john-griffith's profile - activity

2013-12-13 15:30:07 -0600 answered a question Independent Database for Driver

The concept of each driver having it's own DB or DB table even is something we've tried to avoid and I'd like to continue to do so. There are a number of ways to map ID's between the backend device and Cinder.

One common method is to use the Cinder-UUID for the volume-name on the backend. Another option that some folks have used is to create a mapping using VolumeAdminMetadata

2013-02-08 00:09:29 -0600 answered a question Cinder create volume error

Is this running off of a version from Trunk/Master?

It looks like you've got the bits that default to trying to use the Cinder filter scheduler, but you don't have things setup. We inadvertantly submitted a change with the filter enabled by default, since then we've set the default back to just using the simple scheduler and ignoring the filters.

Maybe try a fresh pull from github?

2013-02-08 00:05:40 -0600 answered a question default size disk in an instance

This is setup by the flavor you chose when running "nova boot", based on what you describe here it doesn't sound like you're using/attaching a Cinder volume in this case.

You can get info on Flavors using "nova flavor-list" and it's also possible to create custom flavors. Teh Disk column is what you'll be interested in here.

2012-12-19 15:59:24 -0600 answered a question Basic confusion on cinder development setup

Hi Raj, On devstack we don't install packages etc, we launch each OpenStack process in a screen session. To access type 'screen -x', as the user that you ran stack.sh as.

I think that's what you're looking for, if now lemme know.