问题:有可能节点内存被某进程耗尽,造成多fs的某个mds 损坏
#ceph health detail
HEALTH_ERR mds rank 0 is damaged; mds cluster is degraded
mds.0 is damaged
mds cluster is degraded
解决:ceph fs status查看损坏的fs
Intelligent_Innovation_Labfs - 37 clients
============================
+------+--------+-----------+---------------+-------+-------+
| Rank | State | MDS | Activity | dns | inos |
+------+--------+-----------+---------------+-------+-------+
| 0 | failed | daemon253 | Reqs: 51 /s | 8520k | 8479k |
+------+--------+-----------+---------------+-------+-------+
+-------------------------------------+----------+-------+-------+
| Pool | type | used | avail |
+-------------------------------------+----------+-------+-------+
| Intelligent_Innovation_Lab_metadata | metadata | 214M | 8218G |
| Intelligent_Innovation_Lab_data | data | 27.0T | 8218G |
+-------------------------------------+----------+-------+-------+
最后,执行命令 ceph mds repaired Intelligent_Innovation_Labfs:0
问题解决。
参考:https://blog.csdn.net/mailjoin/article/details/79694965