[sheepdog-users] problem with cluster recover

[MORITA Kazutaka]

At Tue, 12 Feb 2013 17:20:45 +0100,
Valerio Pachera wrote:
> 
> Hi,
> 
> qemu 1.3.50
> sheepdog 0.5.5_101_gee9838f
> 
> 3 nodes 3 copies (with 2 vdi)
> cluster shutdown
> 2 node start + 1 new node
> cluster recover force
> 
> On node 1:
> ---
> collie cluster info
>   Cluster status: running
> 
> collie vdi check squeeze
>   FATAL: failed to read, Waiting for other nodes to join cluster
> ---
> 
> On the "new" node 3
> ---
> collie cluster info
>   Cluster status: Waiting for other nodes to join cluster
> 
> collie vdi list
>   <empty>
> ---

I think you can recover cluster correctly with the latest master
branch.

Thanks,

Kazutaka