Add deploy code, moved from rancher/longhorn-manager

deploy/common.sh

@@ -0,0 +1,28 @@
+#!/bin/bash
+
+cleanup(){
+    name=$1
+    set +e
+    docker rm -vf ${name}
+    set -e
+}
+
+get_container_ip() {
+    container=$1
+    for i in `seq 1 5`
+    do
+        ip=`docker inspect -f '{{range .NetworkSettings.Networks}}{{.IPAddress}}{{end}}' $container`
+        if [ "$ip" != "" ]
+        then
+            break
+        fi
+        sleep 10
+    done
+
+    if [ "$ip" == "" ]
+    then
+        echo cannot find ip for $container
+        exit -1
+    fi
+    echo $ip
+}

deploy/longhorn-deploy-env-etcd.sh

@@ -0,0 +1,36 @@
+#!/bin/bash
+
+set -e
+
+source ./common.sh
+
+network=$1
+
+if [ ${network} == "" ]; then
+        echo Usage: $(basename $0) \<network\>
+        exit 1
+fi
+
+ETCD_SERVER=longhorn-etcd-server
+ETCD_IMAGE=quay.io/coreos/etcd:v3.1.5
+
+cleanup $ETCD_SERVER
+
+docker run -d \
+        --name $ETCD_SERVER \
+        --volume /etcd-data \
+        --network ${network} \
+        $ETCD_IMAGE \
+        /usr/local/bin/etcd \
+        --name longhorn-etcd-server \
+        --data-dir /tmp/etcd-data:/etcd-data \
+        --listen-client-urls http://0.0.0.0:2379 \
+        --advertise-client-urls http://0.0.0.0:2379
+
+etcd_ip=$(get_container_ip $ETCD_SERVER)
+echo etcd server is up at ${etcd_ip}
+echo
+echo Use following command on each node to deploy longhorn
+echo
+echo ./longhorn-deploy-node.sh ${network} ${etcd_ip}
+echo

deploy/longhorn-deploy-env-nfs.sh

@@ -0,0 +1,39 @@
+#!/bin/bash
+
+set -e
+
+network=$1
+
+if [ "${network}" == "" ]; then
+        echo Usage: $(basename $0) \<network\>
+        exit 1
+fi
+
+echo MAKE SURE you have \"nfs-kernel-common\" installed on the host before starting this NFS server
+echo Press Ctrl-C to bail out in 3 seconds
+
+sleep 3
+
+echo WARNING: This NFS server won\'t save any data after you delete the container
+
+source ./common.sh
+
+NFS_SERVER=longhorn-nfs-server
+NFS_IMAGE=docker.io/erezhorev/dockerized_nfs_server
+
+BACKUPSTORE_PATH=/opt/backupstore
+
+docker run -d \
+        --name ${NFS_SERVER} \
+        --net ${network} \
+        --privileged \
+        ${NFS_IMAGE} ${BACKUPSTORE_PATH}
+
+nfs_ip=$(get_container_ip ${NFS_SERVER})
+
+echo NFS server is up
+echo
+echo Use following URL as the Backup Target in the Longhorn:
+echo
+echo nfs://${nfs_ip}:${BACKUPSTORE_PATH}
+echo

deploy/longhorn-deploy-node.sh

@@ -0,0 +1,91 @@
+#!/bin/bash
+
+set -e
+
+source ./common.sh
+
+network=$1
+etcd_ip=$2
+
+if [ "$network" == "" -o "$etcd_ip" == "" ]; then
+        echo usage: $(basename $0) \<network_name\> \<etcd_server_ip\>
+        exit 1
+fi
+
+set +e
+iscsiadm_check=`iscsiadm --version 2>&1`
+if [ $? -ne 0 ]; then
+        echo Cannot find \`iscsiadm\` on the host, please install \`open-iscsi\` package
+        exit 1
+fi
+set -e
+
+LONGHORN_ENGINE_BINARY_NAME="longhorn-engine-binary"
+LONGHORN_ENGINE_IMAGE="rancher/longhorn-engine:046b5a5"
+
+LONGHORN_MANAGER_NAME="longhorn-manager"
+LONGHORN_MANAGER_IMAGE="rancher/longhorn-manager:31b613b"
+
+LONGHORN_DRIVER_NAME="longhorn-driver"
+LONGHORN_DRIVER_IMAGE="imikushin/storage-longhorn:8b1bb5c"
+
+LONGHORN_UI_NAME="longhorn-ui"
+LONGHORN_UI_IMAGE="rancher/longhorn-ui:5528110"
+
+# longhorn-binary first, provides binary to longhorn-manager
+cleanup ${LONGHORN_ENGINE_BINARY_NAME}
+
+docker run --name ${LONGHORN_ENGINE_BINARY_NAME} \
+        --network none \
+        ${LONGHORN_ENGINE_IMAGE} \
+	/bin/bash
+echo ${LONGHORN_ENGINE_BINARY_NAME} is ready
+
+# now longhorn-manager
+cleanup ${LONGHORN_MANAGER_NAME}
+
+docker run -d \
+        --name ${LONGHORN_MANAGER_NAME} \
+        --network ${network} \
+        --privileged \
+        --uts host \
+        -v /dev:/host/dev \
+        -v /var/run:/var/run \
+        -v /var/lib/rancher/longhorn:/var/lib/rancher/longhorn \
+        --volumes-from ${LONGHORN_ENGINE_BINARY_NAME} \
+        ${LONGHORN_MANAGER_IMAGE} \
+        launch-manager -d \
+        --orchestrator docker \
+        --engine-image ${LONGHORN_ENGINE_IMAGE} \
+        --etcd-servers http://${etcd_ip}:2379
+echo ${LONGHORN_MANAGER_NAME} is ready
+
+# finally longhorn-driver
+cleanup ${LONGHORN_DRIVER_NAME}
+
+docker run -d \
+        --name ${LONGHORN_DRIVER_NAME} \
+        --network none \
+        --privileged \
+        -v /run:/run \
+        -v /var/run:/var/run \
+        -v /dev:/host/dev \
+        -v /var/lib/rancher/volumes:/var/lib/rancher/volumes:shared \
+        ${LONGHORN_DRIVER_IMAGE}
+echo ${LONGHORN_DRIVER_NAME} is ready
+
+manager_ip=$(get_container_ip ${LONGHORN_MANAGER_NAME})
+
+host_port=8080
+
+cleanup ${LONGHORN_UI_NAME}
+docker run -d \
+        --name ${LONGHORN_UI_NAME} \
+        --network ${network} \
+        -p ${host_port}:8000/tcp \
+        -e LONGHORN_MANAGER_IP=http://${manager_ip}:9500 \
+        ${LONGHORN_UI_IMAGE}
+echo ${LONGHORN_UI_NAME} is ready
+
+echo
+echo Longhorn is up at port ${host_port}