Openstack云计算项目实施其三（遇到问题以及注意事项）

Posted 2020-11-03

6 

扩展服务

 

 

7 日常问题排查与解决

 

 

7.1 准备工作遇到的问题

 

 

7.1.1 物理主机分区问题

 

 

案例：

 

在物理主机安装系统过程中忘记调整分区，结果在 openstack 部署之后发现 openstack 集群磁盘总量太小，原因是安装系统默认分区/home 分区过大，而/ 分区过小，openstack 集群磁盘识别的是系统/

 

分区

 

 

解决办法：

 

以控制节点为例

 

查看分区

 

[[email protected]YUN-11 ~]# fdisk -l

 

WARNING: GPT (GUID Partition Table) detected on '/dev/sda'! The util fdisk doesn't support GPT. Use GNU Parted.

 

 

 

Disk /dev/sda: 299.6 GB, 299563483136 bytes

 

255 heads, 63 sectors/track, 36419 cylinders

 

Units = cylinders of 16065 * 512 = 8225280 bytes

 

Sector size (logical/physical): 512 bytes / 512 bytes

 

I/O size (minimum/optimal): 512 bytes / 512 bytes

 

Disk identifier: 0x00000000

 

 

 

Device Boot /dev/sda1        Start                End            Blocks Id                         System 

                                                                                             1                 36420         292542463+        ee        GPT

 

Disk /dev/mapper/vg_YUN2-lv_root: 53.7 GB, 53687091200 bytes 

255 heads, 63 sectors/track, 6527 cylinders

 

 

Units = cylinders of 16065 * 512 = 8225280 bytes

 

Sector size (logical/physical): 512 bytes / 512 bytes

 

I/O size (minimum/optimal): 512 bytes / 512 bytes

 

Disk identifier: 0x00000000

 

 

 

 

 

 

Disk /dev/mapper/vg_YUN2-lv_swap: 16.9 GB, 16894656512 bytes 255 heads, 63 sectors/track, 2053 cylinders

 

Units = cylinders of 16065 * 512 = 8225280 bytes

 

Sector size (logical/physical): 512 bytes / 512 bytes

 

I/O size (minimum/optimal): 512 bytes / 512 bytes

 

Disk identifier: 0x00000000

 

 

 

 

 

Disk /dev/mapper/vg_YUN2-lv_home: 228.2 GB, 228241440768 bytes 255 heads, 63 sectors/track, 27748 cylinders

 

Units = cylinders of 16065 * 512 = 8225280 bytes

 

Sector size (logical/physical): 512 bytes / 512 bytes

 

I/O size (minimum/optimal): 512 bytes / 512 bytes

 

Disk identifier: 0x00000000

 

可 以 看 到 /dev/mapper/vg_YUN2-lv_root 太 小 ， 而 /dev/mapper/vg_YUN2-lv_home 比较大，但是云主机磁盘使用的是/dev/mapper/vg_YUN2-lv_root 分区

 

 

 

下面对分区做一下调整

卸载/home 分区（如果/home 分区下有文件，需要先备份，这里尤为

 

重要）

 

# umount /home

 

 

 

检查分区

 

# e2fsck -f /dev/mapper/vg_YUN2-lv_home

 

 

 

调整 home 分区大小

 

# resize2fs -p /dev/mapper/vg_YUN2-lv_home 2G

 

 

 

挂载/home 分区

 

# mount /home

 

 

 

调整/dev/mapper/vg_YUN2-lv_home 大小

 

# lvreduce -L 2G /dev/mapper/vg_YUN2-lv_home

 

 

 

查看分区

 

[[email protected]YUN-11 ~]# df -h

 

 

Filesystem                                    Size Used Avail Use% Mounted on

 

 

/dev/mapper/vg_YUN2-lv_root

 

50G 2.2G 45G 5% /

 

 

tmpfs	16G	4.0K	16G	1%	/dev/shm
/dev/sda2	477M	49M	404M	11%	/boot
/dev/sda1	200M	260K	200M	1%	/boot/efi

 

/srv/loopback-device/swiftloopback

                                                                 1.9G 3.1M 1.8G 1% /srv/node/swiftloopback 

       /dev/mapper/vg_YUN2-lv_home

                                                                 

                                                      1.9G 25M 1.8G 2% /home

 

[[email protected]YUN-11 ~]# vgdisplay
--- Volume group ---
VG Name	vg_YUN2
System ID
Format	lvm2
Metadata Areas	1
Metadata Sequence No	5
VG Access	read/write
VG Status	resizable
MAX LV	0
Cur LV	3
Open LV	3
Max PV	0
Cur PV	1
Act PV	1
VG Size	278.30 GiB
PE Size	4.00 MiB
Total PE	71245
Alloc PE / Size	17340 / 67.73 GiB
Free  PE / Size	53905 / 210.57 GiB
VG UUID	gwRY1M-61we-pric-21Wk-pTyk-2bF9-i5koBQ
--- Volume group ---
VG Name	cinder-volumes
System ID
Format	lvm2
Metadata Areas	1
Metadata Sequence No	1
VG Access	read/write
VG Status	resizable
MAX LV	0
Cur LV	0
Open LV	0
Max PV	0
Cur PV	1
Act PV	1
VG Size	20.60 GiB

 

 

PE Size	4.00 MiB
Total PE	5273
Alloc PE / Size	0 / 0
Free  PE / Size	5273 / 20.60 GiB
VG UUID	sqfFLq-QOvj-awvz-M5HG-MDj3-5SZE-h4Snlc

 

注意上面粗字体的一行，可以发现已经有 210.57 GiB 空闲空间

 

接下来调整/dev/mapper/vg_YUN2-lv_root 分区

 

# lvextend -L +182.27G /dev/mapper/vg_YUN2-lv_root

 

# resize2fs -p /dev/mapper/vg_YUN2-lv_root

 

再次查看分区
[[email protected]YUN-11 ~]# df -h
Filesystem	Size  Used Avail Use% Mounted on
/dev/mapper/vg_YUN2-lv_root
	229G	2.2G	215G	2% /
tmpfs	16G	4.0K	16G	1% /dev/shm
/dev/sda2	477M	49M	404M	11% /boot
/dev/sda1	200M	260K	200M	1% /boot/efi
/srv/loopback-device/swiftloopback
	1.9G	3.1M	1.8G	1% /srv/node/swiftloopback
/dev/mapper/vg_YUN2-lv_home
	1.9G	25M	1.8G	2% /home
[[email protected]YUN-11 ~]# vgdisplay
--- Volume group ---
VG Name	vg_YUN2
System ID
Format	lvm2
Metadata Areas	1
Metadata Sequence No  6
VG Access	read/write
VG Status	resizable
MAX LV	0
Cur LV	3
Open LV	3
Max PV	0
Cur PV	1
Act PV	1
VG Size	278.30 GiB
PE Size	4.00 MiB
Total PE	71245

 

 

Alloc PE / Size		64002 / 250.01 GiB
Free	PE / Size	7243 / 28.29 GiB
VG UUID		gwRY1M-61we-pric-21Wk-pTyk-2bF9-i5koBQ
--- Volume group ---
VG Name		cinder-volumes
System ID
Format		lvm2
Metadata Areas		1
Metadata Sequence No		1
VG Access		read/write
VG Status		resizable
MAX LV		0
Cur LV		0
Open LV		0
Max PV		0
Cur PV		1
Act PV		1
VG Size		20.60 GiB
PE Size		4.00 MiB
Total PE		5273
Alloc PE / Size		0 / 0
Free	PE / Size	5273 / 20.60 GiB
VG UUID		sqfFLq-QOvj-awvz-M5HG-MDj3-5SZE-h4Snlc

 

可以看到 root 分区调整的更大了

 

 

 

7.1.2 网络规划问题

 

在网络规划中最好不要把 openstack 集群内网 IP 地址和物理服务器集群对内服务网卡 IP 地址在同一个网段，容易造成 IP 地址不容易管理

 

 

7.2 部署环境遇到的问题

 

7.2.1 扩展节点出现异常

 

案例 1：

 

在扩展节点完成后，发现扩展节点的 NOVA 服务状态是 down

 

提示信息：

 

查看日志控制节点日志/var/log/nova/compute.log

 

AMQP server on <server>:5672 is unreachable: Socket closed

 

原因：

 

在扩展节点之前没有在配置节点上配置网桥

 

 

 

案例 2：

 

扩展节点的过程中出现错误

 

提示信息：

 

192.168.0.101_keystone.pp: [ ERROR ]

 

 

 

Applying Puppet manifests [ ERROR ]

 

 

 

ERROR : Error appeared during Puppet run: 192.168.0.101_keystone. pp

 

 

 Error: /Stage[main]/Keystone::Roles::Admin/Keystone_user[admin]: Could not evaluate: Execution of '/usr/bin/keystone --os-auth-url http://127.0.0.1:35357/v2.0/ token-get' returned 1: The request you hav e made requires authentication. (HTTP 401)

 

You will find full trace in log /var/tmp/packstack/20150209-133712-oZcG_v/manifests/192.168.0.101_keystone.pp.log

 

Please check log file /var/tmp/packstack/20150209-133712-oZcG_v/o penstack-setup.log for more information

 

 

原因：

 

在 dashboard 修改登录密码没有在配置文件修改解决办法：

 

# vi packstack-answers-20150130-201639.txt

 

CONFIG_KEYSTONE_ADMIN_PW=hndlyptl（参数之后是重新设置的密码）

 

 

 

 

案例 3：

 扩展节点之后，由于实际需要把扩展节点所在的物理主机替换掉，结果在 dashboard 中仍然存在原来的扩展节点

 

解决办法：(以删除节点 YUN-14 为例)

 

在扩展节点上

 

# mysql mysql>use nova;

 

mysql> select id,service_id,host_ip from compute_nodes; +----+------------+--------------+

 

 

| id | service_id | host_ip				|
+----	+------------	+	--------------	+
|	1 |		5 | 192.168.0.101 |
|	2 |		7 | 192.168.0.102 |
|	3 |		8 | 192.168.0.103 |
|	4 |		9 | 192.168.0.104 |
+----	+------------	+	--------------	+

 

4 rows in set (0.00 sec)

 

 

 

mysql> delete from compute_nodes where id=4; Query OK, 1 row affected (0.01 sec)

 

 

mysql> delete from services where host='YUN-14'; Query OK, 1 row affected (0.00 sec)

 

 

 

7.3 后期运维遇到的问题

 

 

 

 

8 注意事项

 

 

注意事项 1：

 

在创建网络时，需要开启内网的 DHCP 功能

 

 

 

注意事项 2：

 

云主机不能修改 IP 地址

 

 

 

注意事项 3：

 

在创建完实例之后要查看资源占用情况，确认云主机的创建和资源没有冲突

 

 

 

注意事项 4：

 

对于各节点来说对待防火墙一定要慎重处理，不能关闭控制节点以及计算节点的防火墙

 

 

9 资源下载

 

 

repo 文件下载链接

 

http://mirrors.163.com/.help/CentOS6-Base-163.repo

 

http://mirrors.zju.edu.cn/epel/6/x86_64/epel-release-6-8.noarch.rpm

 

https://repos.fedorapeople.org/repos/openstack/openstack-icehouse/rdo-release-icehouse-4.noarch.rpm

 

 

10 小结

 

 

11 参考文档

 

 

部署 RDO 参考文档

 

https://www.rdoproject.org/Neutron_with_existing_external_network

 

http://www.it165.net/os/html/201410/9532.html

 

 

网络优化参考文档

 

https://ask.openstack.org/en/question/25306/slow-network-speed-between-vm-and-external/

 

http://www.chenshake.com/how-node-installation-centos-6-4-openstack-havana-ovsgre/