Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

Version 1 Current »

The Fedora Commons data store can be moved and resized with fairly minimal consequences (tested 1/4/16 by moving staging's fedora datastore to a new disk). See the steps below:

  1. Create a new AWS volume(s) to hold the data (if we decide to do a RAID array we'll need to make a group of volumes, then mount them and make the array).
    1. MAKE SURE THE VOLUME IS IN THE SAME REGION AS THE INSTANCE
  2. Transfer the data to the new volume(s). This can be done with a snapshot (see http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/ebs-expand-volume.html) though that has not been tested yet.
    1. If not using snapshots,  we'll need to attach and mount the volume.
    2. Attach the volume in the web console by going to tasks->attach or use the command (ec2-attach-volume volume_id --instance instance_id --device device)
    3. Copy the files over (cp -a oldlocation(s) new location) -a should preserve all attributes and links.
      1. This must be done when nobody is changing or adding data to avoid missing content 
    4. Unmount the old storage volume
    5. Mount the new storage volume in the old volume's location (currently /opt/fedora-data)
    6. Restart tomcat (this will restart Fedora)
    7. Check to make sure the repository is still working by looking at a few items and doing a test search or two.
    8. Delete the old volume.
    9. Set the new volume up for snapshots (this part has not yet been worked out, but we should be able to reassign the volume ID, that would require unmounting and remounting the new storage volume)

 

  • No labels