安装之前的准备我的操作系统[rootceph02 ~]# uname -aLinux ceph026.6.0-105.0.0.111.oe2403sp2.aarch64#1 SMP Wed Aug 13 16:24:53 CST 2025 aarch64 aarch64 aarch64 GNU/Linux[rootceph02 ~]# cat /etc/os-releaseNAMEopenEulerVERSION24.03 (LTS-SP2)IDopenEulerVERSION_ID24.03PRETTY_NAMEopenEuler 24.03 (LTS-SP2)ANSI_COLOR0;31通过网盘分享的文件arm64 ceph安装包链接: https://pan.baidu.com/s/1p4xr1uNzP8sRxdxiFiU-zA?pwdbd6k 提取码: bd6k导入镜像安装cephadmdockerload-iceph-v19.2.6.tardockerload-igrafana.tardockerload-iprometheus.tardockerload-inode-exporter.tardockerload-ialertmanager.tarmvcephadm /usr/sbin/chmodx /usr/sbin/cephadm一、先过掉 cephadm 的时间检查openEuler 24.03 上 cephadm 认不出 chronyd即使 chronyd 在运行也会报 No time sync。必须先创建一个 cephadm 能识别的兼容服务。在 ceph01 上执行cat/etc/systemd/system/chrony.serviceEOF [Unit] DescriptionChrony NTP compat shim for cephadm Afterchronyd.service Requireschronyd.service [Service] Typeoneshot ExecStart/bin/true RemainAfterExityes [Install] WantedBymulti-user.target EOFsystemctl daemon-reload systemctlenable--nowchrony.service二、安装ceph集群cephadm--imagequay.io/ceph/ceph:v19.2.6 bootstrap --mon-ip10.90.11.16 --skip-prepare-host --skip-pull三、加节点ssh-copy-id-f-i/etc/ceph/ceph.pub rootceph02 ssh-copy-id-f-i/etc/ceph/ceph.pub rootceph03#进入容器exportCEPHADM_IMAGEquay.io/ceph/ceph:v19.2.6 cephadm shell ceph configsetmgr mgr/cephadm/use_repo_digestfalse#关闭 repo digest离线部署关键,使用本地镜像ceph orchhostaddceph0210.90.11.17 ceph orchhostaddceph0310.90.11.18 ceph orch apply mon--placementceph01,ceph02,ceph03ceph orch apply mgr--placementceph01,ceph02,ceph03ceph-sceph orchhostls#删除不必要的组件ceph orchrmalertmanager ceph orchrmgrafana ceph orchrmprometheus ceph orchrmnode-exporter确认离线镜像配置cephadm--imagequay.io/ceph/ceph:v19.2.6 shell --\ceph configsetglobal container_image quay.io/ceph/ceph:v19.2.6四、增加osdwipefs-a/dev/nvme0n1#擦除 /dev/nvme0n1这块磁盘上所有已知的文件系统#进入容器[rootceph01 ~]# export CEPHADM_IMAGEquay.io/ceph/ceph:v19.2.6[rootceph01 ~]# cephadm shellInferring fsid 3924d610-c7b0-11f1-b569-fc119386944e Inferring config /var/lib/ceph/3924d610-c7b0-11f1-b569-fc119386944e/mon.ceph01/config[ceph: rootceph01 /]# ceph -scluster: id: 3924d610-c7b0-11f1-b569-fc119386944e health: HEALTH_OK services: mon:3daemons, quorum ceph01,ceph03,ceph02(age 17h)mgr: ceph01.jsgdms(active, since 18h), standbys: ceph03.lfeosf, ceph02.owsuba osd:3osds:3up(since 17h),3in(since 17h)data: pools:1pools,1pgs objects:2objects,577KiB usage:91MiB used,5.2TiB /5.2TiB avail pgs:1activeclean