Force new cluster causes etcd to panic when learner members are added

[galal-hussein]

etcd version: v3.4.13

Issue:

the issue was seen in k3s k3s-io/k3s#2131 when --cluster-reset is passed which internally passing a --force-new-cluster to etcd node, If --force-new-cluster flag is passed to a cluster where it had previous learner members added, etcd panics with the following error:

{"level":"info","ts":"2020-09-11T16:10:24.260Z","caller":"etcdmain/etcd.go:134","msg":"server has been already initialized","data-dir":"/var/lib/etcd","dir-type":"member"}
{"level":"info","ts":"2020-09-11T16:10:24.260Z","caller":"embed/etcd.go:117","msg":"configuring peer listeners","listen-peer-urls":["http://0.0.0.0:2380"]}
{"level":"info","ts":"2020-09-11T16:10:24.262Z","caller":"embed/etcd.go:127","msg":"configuring client listeners","listen-client-urls":["http://0.0.0.0:2379","http://0.0.0.0:4001"]}
{"level":"info","ts":"2020-09-11T16:10:24.263Z","caller":"embed/etcd.go:302","msg":"starting an etcd server","etcd-version":"3.4.13","git-sha":"ae9734ed2","go-version":"go1.12.17","go-os":"linux","go-arch":"amd64","max-cpu-set":2,"max-cpu-available":2,"member-initialized":true,"name":"etcd0","data-dir":"/var/lib/etcd","wal-dir":"","wal-dir-dedicated":"","member-dir":"/var/lib/etcd/member","force-new-cluster":true,"heartbeat-interval":"100ms","election-timeout":"1s","initial-election-tick-advance":true,"snapshot-count":100000,"snapshot-catchup-entries":5000,"initial-advertise-peer-urls":["http://172.31.35.28:2380"],"listen-peer-urls":["http://0.0.0.0:2380"],"advertise-client-urls":["http://172.31.35.28:2379","http://172.31.35.28:4001"],"listen-client-urls":["http://0.0.0.0:2379","http://0.0.0.0:4001"],"listen-metrics-urls":[],"cors":["*"],"host-whitelist":["*"],"initial-cluster":"","initial-cluster-state":"new","initial-cluster-token":"","quota-size-bytes":2147483648,"pre-vote":false,"initial-corrupt-check":false,"corrupt-check-time-interval":"0s","auto-compaction-mode":"periodic","auto-compaction-retention":"0s","auto-compaction-interval":"0s","discovery-url":"","discovery-proxy":""}
2020-09-11 16:10:24.263932 W | pkg/fileutil: check file permission: directory "/var/lib/etcd" exist, but the permission is "drwxr-xr-x". The recommended permission is "-rwx------" to prevent possible unprivileged access to the data.
{"level":"info","ts":"2020-09-11T16:10:24.264Z","caller":"etcdserver/backend.go:80","msg":"opened backend db","path":"/var/lib/etcd/member/snap/db","took":"118.408µs"}
{"level":"panic","ts":"2020-09-11T16:10:24.265Z","caller":"etcdserver/raft.go:704","msg":"unknown ConfChange Type","type":"ConfChangeAddLearnerNode","stacktrace":"go.etcd.io/etcd/etcdserver.getIDs\n\t/tmp/etcd-release-3.4.13/etcd/release/etcd/etcdserver/raft.go:704\ngo.etcd.io/etcd/etcdserver.restartAsStandaloneNode\n\t/tmp/etcd-release-3.4.13/etcd/release/etcd/etcdserver/raft.go:611\ngo.etcd.io/etcd/etcdserver.NewServer\n\t/tmp/etcd-release-3.4.13/etcd/release/etcd/etcdserver/server.go:482\ngo.etcd.io/etcd/embed.StartEtcd\n\t/tmp/etcd-release-3.4.13/etcd/release/etcd/embed/etcd.go:214\ngo.etcd.io/etcd/etcdmain.startEtcd\n\t/tmp/etcd-release-3.4.13/etcd/release/etcd/etcdmain/etcd.go:302\ngo.etcd.io/etcd/etcdmain.startEtcdOrProxyV2\n\t/tmp/etcd-release-3.4.13/etcd/release/etcd/etcdmain/etcd.go:144\ngo.etcd.io/etcd/etcdmain.Main\n\t/tmp/etcd-release-3.4.13/etcd/release/etcd/etcdmain/main.go:46\nmain.main\n\t/tmp/etcd-release-3.4.13/etcd/release/etcd/main.go:28\nruntime.main\n\t/usr/local/go/src/runtime/proc.go:200"}
panic: unknown ConfChange Type

goroutine 1 [running]:
go.uber.org/zap/zapcore.(*CheckedEntry).Write(0xc00013f550, 0xc000205180, 0x1, 0x1)
	/home/ANT.AMAZON.COM/leegyuho/go/pkg/mod/go.uber.org/zap@v1.10.0/zapcore/entry.go:229 +0x546
go.uber.org/zap.(*Logger).Panic(0xc00007b860, 0x10a7a53, 0x17, 0xc000205180, 0x1, 0x1)
	/home/ANT.AMAZON.COM/leegyuho/go/pkg/mod/go.uber.org/zap@v1.10.0/logger.go:225 +0x7f
go.etcd.io/etcd/etcdserver.getIDs(0xc00007b860, 0x0, 0xc00012c900, 0x7, 0x8, 0x0, 0x0, 0x0)
	/tmp/etcd-release-3.4.13/etcd/release/etcd/etcdserver/raft.go:704 +0x582
go.etcd.io/etcd/etcdserver.restartAsStandaloneNode(0x7ffc2b9e6ddd, 0x5, 0x0, 0x0, 0x0, 0x0, 0xc00019e700, 0x2, 0x2, 0xc000125380, ...)
	/tmp/etcd-release-3.4.13/etcd/release/etcd/etcdserver/raft.go:611 +0x255
go.etcd.io/etcd/etcdserver.NewServer(0x7ffc2b9e6ddd, 0x5, 0x0, 0x0, 0x0, 0x0, 0xc00019e700, 0x2, 0x2, 0xc000125380, ...)
	/tmp/etcd-release-3.4.13/etcd/release/etcd/etcdserver/server.go:482 +0x122f
go.etcd.io/etcd/embed.StartEtcd(0xc000202000, 0xc00035c000, 0x0, 0x0)
	/tmp/etcd-release-3.4.13/etcd/release/etcd/embed/etcd.go:214 +0x988
go.etcd.io/etcd/etcdmain.startEtcd(0xc000202000, 0x10963d6, 0x6, 0xc000125701, 0x2)
	/tmp/etcd-release-3.4.13/etcd/release/etcd/etcdmain/etcd.go:302 +0x40
go.etcd.io/etcd/etcdmain.startEtcdOrProxyV2()
	/tmp/etcd-release-3.4.13/etcd/release/etcd/etcdmain/etcd.go:144 +0x2ef9
go.etcd.io/etcd/etcdmain.Main()
	/tmp/etcd-release-3.4.13/etcd/release/etcd/etcdmain/main.go:46 +0x38
main.main()
	/tmp/etcd-release-3.4.13/etcd/release/etcd/main.go:28 +0x20

To reproduce

• start a 1 member etcd cluster, for example:

docker run -d -v /var/lib/etcd:/var/lib/etcd -p 4001:4001 -p 2380:2380 -p 2379:2379  --name etcd quay.io/coreos/etcd:v3.4.13 etcd  --name etcd0  -advertise-client-urls http://172.31.35.28:2379,http://172.31.35.28:4001  -listen-client-urls http://0.0.0.0:2379,http://0.0.0.0:4001  -initial-advertise-peer-urls http://172.31.35.28:2380  -listen-peer-urls http://0.0.0.0:2380  -initial-cluster-token etcd-cluster-1  --data-dir /var/lib/etcd -initial-cluster etcd0=http://172.31.35.28:2380 -initial-cluster-state new

• add another member as a learner

etcdctl member add --learner etcd1 --peer-urls="http://172.31.34.59:2380"

• start the other etcd node

docker run -d -v /var/lib/etcd:/var/lib/etcd -p 4001:4001 -p 2380:2380 -p 2379:2379  --name etcd quay.io/coreos/etcd:v3.4.13 etcd  --name etcd1  -advertise-client-urls http://172.31.34.59:2379,http://172.31.34.59:4001  --data-dir /var/lib/etcd -listen-client-urls http://0.0.0.0:2379,http://0.0.0.0:4001  -initial-advertise-peer-urls http://172.31.34.59:2380  -listen-peer-urls http://0.0.0.0:2380  -initial-cluster-token etcd-cluster-1  -initial-cluster etcd0=http://172.31.35.28:2380,etcd1=http://172.31.34.59:2380  -initial-cluster-state existing

• promote the learner node

etcdctl member promote 9b83f879a67d44eb

• stop both etcd0 and 1 and then start etcd0 with --force-new-cluster

docker run -d -v /var/lib/etcd:/var/lib/etcd -p 4001:4001 -p 2380:2380 -p 2379:2379  --name etcd quay.io/coreos/etcd:v3.4.13 etcd  --name etcd0  -advertise-client-urls http://172.31.35.28:2379,http://172.31.35.28:4001  -listen-client-urls http://0.0.0.0:2379,http://0.0.0.0:4001  -initial-advertise-peer-urls http://172.31.35.28:2380  -listen-peer-urls http://0.0.0.0:2380  -initial-cluster-token etcd-cluster-1  --data-dir /var/lib/etcd -initial-cluster etcd0=http://172.31.35.28:2380 -initial-cluster-state new --force-new-cluster --logger=zap --log-level=debug

The error appears to be in this line

etcd/etcdserver/raft.go

Line 642 in 76e769c

lg.Panic("unknown ConfChange Type", zap.String("type", cc.Type.String()))

, it appears that getIds() function doesnt handle config change "ConfChangeAddLearnerNode" when getting the ID of members in a given snapshot