This is a distributed storage system for the Linux platform. It consists of a kernel module, several userspace management applications and some shell scripts. DRBD is normally used on high availability (HA) clusters.
In very simple terms what happens is DRBD takes over the data, writes it to the local disk and sends it to the other host. Then on that second host, it takes it to the disk there. However, the reads are always carried out locally.
The main point is carrying out the work without the user being affected of any failure.If the primary node fails, the services will be carried out on the secondary host. If the failed node, comes again it has to take the backups from the secondary node.The resynchronization will happen on the failed nodes.
DRBD works on top of block devices.
Fully synchronous
Mirroring can be done tightly coupled. That means that the file system on the active node is notified that the writing of the block was finished only when the block made it to both disks of the cluster.Synchronous mirroring is the right choice for HA clusters.
Asynchronous
Asynchronous means that the entity that issued the write requests is informed about completion as soon as the data is written to the local disk.Asynchronous mirroring is necessary to build mirrors over long distances.
DRBD also offers some advantages over shared cluster storage.
- Shared storage resources usually introduce a single point of failure in the cluster setup. In DRBD, issues like this don't exist as the cluster resource data is replicated.
-Shared storage resources may face situations where both cluster nodes are still alive, but lose all network connectivity between them. In this kind of situation, each cluster node will assume that it is the only node in the cluster. DRBD reduces this problem by keeping two replicated sets.
No comments:
Post a Comment