vnline

小文件较多场景下，raft心跳超时，leader切换异常

1

CFS Version: 2.4 Release SYSTEM: Centos 7.3 NODE: 18台（metanode\datanode混合部署） 服务配置：128G RAM、HDD: 6T*12 no raid 从2.3版本升级到2.4时，更新metanode后出现以下现象，具体日志同步给@Shuoran Liu 在2.2版本之前由于小文件多，dp较多，修改过以下参数： metanode/raft_server.go： raftConf := &raftstore.Config{ NodeID: m.nodeId, RaftPath: m.raftDir, IPAddr: m.localAddr, HeartbeatPort: heartbeatPort, ReplicaPort:...

volume占用空间大于实际数据空间

2

**Describe the bug** 实际数据大小远小于卷占用空间，删除卷中数据，卷占用空间仍无法释放。 **Environment** CentOS Linux release 7.3.1611 (Core) Kernel:3.10.0-514.el7.x86_64 DataNode Filesystem: XFS ChubaoFS Server Version : v2.2.1 Branch : master Commit : 00ebf481a2f78d94607c72cb69d437dd74cf23c4 Build : go1.14.3 linux amd64...

# kubectl apply -f service.yaml service/elasticsearch-discovery created service/kibana created The Service "elasticsearch" is invalid: spec.externalTrafficPolicy: Invalid value: "Cluster": ExternalTrafficPolicy can only be set on NodePort and LoadBalancer service

How can I convert unit millisecond to seconds in access log

2

I want to customize access log , convert %DURATION% and %RESP(X-ENVOY-UPSTREAM-SERVICE-TIME)% millisecond unit to seconds unit. Is there anyway to do it ? json-fields: - "request_time=%DURATION%" - "upstream_response_time=%RESP(X-ENVOY-UPSTREAM-SERVICE-TIME)%" Use seconds...