如何从 CoreOS 上 docker 映像中的快照恢复 etcd 集群？

Question

我在 vmware 上有一个 Kubernetes 集群 (v1.5.6) 和 3 个节点的 etcd 集群（etcd 版本 3.1.5）。此 etcd 节点运行在 vmware 上的 coreos 上的三个 docker 容器（在三个主机上）中。

我尝试使用以下解决方案备份 etcd：

docker run --rm --net=host -v /tmp:/etcd_backup -e ETCDCTL_API=3 quay.io/coreos/etcd:v3.1.5 etcdctl --endpoints=[1.1.1.1:2379,2.2.2.2:2379,3.3.3.3:2379] snapshot save etcd_backup/snapshot.db

备份已成功完成。

我想在另一个 vmware 环境中从零开始创建这个 kubernetes 集群，但我需要为此从快照恢复 etcd。

到目前为止，我还没有找到适用于 docker 容器中 etcd 的正确解决方案。

我尝试用下面的方法恢复，可惜没有成功

首先，我在运行执行以下命令后创建了一个新的 etcd 节点：

docker run --rm --net=host -v /tmp/etcd_bak:/etcd_backup -e ETCDCTL_API=3 registry:5000/quay.io/coreos/etcd:v3.1.5 etcdctl snapshot restore etcd_backup/snapshot.db --name etcd0 --initial-cluster etcd0=http://etcd0:2380,etcd1=http://etcd1:2380,etcd2=http://etcd2:2380 --initial-cluster-token etcd-cluster-1 --initial-advertise-peer-urls http://etcd0:2380

结果：

2018-06-04 09:25:52.314747 I | etcdserver/membership: added member 7ff5c9c6942f82e [http://etcd0:2380] to cluster 5d1b637f4b7740d5
2018-06-04 09:25:52.314940 I | etcdserver/membership: added member 91b417e7701c2eeb [http://etcd2:2380] to cluster 5d1b637f4b7740d5
2018-06-04 09:25:52.315096 I | etcdserver/membership: added member faeb78734ee4a93d [http://etcd1:2380] to cluster 5d1b637f4b7740d5

不幸的是，没有任何反应。

恢复etcd备份有什么好的解决方案？

如何创建一个空的 etcd cluster/node 以及我应该如何恢复快照？

Answer 1

根据 Etcd Disaster Recovery 文档，您需要使用像您这样的命令从快照中恢复所有三个 etcd 节点，然后运行使用这样的命令恢复三个节点：

etcd \
  --name m1 \
  --listen-client-urls http://host1:2379 \
  --advertise-client-urls http://host1:2379 \
  --listen-peer-urls http://host1:2380 &

另外，你可以从镜像中提取etcdctl，像这样：

docker run --rm -v /opt/bin:/opt/bin registry:5000/quay.io/coreos/etcd:v3.1.5 cp /usr/local/bin/etcdctl /opt/bin

然后使用etcdctl恢复快照：

# ETCDCTL_API=3 ./etcdctl snapshot restore snapshot.db \
  --name m1 \
  --initial-cluster m1=http://host1:2380,m2=http://host2:2380,m3=http://host3:2380 \
  --initial-cluster-token etcd-cluster-1 \
  --initial-advertise-peer-urls http://host1:2380 \
  --data-dir /var/lib/etcd

这会将快照还原到 /var/lib/etcd 目录。然后使用 docker 启动 etcd，不要忘记将 /var/lib/etcd 挂载到您的容器中，并为其指定 --data-dir .

Answer 2

kubernetes 中的 Ectd 是 Docker 容器中的运行，这是我恢复集群所做的工作：

检索 Etcd 集群元数据

docker inspect etcd1

你会得到类似下面的东西：

"Binds": [
    "/etc/ssl/certs:/etc/ssl/certs:ro",
    "/etc/ssl/etcd/ssl:/etc/ssl/etcd/ssl:ro",
    "/var/lib/etcd:/var/lib/etcd:rw"
],
...
"Env": [
    "ETCD_DATA_DIR=/var/lib/etcd",
    "ETCD_ADVERTISE_CLIENT_URLS=https://172.16.60.1:2379",
    "ETCD_INITIAL_ADVERTISE_PEER_URLS=https://172.16.60.1:2380",
    "ETCD_INITIAL_CLUSTER_STATE=existing",
    "ETCD_METRICS=basic",
    "ETCD_LISTEN_CLIENT_URLS=https://172.16.60.1:2379,https://127.0.0.1:2379",
    "ETCD_ELECTION_TIMEOUT=5000",
    "ETCD_HEARTBEAT_INTERVAL=250",
    "ETCD_INITIAL_CLUSTER_TOKEN=k8s_etcd",
    "ETCD_LISTEN_PEER_URLS=https://172.16.60.1:2380",
    "ETCD_NAME=etcd1",
    "ETCD_PROXY=off",
    "ETCD_INITIAL_CLUSTER=etcd1=https://172.16.60.1:2380,etcd2=https://172.16.60.2:2380,etcd3=https://172.16.60.2:2380",
    "ETCD_AUTO_COMPACTION_RETENTION=8",
    "ETCD_TRUSTED_CA_FILE=/etc/ssl/etcd/ssl/ca.pem",
    "ETCD_CERT_FILE=/etc/ssl/etcd/ssl/member-node01.pem",
    "ETCD_KEY_FILE=/etc/ssl/etcd/ssl/member-node01-key.pem",
    "ETCD_PEER_TRUSTED_CA_FILE=/etc/ssl/etcd/ssl/ca.pem",
    "ETCD_PEER_CERT_FILE=/etc/ssl/etcd/ssl/member-node01.pem",
    "ETCD_PEER_KEY_FILE=/etc/ssl/etcd/ssl/member-node01-key.pem",
    "ETCD_PEER_CLIENT_CERT_AUTH=true",
    "PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin"
],
"Cmd": [
    "/usr/local/bin/etcd"
],

复制etcd snapshotdb到其他etcd节点

scp snapshotdb_20180913 node02:/root/  
scp snapshotdb_20180913 node03:/root/

使用原始信息重建新集群

# etcd1
docker stop etcd1
rm -rf /var/lib/etcd

ETCDCTL_API=3 etcdctl snapshot restore snapshotdb_20180913 \
  --cacert /etc/ssl/etcd/ssl/ca.pem \
  --cert /etc/ssl/etcd/ssl/member-node01.pem \
  --key /etc/ssl/etcd/ssl/member-node01-key.pem \
  --name etcd1 \
  --initial-cluster etcd1=https://node01:2380,etcd2=https://node02:2380,etcd3=https://node03:2380 \
  --initial-cluster-token k8s_etcd \
  --initial-advertise-peer-urls https://node01:2380 \
  --data-dir /var/lib/etcd

# etcd2
docker stop etcd2
rm -rf /var/lib/etcd

ETCDCTL_API=3 etcdctl snapshot restore snapshotdb_20180913 \
  --cacert /etc/ssl/etcd/ssl/ca.pem \
  --cert /etc/ssl/etcd/ssl/member-node02.pem \
  --key /etc/ssl/etcd/ssl/member-node02-key.pem \
  --name etcd2 \
  --initial-cluster etcd1=https://node01:2380,etcd2=https://node02:2380,etcd3=https://node03:2380 \
  --initial-cluster-token k8s_etcd \
  --initial-advertise-peer-urls https://node02:2380 \
  --data-dir /var/lib/etcd

# etcd3
docker stop etcd3
rm -rf /var/lib/etcd

ETCDCTL_API=3 etcdctl snapshot restore snapshotdb_20180913 \
  --cacert /etc/ssl/etcd/ssl/ca.pem \
  --cert /etc/ssl/etcd/ssl/member-node03.pem \
  --key /etc/ssl/etcd/ssl/member-node03-key.pem \
  --name etcd3 \
  --initial-cluster etcd1=https://node01:2380,etcd2=https://node02:2380,etcd3=https://node03:2380 \
  --initial-cluster-token k8s_etcd \
  --initial-advertise-peer-urls https://node03:2380 \
  --data-dir /var/lib/etcd

启动容器并检查集群状态

cd /etc/ssl/etcd/ssl
etcdctl \
  --endpoints=https://node01:2379 \
  --ca-file=./ca.pem \
  --cert-file=./member-node01.pem \
  --key-file=./member-node01-key.pem \
  member list

如何从 CoreOS 上 docker 映像中的快照恢复 etcd 集群？

How can I restore etcd cluster from snapshot in docker image on CoreOS?

docker

coreos

etcd

kubernetes