rm排除某文件 openstack状态排查

最新推荐文章于 2024-05-31 10:51:47 发布

wangxuesonger

最新推荐文章于 2024-05-31 10:51:47 发布

阅读量294

点赞数

CC 4.0 BY-SA版权

本文链接：https://blog.youkuaiyun.com/wangxuesonger/article/details/79020660

本文提供了一套全面的OpenStack故障排查流程，包括Neutron、Nova、Cinder等关键组件的状态检查与服务重启策略，适用于快速定位并解决OpenStack平台运行中出现的问题。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

# rm -rf !(audit_siteinfo)

首先检查neutron，保证alive :-) 不行就到compute节点重启

[root@ncontroller heat]# neutron agent-list

检查nova都是up，不行到compute节点重启openstack-nova-compute

[root@compute1 ~]# systemctl list-unit-files | grep openstack

openstack-cinder-api.service disabled

openstack-cinder-backup.service disabled

openstack-cinder-scheduler.service disabled

openstack-cinder-volume.service enabled

openstack-nova-compute.service enabled

直接用systemctl也能查询

[root@compute1 ~]# systemctl | grep cinder
openstack-cinder-volume.service

[root@ncontroller heat]# nova service-list

检查cinder都是up，如果不是到compute节点systemctl restart openstack-cinder-volume，或者注意同步时间

[root@ncontroller heat]# cinder service-list

检查ceph

[root@ncontroller heat]# ceph health

HEALTH_OK

不是OK的话

[root@ncontroller heat]# ceph osd tree

ID WEIGHT TYPE NAME UP/DOWN REWEIGHT PRIMARY-AFFINITY

-1 8.10883 root default

-2 0.54059 host compute1

0 0.54059 osd.0 up 1.00000 1.00000

-3 0.54059 host compute2

有down的话，到compute节点

后来总结干脆重启所有服务

#!/bin/bash

for i in {1..15}
do
ssh -o StrictHostKeyChecking=no compute$i systemctl restart openstack-cinder-volume.service
done
sleep 2
systemctl restart openstack-cinder-api.service
sleep 2
systemctl restart openstack-cinder-scheduler.service
sleep 2
systemctl restart openstack-nova-conductor
sleep 2
systemctl restart openstack-nova-consoleauth
sleep 2
systemctl restart openstack-nova-scheduler
sleep 2
systemctl restart neutron-dhcp-agent
sleep 2
systemctl restart neutron-l3-agent
sleep 2
systemctl restart neutron-metadata-agent
sleep 2
systemctl restart neutron-openvswitch-agent