In today’s entry I would like to introduce quorum and epsilon concepts. Both are related with cluster health, needed in case of potential communications and connectivity issues.
When a cluster is in quorum, it means that majority of nodes are healthy and can communicate with each other. In other words, quorum is a precondition for a functioning cluster. When quorum is lost, the cluster does not have the ability to accomplish normal operations. Continue reading
In my previous entry I have briefly describe how you can protect SVM root volume by using load-sharing mirrors. (The post can be found here: NetApp cDOT – SVM root volume protection). If you haven’t read it, and you are not sure what are the consequences of root volume not being accessible, I would encourage to give the article a try. Long story short, each SVM (Storage Virtual Machine, a.k.a vserver), has it’s own namespace. Root volume is a root (/) path of SVM. If SVM root volume becomes unavailable all NAS (CIFS/NFS) clients will lose access to all shares from that particular SVM. If you want to read a little bit more into namespace concept, check out my other entry: NetApp cDOT – Namespace, junction path.
Remember: when a SVM root volume became unavailable, it will be disruptive for all NAS clients! Never “experiment” on production environment.
In this entry I would like briefly describe how to protect the SVM (vserver) root volume by creating load-sharing mirrors on every node of a cluster. If you are unfamiliar with SVMs, check out my article NetApp cDOT – what is SVM or vserver ?. Every vserver has it’s root volume, which is an entry point fo a namespace provided by that SVM (You can read more about namespaces in NetApp cDOT – Namespace, junction path entry). Continue reading