【 TiDB 使用环境】测试
【 TiDB 版本】6.5.0
【复现路径】通过tidb-operator的tidbcluster.yaml开启TLS
【遇到的问题：通过tidb-operator部署TIDB集群，PD启动异常】
【资源配置】进入到 TiDB Dashboard -集群信息 (Cluster Info) -主机(Hosts) 截图此页面
【附件：截图/日志/监控】
1、目前所有POD可调度
kubectl -n csc get pod -owide | grep basic
basic-discovery-7d7f9f8644-kj7lh 1/1 Running 0 4h36m 100.65.108.98 rxx-ipv6-02
basic-pd-0 1/1 Running 0 4h36m 100.107.182.85 rxx-ipv6-03
basic-pd-1 1/1 Running 0 4h36m 100.64.175.82 rxx-ipv6-01
basic-pd-2 1/1 Running 5 4h35m 100.65.108.66 rxx-ipv6-02
basic-tidb-0 2/2 Running 0 4h35m 100.107.182.73 rxx-ipv6-03
basic-tidb-1 2/2 Running 0 4h35m 100.65.108.103 rxx-ipv6-02
basic-tidb-2 2/2 Running 0 4h35m 100.64.175.77 rxx-ipv6-01
basic-tidb-initializerd-6jh88 0/1 Completed 0 4h49m 100.65.108.93 rxx-ipv6-02
basic-tikv-0 1/1 Running 0 4h35m 100.65.108.101 rxx-ipv6-02
basic-tikv-1 1/1 Running 0 4h35m 100.64.175.83 rxx-ipv6-01
basic-tikv-2 1/1 Running 0 4h35m 100.107.182.74 rxx-ipv6-03
2、tidb\tikv由于PD异常日志报错；
3、PD集群无法构建，选不出leader
[2024/07/30 03:45:42.780 +00:00] [ERROR] [member.go:160] [“no etcd leader, check pd leader later”] [error=“[PD:member:ErrEtcdLeaderNotFound]etcd leader not found”]
[2024/07/30 03:45:42.980 +00:00] [ERROR] [member.go:160] [“no etcd leader, check pd leader later”] [error=“[PD:member:ErrEtcdLeaderNotFound]etcd leader not found”]
[2024/07/30 03:45:43.148 +00:00] [WARN] [v3_server.go:814] [“waiting for ReadIndex response took too long, retrying”] [sent-request-id=660218256426337256] [retry-timeout=500ms]
[2024/07/30 03:45:43.181 +00:00] [ERROR] [member.go:160] [“no etcd leader, check pd leader later”] [error=“[PD:member:ErrEtcdLeaderNotFound]etcd leader not found”]
[2024/07/30 03:45:43.381 +00:00] [ERROR] [member.go:160] [“no etcd leader, check pd leader later”] [error=“[PD:member:ErrEtcdLeaderNotFound]etcd leader not found”]
[2024/07/30 03:45:43.435 +00:00] [WARN] [peer_status.go:68] [“peer became inactive (message send to peer failed)”] [peer-id=fc6f0e3e761ea26] [error=“failed to heartbeat fc6f0e3e761ea26 on stream Message (write tcp 100.107.182.85:2380->100.64.175.82:45216: write: broken pipe)”]
[2024/07/30 03:45:43.437 +00:00] [WARN] [stream.go:193] [“lost TCP streaming connection with remote peer”] [stream-writer-type=“stream Message”] [local-member-id=bad55929a0b50929] [remote-peer-id=fc6f0e3e761ea26]
[2024/07/30 03:45:43.440 +00:00] [WARN] [stream.go:193] [“lost TCP streaming connection with remote peer”] [stream-writer-type=“stream MsgApp v2”] [local-member-id=bad55929a0b50929] [remote-peer-id=fc6f0e3e761ea26]
[2024/07/30 03:45:43.582 +00:00] [ERROR] [member.go:160] [“no etcd leader, check pd leader later”] [error=“[PD:member:ErrEtcdLeaderNotFound]etcd leader not found”]
[2024/07/30 03:45:43.648 +00:00] [WARN] [v3_server.go:814] [“waiting for ReadIndex response took too long, retrying”] [sent-request-id=660218256426337256] [retry-timeout=500ms]
4、pd-2 POD报错
waiting for discovery service to return start args …
waiting for discovery service to return start args …
waiting for discovery service to return start args …
waiting for discovery service to return start args …
waiting for discovery service to return start args …
waiting for discovery service to return start args …
waiting for discovery service to return start args …
waiting for discovery service to return start args …
POD内从discovery 获取信息使用的是http，目前已经开启TLS，此处应该用https？
5、集群使用tidb-operator1.4.3、tidb6.5.0、helm3.2.1部署；
6、tidb-cluster.yaml配置如下：
apiVersion: pingcap.com/v1alpha1
kind: TidbCluster
metadata:
name: basic
namespace: csc
spec:
version: v6.5.0
timezone: UTC
pvReclaimPolicy: Retain
discovery: {}
tlsCluster:
enabled: true
pd:
podManagementPolicy: Parallel
baseImage: pingcap/pd
replicas: 3
tlsClientSecretName: basic-pd-cluster-secret
mountClusterClientSecret: true
# if storageClassName is not set, the default Storage Class of the Kubernetes cluster will be used
storageClassName: tidb-pd-storage-csc
requests:
storage: “1Gi”
config: {}
nodeSelector:
node-role.kubernetes.io/master: “”
affinity:
podAntiAffinity:
requiredDuringSchedulingIgnoredDuringExecution:
- labelSelector:
matchExpressions:
- key: app.kubernetes.io/component
operator: “In”
values:
- “pd”
topologyKey: “kubernetes.io/hostname”
tikv:
podManagementPolicy: Parallel
baseImage: pingcap/tikv
replicas: 3
mountClusterClientSecret: true
# if storageClassName is not set, the default Storage Class of the Kubernetes cluster will be used
storageClassName: tidb-kv-storage-csc
requests:
storage: “20Gi”
config:
raftdb:
compaction-readahead-size: 2MiB
defaultcf:
max-write-buffer-number: 10
target-file-size-base: 32MiB
raftstore:
sync-log: true
readpool:
coprocessor:
use-unified-pool: true
storage:
use-unified-pool: true
unified:
max-thread-count: 10
rocksdb:
compaction-readahead-size: 2MiB
defaultcf:
max-write-buffer-number: 10
target-file-size-base: 32MiB
storage:
block-cache:
capacity: 2GiB
shared: true
strict-capacity-limit: true
reserve-space: 0MB
nodeSelector:
node-role.kubernetes.io/master: “”
affinity:
podAntiAffinity:
requiredDuringSchedulingIgnoredDuringExecution:
- labelSelector:
matchExpressions:
- key: app.kubernetes.io/component
operator: “In”
values:
- “tikv”
topologyKey: “kubernetes.io/hostname”
tidb:
podManagementPolicy: Parallel
baseImage: pingcap/tidb
replicas: 3
tlsClient:
enabled: true
limits:
memory: 8Gi
service:
type: ClusterIP
config: |
[instance]
tidb_slow_log_threshold = 10000
nodeSelector:
node-role.kubernetes.io/master: “”
affinity:
podAntiAffinity:
requiredDuringSchedulingIgnoredDuringExecution:
- labelSelector:
matchExpressions:
- key: app.kubernetes.io/component
operator: “In”
values:
- “tidb”
topologyKey: “kubernetes.io/hostname”
7、集群为3节点：pd3、kv3、tidb*3，crd.yaml官方下载未修改，operator values.yaml使用k8s scheduler；
