[ubuntu-cloud] Separate CLC.CC servers not seeing nodes.

[Ahmed Kamal]

On 03/28/2011 11:13 AM, Richard Harvey wrote:
> Hi all,
>
> I have separate servers running my cloud components. I followed this 
> guide: https://help.ubuntu.com/community/UEC/PackageInstallSeparate
>
> My set up is:
>
> server1: CLC and Walrus
> server2: CC and CS
> server3,4,5,6: nodes
>
> I can successfully log into the cloud controller webgui and install an 
> image. However when I try and start this image it fails with:
>
> Not enough resources (VmTypeAvailability{type=VmType{name='m1.small', 
> cpu=1, disk=2, mem=192}, max=0, available=0}
> < 1: vm instances.
>
> Listing the resources on the CLC I can see:
>
> root at c4p-uec01:/etc# euca_conf --list-walruses --list-scs 
> --list-clusters --list-nodes
> registered walruses:
>    walrus  192.168.92.20
> registered storage controllers:
>    c4p-cluster1  192.168.92.30
> registered clusters:
>    c4p-cluster1  192.168.92.30
>
> As you can see it can't see the nodes. On the CC you can't 
> run euca_conf --list-nodes as there is no database. But I can run 
> euca_conf --discover-nodes and add the nodes like that. this creates 
> /var/lib/eucalyptus/node.list and populates the values. In the logs it 
> looks like the nodes are registered happily.
>
> Can anyone see what I'm missing here and why its reporting not enough 
> resources?
>
Hi Richard,
So if the nodes are being discovered happily, but not really showing up, 
it might be because the old data is still cached. For that you probably 
need to restart eucalyptus service with the CLEAN=1 parameter .. like
restart eucalyptus CLEAN=1  #so on for other services
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.ubuntu.com/archives/ubuntu-cloud/attachments/20110328/7a061c44/attachment.html>