
                                       \"ExceptionList\":  [\r\n
                                    \"10.0.0.0/8\",\r\n
                           \"10.240.0.0/12\"\r\n
                ]\r\n                                                             }\r\n
                   },\r\n                                               {\r\n
             \"Name\":  \"EndpointPolicy\",\r\n                                                   \"Value\":  {\r\n
                                                                 \"Type\":  \"ROUTE\",\r\n
                                        \"DestinationPrefix\":  \"10.0.0.0/16\",\r\n
                                  \"NeedEncap\":  true\r\n
    }\r\n                                               },\r\n                                               {\r\n
                                                  \"Name\":  \"EndpointPolicy\",\r\n
                    \"Value\":  {\r\n                                                                 \"Type\":  \"
ACL\",\r\n                                                                 \"Protocols\":  \"6\",\r\n
                                                   \"Action\":  \"Block\",\r\n
                            \"Direction\":  \"Out\",\r\n
      \"RemoteAddresses\":  \"168.63.129.16/32\",\r\n
   \"RemotePorts\":  \"80\",\r\n                                                                 \"Priority\":  200
,\r\n                                                                 \"RuleType\":  \"Switch\"\r\n
                                             }\r\n                                               },\r\n
                                   {\r\n                                                   \"Name\":  \"EndpointPol
icy\",\r\n                                                   \"Value\":  {\r\n
                            \"Type\":  \"ACL\",\r\n
 \"Action\":  \"Allow\",\r\n                                                                 \"Direction\":  \"In\"
,\r\n                                                                 \"Priority\":  65500\r\n
                                        }\r\n                                               },\r\n
                              {\r\n                                                   \"Name\":  \"EndpointPolicy\"
,\r\n                                                   \"Value\":  {\r\n
                       \"Type\":  \"ACL\",\r\n                                                                 \"Ac
tion\":  \"Allow\",\r\n                                                                 \"Direction\":  \"Out\",\r\
n                                                                 \"Priority\":  65500\r\n
                                    }\r\n                                               }\r\n
                     ]\r\n                    }\r\n                ]\r\n}\r\n"
        },
        "IFName": "eth0"
      }
    ]
  },
  "config": {
    "containerd": {
      "snapshotter": "windows",
      "defaultRuntimeName": "default",
      "defaultRuntime": {
        "runtimeType": "io.containerd.runhcs.v1",
        "runtimeEngine": "",
        "PodAnnotations": null,
        "ContainerAnnotations": null,
        "runtimeRoot": "",
        "options": {
          "Debug": true,
          "DebugType": 2,
          "SandboxImage": "mcr.microsoft.com/oss/kubernetes/pause:3.4.1-windows-1809-amd64",
          "SandboxIsolation": 0,
          "SandboxPlatform": "windows/amd64"
        },
        "privileged_without_host_devices": false,
        "baseRuntimeSpec": ""
      },
      "untrustedWorkloadRuntime": {
        "runtimeType": "",
        "runtimeEngine": "",
        "PodAnnotations": null,
        "ContainerAnnotations": null,
        "runtimeRoot": "",
        "options": null,
        "privileged_without_host_devices": false,
        "baseRuntimeSpec": ""
      },
      "runtimes": {
        "default": {
          "runtimeType": "io.containerd.runhcs.v1",
          "runtimeEngine": "",
          "PodAnnotations": null,
          "ContainerAnnotations": null,
          "runtimeRoot": "",
          "options": {
            "Debug": true,
            "DebugType": 2,
            "SandboxImage": "mcr.microsoft.com/oss/kubernetes/pause:3.4.1-windows-1809-amd64",
            "SandboxIsolation": 0,
            "SandboxPlatform": "windows/amd64"
          },
          "privileged_without_host_devices": false,
          "baseRuntimeSpec": ""
        },
        "runhcs-wcow-process": {
          "runtimeType": "io.containerd.runhcs.v1",
          "runtimeEngine": "",
          "PodAnnotations": null,
          "ContainerAnnotations": null,
          "runtimeRoot": "",
          "options": {
            "Debug": true,
            "DebugType": 2,
            "SandboxImage": "mcr.microsoft.com/oss/kubernetes/pause:3.4.1-windows-1809-amd64",
            "SandboxPlatform": "windows/amd64"
          },
          "privileged_without_host_devices": false,
          "baseRuntimeSpec": ""
        }
      },
      "noPivot": false,
      "disableSnapshotAnnotations": false,
      "discardUnpackedLayers": true
    },
    "cni": {
      "binDir": "c:/k/azurecni/bin",
      "confDir": "c:/k/azurecni/netconf",
      "maxConfNum": 1,
      "confTemplate": ""
    },
    "registry": {
      "configPath": "",
      "mirrors": {
        "docker.io": {
          "endpoint": [
            "https://registry-1.docker.io"
          ]
        }
      },
      "configs": null,
      "auths": null,
      "headers": null
    },
    "imageDecryption": {
      "keyModel": "node"
    },
    "disableTCPService": true,
    "streamServerAddress": "127.0.0.1",
    "streamServerPort": "0",
    "streamIdleTimeout": "4h0m0s",
    "enableSelinux": false,
    "selinuxCategoryRange": 0,
    "sandboxImage": "mcr.microsoft.com/oss/kubernetes/pause:3.4.1-windows-1809-amd64",
    "statsCollectPeriod": 10,
    "systemdCgroup": false,
    "enableTLSStreaming": false,
    "x509KeyPairStreaming": {
      "tlsCertFile": "",
      "tlsKeyFile": ""
    },
    "maxContainerLogSize": 16384,
    "disableCgroup": false,
    "disableApparmor": false,
    "restrictOOMScoreAdj": false,
    "maxConcurrentDownloads": 3,
    "disableProcMount": false,
    "unsetSeccompProfile": "",
    "tolerateMissingHugetlbController": false,
    "disableHugetlbController": false,
    "ignoreImageDefinedVolumes": false,
    "netnsMountsUnderStateDir": false,
    "containerdRootDir": "C:\\Program Files\\Git\\var\\lib\\containerd-test",
    "containerdEndpoint": "//./pipe//run/containerd-test/containerd",
    "rootDir": "C:\\Program Files\\Git\\var\\lib\\containerd-test\\io.containerd.grpc.v1.cri",
    "stateDir": "C:\\Program Files\\Git\\run\\containerd-test\\io.containerd.grpc.v1.cri"
  },
  "golang": "go1.16.4",
  "lastCNILoadStatus": "OK"
}

uname -a
$ uname -a

Metadata

Metadata

Assignees

Labels

kind/bugplatform/windowsWindows

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions