我创建了一个ingress，然后应用：
ceph orch apply -i nfs-ingress.yaml
# cat nfs-ingress.yaml
service_type: ingress
service_id: nfs.nfs-cephfs-ha
placement:
hosts:
- ww209 # 您要部署的主机名
spec:
backend_service: nfs.nfs-cephfs # 您后端的 NFS 服务名
virtual_ip: 10.0.19.203/24 # 必须带子网掩码，例如 /24、/23 等
frontend_port: 2049 # 客户端访问的 NFS 端口
monitor_port: 9000 # HAProxy 统计页面端口（可选）
但是一直卡主，ceph orch ls | grep nfs
# ceph orch ls | grep nfs
ingress.nfs.nfs-cephfs-ha 10.0.19.203:2049,9000 0/2 - 20m ww209
nfs.nfs-cephfs ?:2049 1/1 7m ago 15h count:1
查看明细是
# ceph orch ls --service-name ingress.nfs.nfs-cephfs-ha --format yaml
service_type: ingress
service_id: nfs.nfs-cephfs-ha
service_name: ingress.nfs.nfs-cephfs-ha
placement:
hosts:
- ww209
spec:
backend_service: nfs.nfs-cephfs
first_virtual_router_id: 50
frontend_port: 2049
monitor_port: 9000
virtual_ip: 10.0.19.203/24
status:
created: '2025-11-28T01:20:14.776574Z'
ports:
- 2049
- 9000
running: 0
size: 2
virtual_ip: 10.0.19.203/24
events:
- '2025-11-28T01:37:23.031504Z service:ingress.nfs.nfs-cephfs-ha [ERROR] "Failed while
placing keepalived.nfs.nfs-cephfs-ha.ww209.keldtu on ww209: Failed to generate keepalived.conf:
No daemons deployed for ingress.nfs.nfs-cephfs-ha"'
- "2025-11-28T01:38:51.752066Z service:ingress.nfs.nfs-cephfs-ha [ERROR] \"Failed\
\ while placing haproxy.nfs.nfs-cephfs-ha.ww209.mizxcw on ww209: cephadm exited\
\ with an error code: 1, stderr: Non-zero exit code 1 from /usr/bin/docker container\
\ inspect --format {{.State.Status}} ceph-52eabf04-b7ba-11ef-8287-bba5b8705822-haproxy-nfs-nfs-cephfs-ha-ww209-mizxcw\n\
/usr/bin/docker: stdout \n/usr/bin/docker: stderr Error response from daemon: No\
\ such container: ceph-52eabf04-b7ba-11ef-8287-bba5b8705822-haproxy-nfs-nfs-cephfs-ha-ww209-mizxcw\n\
Non-zero exit code 1 from /usr/bin/docker container inspect --format {{.State.Status}}\
\ ceph-52eabf04-b7ba-11ef-8287-bba5b8705822-haproxy.nfs.nfs-cephfs-ha.ww209.mizxcw\n\
/usr/bin/docker: stdout \n/usr/bin/docker: stderr Error response from daemon: No\
\ such container: ceph-52eabf04-b7ba-11ef-8287-bba5b8705822-haproxy.nfs.nfs-cephfs-ha.ww209.mizxcw\n\
Deploy daemon haproxy.nfs.nfs-cephfs-ha.ww209.mizxcw ...\nVerifying port 10.0.19.203:2049\
\ ...\nCannot bind to IP 10.0.19.203 port 2049: [Errno 98] Address already in use\n\
Verifying port 0.0.0.0:9000 ...\nERROR: TCP Port(s) '10.0.19.203:2049,0.0.0.0:9000'\
\ required for haproxy already in use\""
- '2025-11-28T01:38:51.756738Z service:ingress.nfs.nfs-cephfs-ha [ERROR] "Failed while
placing keepalived.nfs.nfs-cephfs-ha.ww209.mcznzd on ww209: Failed to generate keepalived.conf:
No daemons deployed for ingress.nfs.nfs-cephfs-ha"'
- "2025-11-28T01:40:19.799027Z service:ingress.nfs.nfs-cephfs-ha [ERROR] \"Failed\
\ while placing haproxy.nfs.nfs-cephfs-ha.ww209.tijdnt on ww209: cephadm exited\
\ with an error code: 1, stderr: Non-zero exit code 1 from /usr/bin/docker container\
\ inspect --format {{.State.Status}} ceph-52eabf04-b7ba-11ef-8287-bba5b8705822-haproxy-nfs-nfs-cephfs-ha-ww209-tijdnt\n\
/usr/bin/docker: stdout \n/usr/bin/docker: stderr Error response from daemon: No\
\ such container: ceph-52eabf04-b7ba-11ef-8287-bba5b8705822-haproxy-nfs-nfs-cephfs-ha-ww209-tijdnt\n\
Non-zero exit code 1 from /usr/bin/docker container inspect --format {{.State.Status}}\
\ ceph-52eabf04-b7ba-11ef-8287-bba5b8705822-haproxy.nfs.nfs-cephfs-ha.ww209.tijdnt\n\
/usr/bin/docker: stdout \n/usr/bin/docker: stderr Error response from daemon: No\
\ such container: ceph-52eabf04-b7ba-11ef-8287-bba5b8705822-haproxy.nfs.nfs-cephfs-ha.ww209.tijdnt\n\
Deploy daemon haproxy.nfs.nfs-cephfs-ha.ww209.tijdnt ...\nVerifying port 10.0.19.203:2049\
\ ...\nCannot bind to IP 10.0.19.203 port 2049: [Errno 98] Address already in use\n\
Verifying port 0.0.0.0:9000 ...\nERROR: TCP Port(s) '10.0.19.203:2049,0.0.0.0:9000'\
\ required for haproxy already in use\""
- '2025-11-28T01:40:19.803933Z service:ingress.nfs.nfs-cephfs-ha [ERROR] "Failed while
placing keepalived.nfs.nfs-cephfs-ha.ww209.jwsiyg on ww209: Failed to generate keepalived.conf:
No daemons deployed for ingress.nfs.nfs-cephfs-ha"'
数量一致是0/2，请问我该怎么办。
你的日志显示是：端口 2049 冲突。
日志中关键的错误信息是：
原因分析
nfs.nfs-cephfs...) 当前也配置为监听 2049 端口，并且它已经正在运行中。解决方案
你需要将后端的 NFS 服务修改为监听一个非标准端口（例如
12049），把2049端口让给 Ingress (HAProxy) 使用。按照以下步骤操作：
第一步：导出当前的 NFS 服务配置
找到你的后端 NFS 服务名称（通常是
nfs.nfs-cephfs，根据你的ceph orch ps输出）：ceph orch ls --service-name nfs.nfs-cephfs --export > nfs_backend.yaml第二步：修改配置文件
使用文本编辑器（如
vi或nano）打开nfs_backend.yaml。你需要添加或修改
spec下的port字段。将其设置为一个非 2049 的端口（例如 12049）。修改后的文件应该类似这样：
service_type: nfs service_id: nfs-cephfs service_name: nfs.nfs-cephfs placement: count: 1 spec: port: 2049第三步：应用新配置
将修改后的配置应用到集群：
ceph orch apply -i nfs_backend.yaml第四步：验证 NFS 是否重启
应用配置后，Ceph 编排器会重启后端的 NFS 守护进程。
运行以下命令，确认 NFS 守护进程现在是否正在监听新端口（PORTS 列应该显示
*:12049）：ceph orch ps --service_name nfs.nfs-cephfs第五步：验证 Ingress 是否恢复
一旦后端 NFS 释放了 2049 端口，Ingress 服务（HAProxy）应该会自动尝试重启并成功绑定。
运行以下命令查看状态：
ceph orch ps --service_name ingress.nfs.nfs-cephfs-ha或者查看列表：
ceph orch ls --service-name ingress.nfs.nfs-cephfs-ha如果
RUNNING列显示2/2，说明问题已解决。总结
Ingress 架构是：
客户端 (2049)->HAProxy (监听 2049)->NFS Ganesha (监听 12049)你目前两个都试图监听 2049，所以必须把后面那个改成别的端口。
你的答案