Unable to create a VDC

Community Support
1

Hi.

i´m not sure whats happening, maybe any here can give me any idea.

I have the following scenario

1 user: that just want to send MV (using oneflow service)
2 host: not shared, just 4 the service.
1 datastore: not assigned to any cluster/group
1 network: private, not assigned too

First, I create the cluster

    [oneadmin@cloud one]$ onecluster show 109
CLUSTER 109 INFORMATION
ID                : 109
NAME              : comunica_C

CLUSTER TEMPLATE
RESERVED_CPU=""
RESERVED_MEM=""

HOSTS
55
56

VNETS

DATASTORES

I only give 2 host

after that… i created the group

[oneadmin@cloud one]$ onegroup show 112
GROUP 112 INFORMATION
ID             : 112
NAME           : comunica_G

GROUP TEMPLATE
DEFAULT_VIEW="user"
GROUP_ADMINS="user_U"
SUNSTONE_VIEWS="user,vdcadmin,admin,cloud"

USERS
ID
175

RESOURCE PROVIDERS
   ZONE CLUSTER
      0     109

RESOURCE USAGE & QUOTAS

    NUMBER OF VMS               MEMORY                  CPU        VOLATILE_SIZE
      1 /       -        6G /        -      2.00 /        -        0M /        -

  NETWORK ID               LEASES
          18         1 /        -

    IMAGE ID          RUNNING VMS
        723         1 /        -

d

technically the VDC is set up???

User_U send a VM but… never start.

     45723 user_U              comunica TEST                                     pend    0      0K                   0d 00h48
[oneadmin@cloud one]$ onevm show 45723
VIRTUAL MACHINE 45723 INFORMATION
ID                  : 45723
NAME                : TEST
USER                : user_U
GROUP               : comunica_G
STATE               : PENDING
LCM_STATE           : LCM_INIT
RESCHED             : No
START TIME          : 02/20 11:09:50
END TIME            : -
DEPLOY ID           : -

VIRTUAL MACHINE MONITORING
NET_RX              : 0K
USED MEMORY         : 0K
NET_TX              : 0K
USED CPU            : 0

PERMISSIONS
OWNER               : um-
GROUP               : ---
OTHER               : ---

VM DISKS
 ID TARGET IMAGE                               TYPE SAVE SAVE_AS
  0 hda    Debian-7-one-4.8 - kvm -  - S file   NO       -

VM NICS
 ID NETWORK              VLAN BRIDGE       IP              MAC
  0 Default-Private        no virbrDEVEL   10.38.2.XX       02:00:0a:26:02XX

USER TEMPLATE

SCHED_REQUIREMENTS="CLUSTER_ID = 109"

VIRTUAL MACHINE TEMPLATE
AUTOMATIC_REQUIREMENTS="!(PUBLIC_CLOUD = YES)"
CONTEXT=[
  DATA OMITED 

  VMID="45723" ]
CPU="2"
GRAPHICS=[
  LISTEN="0.0.0.0",
  PORT="XXXX",
  TYPE="VNC" ]
MEMORY="6144"
OS=[
  ARCH="x86_64" ]
TEMPLATE_ID="XXX"
VCPU="2"
VMID="45723"

As last try… I set an ACL to user_U

105     #175     -H---------  %109     u---     *

user_U can USE host at cluster 109

but MV remains in pending state… and never start… and never give me an error…

any idea?

pd: host seems ok.

  55 nubacesga-05-4  comunica_C        0       0 / 800 (0%)    0K / 15.7G (0%) on
  56 nubacesga-05-2  comunica_C        0       0 / 800 (0%)    0K / 15.7G (0%) on

Thanks again!

UPDATED:

I just created all again from scratch… with new user with auth core… the same config…

2host 1cluster and 1Group access to cluster with 1 user…

after instantiate the template… the error now is:

User is not authorized to use any host

2

Hi,

To deploy the VM, the scheduler needs to have a suitable host and system datastore. For legacy reasons, any system DS in the default (-1) cluster can’t be used, such as the default system (0).

You need to assign a system DS to that cluster, to “pair” it with the host, and grant that group permissions to use it.

3

oh Thanks, I thought that datastores and networks can be used as general (whitout cluster unassigned)

Well, I will try, but change datastores now for us will be a problem

Thanks again Carlos.

4

Hi,

This made me realize that when a system DS is missing, it should be made more clear to the admin. I’ve opened a ticket to fix this:
http://dev.opennebula.org/issues/3638

Regards.

1 Like