nova.compute.manager ValueError

admin

ERROR nova.compute.manager ValueError: failed to detect a valid IP address from 'compute'
遇到的完整报错如下：
; ?! i. C9 p3 Z# B2 f$ S3 a& C2017-10-21 21:03:46.781 26219 ERROR nova.compute.manager [req-243abb92-155d-42a7-b5ec-4c767dffaa36 - - - - -] No com
pute node record for host compute
2017-10-21 21:03:47.134 26219 WARNING nova.compute.resource_tracker [req-243abb92-155d-42a7-b5ec-4c767dffaa36 - - -
( h! _* v. l1 ?; f* S3 q- -] No compute node record for compute:compute
2017-10-21 21:03:47.135 26219 ERROR nova.compute.manager [req-243abb92-155d-42a7-b5ec-4c767dffaa36 - - - - -] Error
5 U* ?9 W. K$ Eupdating resources for node compute.
- [( l# y1 a7 _- N2017-10-21 21:03:47.135 26219 ERROR nova.compute.manager Traceback (most recent call last):
; Z- N  Q4 F; u+ ?$ ~2017-10-21 21:03:47.135 26219 ERROR nova.compute.manager   File "/usr/lib/python2.7/site-packages/nova/compute/manag
2 f. c5 W7 M! i1 {8 Xer.py", line 6571, in update_available_resource_for_node
2017-10-21 21:03:47.135 26219 ERROR nova.compute.manager     rt.update_available_resource(context, nodename)
2017-10-21 21:03:47.135 26219 ERROR nova.compute.manager   File "/usr/lib/python2.7/site-packages/nova/compute/resou
0 r8 W& Y$ K6 V2 k1 J) Trce_tracker.py", line 551, in update_available_resource
2017-10-21 21:03:47.135 26219 ERROR nova.compute.manager     self._update_available_resource(context, resources)
  y" a5 u. a5 z0 R: R2017-10-21 21:03:47.135 26219 ERROR nova.compute.manager   File "/usr/lib/python2.7/site-packages/oslo_concurrency/l
ockutils.py", line 271, in inner
0 z4 ?) u6 G  n4 H! w2017-10-21 21:03:47.135 26219 ERROR nova.compute.manager     return f(*args, **kwargs)
2017-10-21 21:03:47.135 26219 ERROR nova.compute.manager   File "/usr/lib/python2.7/site-packages/nova/compute/resou
rce_tracker.py", line 575, in _update_available_resource
, p& A' A+ Z, H, t9 t2017-10-21 21:03:47.135 26219 ERROR nova.compute.manager     self._init_compute_node(context, resources)
7 S0 ?5 W* c8 y( G0 Q4 g2017-10-21 21:03:47.135 26219 ERROR nova.compute.manager   File "/usr/lib/python2.7/site-packages/nova/compute/resou
rce_tracker.py", line 456, in _init_compute_node
2 Z# k$ v2 S6 F2017-10-21 21:03:47.135 26219 ERROR nova.compute.manager     self._copy_resources(cn, resources)
2017-10-21 21:03:47.135 26219 ERROR nova.compute.manager   File "/usr/lib/python2.7/site-packages/nova/compute/resou
; g; {1 i* e  z* nrce_tracker.py", line 491, in _copy_resources
2017-10-21 21:03:47.135 26219 ERROR nova.compute.manager     compute_node.update_from_virt_driver(resources)
+ k3 w9 v  g  m2017-10-21 21:03:47.135 26219 ERROR nova.compute.manager   File "/usr/lib/python2.7/site-packages/nova/objects/compu
" e8 G! Z4 \) I, Y: Yte_node.py", line 338, in update_from_virt_driver
2017-10-21 21:03:47.135 26219 ERROR nova.compute.manager     setattr(self, key, resources[key])
4 }7 i# S% m5 O, z2017-10-21 21:03:47.135 26219 ERROR nova.compute.manager   File "/usr/lib/python2.7/site-packages/oslo_versionedobje
) T) N$ G. y0 Ncts/base.py", line 72, in setter
2017-10-21 21:03:47.135 26219 ERROR nova.compute.manager     field_value = field.coerce(self, name, value)
5 v5 }" f2 a. H% D8 D( M. C& v2017-10-21 21:03:47.135 26219 ERROR nova.compute.manager   File "/usr/lib/python2.7/site-packages/oslo_versionedobje
cts/fields.py", line 195, in coerce
2017-10-21 21:03:47.135 26219 ERROR nova.compute.manager     return self._type.coerce(obj, attr, value)
2017-10-21 21:03:47.135 26219 ERROR nova.compute.manager   File "/usr/lib/python2.7/site-packages/oslo_versionedobje
8 q: U; a( n/ N4 W9 {cts/fields.py", line 497, in coerce
2017-10-21 21:03:47.135 26219 ERROR nova.compute.manager     raise ValueError(six.text_type(e))
/ e6 W7 ~6 c* G6 o- x: |  t$ d2017-10-21 21:03:47.135 26219 ERROR nova.compute.manager ValueError: failed to detect a valid IP address from 'compu
te'
  {/ k1 d1 z0 F/ a& \  C2017-10-21 21:03:47.135 26219 ERROR nova.compute.manager



检查计算节点nova.conf文件my_ip是否写对：
! b2 f. A4 s1 n; p& s! F  my_ip = 本地ip地址

写成本地IP地址，就不再报错。
* b, b" c" d' }2 o  H
   这个地方有些不完美，我明明写了hosts域名解析了，为什么不解析呢？
" p0 f7 Z6 @, S" ^' E/ g4 u

admin

ping compute
PING compute (192.168.2.2) 56(84) bytes of data.
* B; P9 N$ H. y) [+ c  J- V64 bytes from compute (192.168.2.2): icmp_seq=1 ttl=64 time=0.077 ms
& @( ?4 ~, s  O) p% ?64 bytes from compute (192.168.2.2): icmp_seq=2 ttl=64 time=0.039 ms
c64 bytes from compute (192.168.2.2): icmp_seq=3 ttl=64 time=0.053 ms
64 bytes from compute (192.168.2.2): icmp_seq=4 ttl=64 time=0.041 ms

% m+ \, w  y! C' N! i9 V* V1 t
这个my_ip理论来讲应该是写compute这个地址也是对的，可是偏偏报了一大堆的错误，实在难解释。

admin

[root@controller ~]# openstack hypervisor list

[root@controller ~]# openstack hypervisor list

怎么都不显示，以为自己的配置有问题，从头到尾的再弄了一遍，还是这个问题，看到原始文档，发现有个说：

Note: If the nova-compute service fails to start, check /var/log/nova/nova-compute.log. The error messageAMQP server on controller:5672 is unreachable likelyindicatesthatthefirewallonthecontroller node is preventing access to port 5672. Configure the firewall to open port 5672 on the controller node and restart nova-compute service on the compute node.

以为是nova里的配置文件中有rabbitmq配置有问题，结果不是，发现是compute的这个域名不能解析，bug不合理的地方。控制节点都没有问题的，不知道怎么解释。

( k$ A1 @5 @- v  r/ F
修改计算节点的配置之后，发现 openstack hypervisor list 获取到结果了。
% ^3 m  k. ?. Y

# openstack hypervisor list

+----+---------------------+-----------------+---------------+-------+

| ID | Hypervisor Hostname | Hypervisor Type| Host IP       | State |

+----+---------------------+-----------------+---------------+-------+

|  1 |compute         | QEMU            | 192.168.0.13 | up    |

+----+---------------------+-----------------+---------------+-------+

% [6 ~! w8 ~0 D! u5 S7 c

admin

问题总算解决了 ，获取到结果已经很不错了，compute.log日志也不在报错了。
% i" p2 e+ {$ z" d* Q2017-10-21 21:19:29.765 26377 INFO nova.compute.resource_tracker [req-c0e8a3ec-87a8-452c-96d0-d666c4f83205 - - - - -] Final resource view: name=compute phys_ram=65502MB used_ram=512MB phys_disk=10022GB used_disk=0GB total_vcpus=32 used_vcpus=0 pci_stats=[]

欢迎大家一起讨论。
& Z! w* m* S4 u$ r9 o9 h8 w