After my previous post where I introduced the Cohesity backup and secondary storage product, I want to share some experience in configuration and basic configuration using the Cohesity software.
Installation – from box to Cohesity GUI in 1,5 hour
First something on the installation, but not to much. Why? It only took me 1,5 hours including transportation to the datacenter (15 min drive), so there is not really much to it.
The steps you need to do are fairly easy. For the four node C2500 cluster we installed (one chassis) you need to install it in 2U rackspace, connect the dual power supplies, connect the eight 10Gbps network connectors (two per blade to preferable two different switches) and four 1 Gbps (IPMI) interfaces. All 10Gbps interfaces are connected to one vlan. IPMI interfaces are connected to a different network (vlan).
From a DHCP enabled network and a client with bonjour service the installation is then done via a webinterface. The wizard runs you to some steps, where you need the following information:
- Set the IP space for the each node (we use a /24 from a 10.0.0.0/8 IP space)
- Set the virtual IP space for the cluster (we use another /24 from a 10.0.0.0/8 IP space)
- Set the IPMI IP’s for KVM access.
- Set the general network settings: hostnames, cluster name, ntp and smtp
- Set the username and password settings for cluster access and KVM.
As said, everything runs via a web based wizard. The installation guide and worksheet is just over 2 pages, light and easy 🙂
Basic configuration on storage
To get your environment up and running you need to some configuration. The usable space in the cluster needs to be designated to one ore more partitions.
In my case I use one partition, since this partition can be divided later into customer specific allocation where you can choose to set compression, encryption and inline deduplication as an addtional setting.
After setiing up the partition, you need to set up a view box. The view box is Cohesity’s approach to create scalable storage slices from the partition. The view box is an administrative setting and allows you yo set specific characteristics:
- Deduplication: inline or after data has been written to the view box
- Compression: inline or after data has been written to the view box
- Cloud Tier: Setting up the cloud tier allows you offloading of cold (not often accessed data) to external targets like a NAS, public cloud storage from Azure, Google and AWS).
With the setup of this view box you are set to let backups land on the view box. You are also able to create file shares (views). This is where the actual fun begins. I will post more on this subject soon!