Configuring storage is one of the most important parts of setting up a Red Hat Enterprise Virtualization deployment. However, if you don't do it correctly the first time, storage configuration problems are among the most difficult to correct. Let's take a look at how you can avoid some of the most common storage related configuration problems in RHEV.
The first step to be sure all the hosts in your RHEV data center group are connected to the same storage. It is possible to use different types of storage in one RHEV environment, but not within the same data center group. If for instance you need a couple of hosts to use the iSCSI storage area network (SAN), and other hosts to use
More resources on RHEV
RHEV management guide
Connecting shared storage to RHEV
Understanding RHEV live migration
When setting up a RHEV environment, you’ll create a data center group and assign it a specific type of storage, such as Network File System (NFS), iSCSI or local storage. Once you’ve selected a specific type of storage, you cannot change it later, so make that you make the right choice while setting up the data center group.
To install VMs, you need access to ISO files, which is accomplished by creating an ISO storage repository that you access using an NFS. It doesn’t matter where this NFS share resides, as long as all hosts can access it. Putting ISO files in this ISO storage repository can be a bit dull. There is no graphical interface to upload ISO files, so you’ll need to use the RHEV ISO uploader tool, using the rhev-iso-uploader command. The tool isn’t exactly user friendly and it is slow.
The uploader tool automatically creates a subdirectory on the NFS share and uploads the first ISO there. Once the uploader tool creates the structure for uploading ISO files, you can manually copy the other ISO files to the ISO repository to make them available to install VMs.
Even if it’s not possible to change the default storage type for a data center group, it is possible to change the default storage type in your RHEV environment with a workaround. This workaround involves setting up a second data center group with the storage type you want to use. Next, you’ll need to assign RHEV hosts to the new data center group. You can do this by putting a host in maintenance mode and editing its properties to assign it to the other data center group. However, this workaround is not without its drawbacks. When you change the storage type in a data center group all, existing VMs in that group will lose access to files stored on the original storage type. Your best option is to ensure you select the correct storage system the first time to avoid having to rely on a workaround.
This was first published in November 2012