[one-users] 回复: RE: VM can not start, STATE is always PEND, NO error info output( system locale cause it!!!)

[shamaya]

Thanks for your replay!
I have resolved this problem by change the system locale to English. Now My "onehost list" output is nice:

$ onehost list
  ID NAME              CLUSTER  RVM   TCPU   FCPU   ACPU    TMEM    FMEM STAT
   5 work-node1        default    0    100     99    100  497.3M  369.9M   on
   9 front-end         default    0    100     94    100 1002.4M  824.7M   on

Yuelin
2011-02-22



shamaya



发件人: <Carsten.Friedrich at csiro.au>
发送时间: 2011-02-22 10:11
主 题: RE: [one-users] VM can not start, STATE is always PEND, NO error info output( system locale cause it!!!)
收件人: <shamaya at 126.com>, <users at lists.opennebula.org>



Your output of “onehost list” does not look good. It is not reporting TCPU or FCPU correctly and ACPU is very small. Also TMEM seems incorrect and TMEM very small. Your VM is most likely “pending” as OpenNebula cannot find a cloud node with enough  resources to fulfil the requirements of the VM template. This is not surprising when looking at the “onehost list” output.
How many cores and how much RAM does work-node1 actually have? If this number is big enough to fulfil the requirements of the VM template you should be able to launch the VM once you fix the “onehost list” problem.
Carsten
 
From: users-bounces at lists.opennebula.org [mailto:users-bounces at lists.opennebula.org] On Behalf Of shamaya
Sent: Tuesday, 22 February 2011 12:39
To: users
Subject: [one-users] VM can not start, STATE is always PEND, NO error info output( system locale cause it!!!)
 
Hi,
  
Problem:
I had installed openNebula in ubuntu 10.04. but can not start any vm. the output of "onevm list" shows that the STAT is always PEND. and I couldn't find any error in log files. 
 
$ onehost list
  ID NAME              CLUSTER  RVM   TCPU   FCPU   ACPU    TMEM    FMEM STAT
   5 work-node1        default    0      0      0    100      0K  366.6M   on
 
$ onevm list
   ID     USER     NAME STAT CPU     MEM        HOSTNAME        TIME
    9 oneadmin ttylinux pend   0      0K                 00 00:00:33
 
Cause:
finally I found what cause this problem: It is locale that cause this!
In my ubuntu:
$ cat /etc/default/locale 
     LANG="zh_CN.UTF-8"
     LANGUAGE="zh_CN:zh"
This would cause IM scripts (e.g. /var/tmp/one/im/kvm.d/kvm.rb) not able to get correct node info. Because in OpenNebula node info gathered by running shell command like "virsh -c qemu:///system nodeinfo". 
Locale besides English  will cause problem!
 
Solution:
modify /etc/default/locale and add the following to it
LANG=en_US.UTF-8
LANGUAGE=en_US.en
  2.  echo “export LANGUAGE=en_US:en” >> ~/.bashrc 
2011-02-22



shamaya
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opennebula.org/pipermail/users-opennebula.org/attachments/20110222/9bd54c99/attachment.htm>