Merge pull request #443 from rancher/v0.4.1

V0.4.1
2019-03-22 20:11:35 -07:00 · 2019-03-22 20:11:35 -07:00 · 7691a575a5
commit 7691a575a5
parent a28e8cb8b9 2c07481b75
7 changed files with 247 additions and 9 deletions
--- a/README.md
+++ b/README.md
@ -10,7 +10,7 @@ You can read more details of Longhorn and its design [here](http://rancher.com/m

 Longhorn is a work in progress. It's an alpha quality software at the moment. We appreciate your comments as we continue to work on it.

-The latest release of Longhorn is **v0.4.0**, shipped with Longhorn Engine **v0.4.0** as the default engine image.
+The latest release of Longhorn is **v0.4.1**, shipped with Longhorn Engine **v0.4.1** as the default engine image.

 ## Source code
 Longhorn is 100% open source software. Project source code is spread across a number of repos:
@ -264,6 +264,8 @@ Longhorn will always try to maintain at least given number of healthy replicas f
 ### [Google Kubernetes Engine](./docs/gke.md)
 ### [Upgrade](./docs/upgrade.md)
 ### [Deal with Kubernetes node failure](./docs/node-failure.md)
+### [Use CSI driver on RancherOS + RKE](./docs/rancheros.md)
+### [Restore a backup to an image file](./docs/restore-to-file.md)

 ## Troubleshooting
 You can click `Generate Support Bundle` link at the bottom of the UI to download a zip file contains Longhorn related configuration and logs.
--- a/deploy/longhorn.yaml
+++ b/deploy/longhorn.yaml
@ -181,7 +181,7 @@ spec:
    spec:
      containers:
      - name: longhorn-manager
-        image: rancher/longhorn-manager:v0.4.0
+        image: rancher/longhorn-manager:v0.4.1
        imagePullPolicy: Always
        securityContext:
          privileged: true
@ -190,9 +190,9 @@ spec:
        - -d
        - daemon
        - --engine-image
-        - rancher/longhorn-engine:v0.4.0
+        - rancher/longhorn-engine:v0.4.1
        - --manager-image
-        - rancher/longhorn-manager:v0.4.0
+        - rancher/longhorn-manager:v0.4.1
        - --service-account
        - longhorn-service-account
        ports:
@ -269,7 +269,7 @@ spec:
    spec:
      containers:
      - name: longhorn-ui
-        image: rancher/longhorn-ui:v0.4.0
+        image: rancher/longhorn-ui:v0.4.1
        ports:
        - containerPort: 8000
        env:
@ -308,23 +308,26 @@ spec:
    spec:
      initContainers:
        - name: wait-longhorn-manager
-          image: rancher/longhorn-manager:v0.4.0
+          image: rancher/longhorn-manager:v0.4.1
          command: ['sh', '-c', 'while [ $(curl -m 1 -s -o /dev/null -w "%{http_code}" http://longhorn-backend:9500/v1) != "200" ]; do echo waiting; sleep 2; done']
      containers:
        - name: longhorn-driver-deployer
-          image: rancher/longhorn-manager:v0.4.0
+          image: rancher/longhorn-manager:v0.4.1
          imagePullPolicy: Always
          command:
          - longhorn-manager
          - -d
          - deploy-driver
          - --manager-image
-          - rancher/longhorn-manager:v0.4.0
+          - rancher/longhorn-manager:v0.4.1
          - --manager-url
          - http://longhorn-backend:9500/v1
          # manually choose "flexvolume" or "csi"
          #- --driver
          #- flexvolume
+          # manually set root directory for csi
+          #- --kubelet-root-dir
+          #- /var/lib/rancher/k3s/agent/kubelet
          env:
          - name: POD_NAMESPACE
            valueFrom:
--- a/docs/rancheros.md
+++ b/docs/rancheros.md
@ -0,0 +1,64 @@
+# Longhorn CSI on RancherOS + RKE
+
+## Requirements
+  1. Kubernetes v1.11 or higher.
+  2. Longhorn v0.4.1 or higher.
+  3. RancherOS Ubuntu console.
+
+
+## Instruction
+### For Kubernetes v1.11 only 
+  The following step is not needed for Kubernetes v1.12+.
+
+  Add extra_binds for kubelet in RKE `cluster.yml`:
+  ```
+   services:
+     kubelet:
+       extra_binds:
+       - "/opt/rke/var/lib/kubelet/plugins:/var/lib/kubelet/plugins" 
+   ```
+   
+### For each node:
+  #### 1. Switch to ubuntu console
+
+  `sudo ros console switch ubuntu`, then type `y`
+
+  #### 2. Install open-iscsi for each node. 
+  ```
+  sudo su
+  apt update
+  apt install -y open-iscsi
+  ```
+  #### 3. Modify configuration for iscsi. 
+    
+  1. Open config file `/etc/iscsi/iscsid.conf`
+  2. Comment `iscsid.startup = /bin/systemctl start iscsid.socket`
+  3. Uncomment `iscsid.startup = /sbin/iscsid`
+    
+
+
+## Background 
+CSI doesn't work with RancherOS + RKE before Longhorn v0.4.1. The reason is:
+
+1. RancherOS sets argument `root-dir=/opt/rke/var/lib/kubelet` for kubelet, , which is different from the default value `/var/lib/kubelet`.
+                                                                             
+2. **For k8s v1.12+**
+
+     Kubelet will detect the `csi.sock` according to argument `<--kubelet-registration-path>` passed in by Kubernetes CSI driver-registrar, and `<drivername>-reg.sock` (for Longhorn, it's `io.rancher.longhorn-reg.sock`) on kubelet path `<root-dir>/plugins`.
+   
+   **For k8s v1.11**
+   
+     Kubelet will find both sockets on kubelet path `/var/lib/kubelet/plugins`.
+   
+3. By default, Longhorn CSI driver create and expose these 2 sock files on host path `/var/lib/kubelet/plugins`.
+
+4. Then kubelet cannot find `<drivername>-reg.sock`, so CSI driver doesn't work.
+
+5. Furthermore, kubelet will instruct CSI plugin to mount Longhorn volume on `<root-dir>/pods/<pod-name>/volumes/kubernetes.io~csi/<volume-name>/mount`.
+
+   But this path inside CSI plugin container won't be binded mount on host path. And the mount operation for Longhorn volume is meaningless.
+   
+   Hence Kubernetes cannot connect to Longhorn using CSI driver.
+
+## Reference
+https://github.com/kubernetes-csi/driver-registrar
--- a/docs/restore-to-file.md
+++ b/docs/restore-to-file.md
@ -0,0 +1,27 @@
+# Use command restore-to-file
+This command gives users the ability to restore a backup to a `raw` image or a `qcow2` image. If the backup is based on a backing file, users should provide the backing file as a `qcow2` image with `--backing file` parameter.
+
+## Instruction
+1. Copy the yaml template
+
+    1.1 Volume has no base image: Make a copy of `examples/restore_to_file.yaml.template` as e.g. `restore.yaml`.
+    
+    1.2 Volume has a base image: Make a copy of `examples/restore_to_file_with_base_image.yaml.template` as e.g. `restore.yaml`, and set argument `backing-file` by replacing `<BASE_IMAGE>` with your base image, e.g. `rancher/longhorn-test:baseimage-ext4`.
+    
+2. Set the node which the output file should be placed on by replacing `<NODE_NAME>`, e.g. `node1`.
+
+3. Specify the host path of output file by modifying field `hostpath` of volume `disk-directory`. By default the directory is `/tmp/restore/`.
+
+4. Set the first argument (backup url) by replacing `<BACKUP_URL>`, e.g. `s3://backupbucket@us-east-1/backupstore?backup=backup-bd326da2c4414b02&volume=volumeexamplename`. Do not delete `''`.
+
+5. Set argument `output-file` by replacing `<OUTPUT_FILE>`, e.g. `volume.raw` or `volume.qcow2`.
+
+6. Set argument `output-format` by replacing `<OUTPUT_FORMAT>`. Now support `raw` or `qcow2` only.
+
+7. Set S3 Credential Secret by replacing `<S3_SECRET_NAME>`, e.g. `minio-secret`. 
+
+8. Execute the yaml using e.g. `kubectl create -f restore.yaml`.
+
+9. Watching the result using `kubectl -n longhorn-system get pod restore-to-file -w`
+
+After the pod status changed to `Completed`, you should able to find `<OUTPUT_FILE>` at e.g. `/tmp/restore` on the `<NODE_NAME>`.
--- a/examples/restore_to_file.yaml.template
+++ b/examples/restore_to_file.yaml.template
@ -0,0 +1,48 @@
+apiVersion: v1
+kind: Pod
+metadata:
+  name: restore-to-file
+  namespace: longhorn-system
+spec:
+  nodeName: <NODE_NAME>
+  containers:
+  - name: restore-to-file
+    command:
+    # set restore-to-file arguments here
+    - /bin/sh
+    - -c
+    - longhorn backup restore-to-file
+      '<BACKUP_URL>'
+      --output-file '/tmp/restore/<OUTPUT_FILE>'
+      --output-format <OUTPUT_FORMAT>
+    # the version of longhorn engine should be v0.4.1 or higher
+    image: rancher/longhorn-engine:v0.4.1
+    imagePullPolicy: IfNotPresent
+    securityContext:
+      privileged: true
+    volumeMounts:
+    - name: disk-directory
+      mountPath: /tmp/restore  # the argument <output-file> should be in this directory
+    env:
+    # set Backup Target Credential Secret here.
+    - name: AWS_ACCESS_KEY_ID
+      valueFrom:
+        secretKeyRef:
+          name: <S3_SECRET_NAME>
+          key: AWS_ACCESS_KEY_ID
+    - name: AWS_SECRET_ACCESS_KEY
+      valueFrom:
+        secretKeyRef:
+          name: <S3_SECRET_NAME>
+          key: AWS_SECRET_ACCESS_KEY
+    - name: AWS_ENDPOINTS
+      valueFrom:
+        secretKeyRef:
+          name: <S3_SECRET_NAME>
+          key: AWS_ENDPOINTS
+  volumes:
+    # the output file can be found on this host path
+    - name: disk-directory
+      hostPath:
+        path: /tmp/restore
+  restartPolicy: Never
--- a/examples/restore_to_file_with_base_image.yaml.template
+++ b/examples/restore_to_file_with_base_image.yaml.template
@ -0,0 +1,94 @@
+apiVersion: v1
+kind: Pod
+metadata:
+  name: restore-to-file
+  namespace: longhorn-system
+spec:
+  nodeName: <NODE_NAME>
+  initContainers:
+  - name: prime-base-image
+    # set base image here
+    command:
+      - /bin/sh
+      - -c
+      - echo primed-base-image
+    # set base image here
+    image: <BASE_IMAGE>
+    imagePullPolicy: Always
+  containers:
+  - name: base-image
+    command:
+    - /bin/sh
+    - -c
+    - mkdir -p /share/base_image &&
+      mount --bind /base_image/ /share/base_image &&
+      echo base image mounted at /share/base_image &&
+      trap 'umount /share/base_image && echo unmounted' TERM &&
+      while true; do $(ls /talk/done 2>&1); if [ $? -eq 0 ]; then break;
+      fi; echo waiting; sleep 1; done;
+      umount /share/base_image && echo unmounted
+    # set base image here
+    image: <BASE_IMAGE>
+    imagePullPolicy: IfNotPresent
+    securityContext:
+      privileged: true
+    volumeMounts:
+    - name: share
+      mountPath: /share
+      mountPropagation: Bidirectional
+    - name: talk
+      mountPath: /talk
+  - name: restore-to-file
+    command:
+    # set restore-to-file arguments here
+    - /bin/sh
+    - -c
+    - while true; do list=$(ls /share/base_image/* 2>&1); if [ $? -eq 0 ]; then break;
+      fi; echo waiting; sleep 1; done; echo Directory found $list;
+      longhorn backup restore-to-file
+      '<BACKUP_URL>'
+      --backing-file $list
+      --output-file '/tmp/restore/<OUTPUT_FILE>'
+      --output-format <OUTPUT_FORMAT>
+      && touch /talk/done && chmod 777 /talk/done && echo created /share/done
+    # the version of longhorn engine should be v0.4.1 or higher
+    image: rancher/longhorn-engine:v0.4.1
+    imagePullPolicy: IfNotPresent
+    securityContext:
+      privileged: true
+    volumeMounts:
+    - name: share
+      mountPath: /share
+      mountPropagation: HostToContainer
+      readOnly: true
+    - name: talk
+      mountPath: /talk
+    - name: disk-directory
+      mountPath: /tmp/restore  # the argument <output-file> should be in this directory
+    env:
+    # set Backup Target Credential Secret here.
+    - name: AWS_ACCESS_KEY_ID
+      valueFrom:
+        secretKeyRef:
+          name: <S3_SECRET_NAME>
+          key: AWS_ACCESS_KEY_ID
+    - name: AWS_SECRET_ACCESS_KEY
+      valueFrom:
+        secretKeyRef:
+          name: <S3_SECRET_NAME>
+          key: AWS_SECRET_ACCESS_KEY
+    - name: AWS_ENDPOINTS
+      valueFrom:
+        secretKeyRef:
+          name: <S3_SECRET_NAME>
+          key: AWS_ENDPOINTS
+  volumes:
+    - name: share
+      emptyDir: {}
+    - name: talk
+      emptyDir: {}
+    # the output file can be found on this host path
+    - name: disk-directory
+      hostPath:
+        path: /tmp/restore
+  restartPolicy: Never
--- a/uninstall/uninstall.yaml
+++ b/uninstall/uninstall.yaml
@ -12,7 +12,7 @@ spec:
    spec:
      containers:
      - name: longhorn-uninstall
-        image: rancher/longhorn-manager:v0.4.0
+        image: rancher/longhorn-manager:v0.4.1
        imagePullPolicy: Always
        command:
        - longhorn-manager