Method for adding disks in a raid system having a protection pool of storage units
A method of adding a disk in a redundant array of independent disks (RAID) system with a protection pool of storage units includes receiving a request to add a new physical disk to a storage cluster and partitioning the new physical disk into a set of physical extents. The method further includes allocating one or more physical extents of the set of physical extents of the new physical disk to at least one array of physical extents, wherein the array of physical extents is distributed across physical disks of the storage cluster.
1. A method comprising:
receiving a request to add a new physical disk to a storage cluster;
partitioning, by a processing device, the new physical disk into one or more partitions, each partition comprising a plurality of physical extents;
allocating, by the processing device, one or more physical extents of the plurality of physical extents of the new physical disk to at least one array of physical extents, wherein the array of physical extents is distributed across a plurality of physical disks of the storage cluster;
generating a physical extent map for the plurality of physical extents of the new physical disk, wherein each one of the one or more partitions includes metadata storing mapping information from the physical extent map, the mapping information including mapping information between the plurality of physical extents of each partition and corresponding locations on the new physical disk;
adding the plurality of physical extents to a list of free physical extents available to be allocated to an array;
determining whether a local disk manager is available to manage the new physical disk, wherein determining whether the local disk manager is available comprises:
issuing a request to a plurality of local disk managers in the storage cluster for a current state of each of the plurality of local disk managers, and
receiving a reply to the request from one or more of the plurality of local disk managers, wherein the local disk manager being available is based on the reply; and
in response to determining based on the reply, that the local disk manager is available, adding the new physical disk to the local disk manager.
2. The method of claim 1 , further comprising:
in response to determining, based on the reply, that the local disk manager is not available:
creating a new local disk manager to manage the new physical disk; and
adding the new physical disk to the new local disk manager.
3. The method of claim 2 , further comprising:
registering the new local disk manager with one or more nodes of the storage cluster to make the plurality of physical extents of the new physical disk available to be allocated to arrays of the one or more nodes.
4. The method of claim 1 , wherein allocating the one or more physical extents of the plurality of physical extents of the new physical disk to at least one array of physical extents comprises:
determining to migrate one or more source physical extents of another physical disk to the one or more physical extents of the new physical disk, wherein the one or more source physical extents are associated with the at least one array;
replacing the one or more source physical extents in the at least one array with the one or more physical extents of the new physical disk; and
migrating data from the one or more source physical extents to the one or more physical extents of the new physical disk.
5. A system comprising:
a processor; and
a memory to store instructions, which when executed by the processor, cause the processor to perform operations comprising:
receiving a request to add a new physical disk to a storage cluster;
partitioning the new physical disk into one or more partitions, each partition comprising a plurality of physical extents;
allocating one or more physical extents of the plurality of physical extents of the new physical disk to at least one array of physical extents, wherein the array of physical extents is distributed across a plurality of physical disks of the storage cluster;
generating a physical extent map for the plurality of physical extents of the new physical disk, wherein one of the one or more partitions includes metadata storing mapping information from the physical extent map, the mapping information including mapping information between the plurality of physical extents of each partition and corresponding locations on the new physical disk;
adding the plurality of physical extents to a list of free physical extents available to be allocated to an array;
determining whether a local disk manager is available to manage the new physical disk, wherein determining whether the local disk manager is available comprises:
issuing a request to a plurality of local disk managers in the storage cluster for a current state of each of the plurality of local disk managers; and
receiving a reply to the request from one or more of the plurality of local disk managers, wherein the local disk manager being available is based on the reply; and
in response to determining based on the reply, that the local disk manager is available, adding the new physical disk to the local disk manager.
6. The system of claim 5 , further comprising:
in response to determining based on the reply, that the local disk manager is not available:
creating a new local disk manager to manage the new physical disk; and
adding the new physical disk to the new local disk manager.
7. The system of claim 6 , further comprising:
registering the new local disk manager with one or more nodes of the storage cluster to make the plurality of physical extents of the new physical disk available to be allocated to arrays of the one or more nodes.
8. The system of claim 5 , wherein allocating the one or more physical extents of the plurality of physical extents of the new physical disk to at least one array of physical extents comprises:
determining to migrate one or more source physical extents of another physical disk to the one or more physical extents of the new physical disk, wherein the one or more source physical extents are associated with the at least one array;
replacing the one or more source physical extents in the at least one array with the one or more physical extents of the new physical disk; and
migrating data from the one or more source physical extents to the one or more physical extents of the new physical disk.
9. A non-transitory machine-readable medium having instructions stored therein, which when executed by a processor, cause the processor to perform operations comprising:
receiving a request to add a new physical disk to a storage cluster;
partitioning, by the processor, the new physical disk into one or more partitions, each partition comprising a plurality of physical extents;
allocating, by the processor, one or more physical extents of the plurality of physical extents of the new physical disk to at least one array of physical extents, wherein the array of physical extents is distributed across a plurality of physical disks of the storage cluster;
generating a physical extent map for the plurality of physical extents of the new physical disk, wherein one of the one or more partitions includes metadata storing mapping information from the physical extent map, the mapping information including mapping information between the plurality of physical extents of each partition and corresponding locations on the new physical disk;
adding the plurality of physical extents to a list of free physical extents available to be allocated to an array;
determining whether a local disk manager is available to manage the new physical disk, wherein determining whether the local disk manager is available comprises:
issuing a request to a plurality of local disk managers in the storage cluster for a current state of each of the plurality of local disk managers; and
receiving a reply to the request from one or more of the plurality of local disk managers, wherein the local disk manager being available is based on the reply; and
in response to determining based on the reply, that the local disk manager is available, adding the new physical disk to the local disk manager.
10. The non-transitory machine-readable medium of claim 9 , further comprising: in response to determining, based on the reply, that the local disk manager 1s not available: creating a new local disk manager to manage the new physical disk; and adding the new physical disk to the new local disk manager.
11. The non-transitory machine-readable medium of claim 10 , further comprising:
registering the new local disk manager with one or more nodes of the storage cluster to make the plurality of physical extents of the new physical disk available to be allocated to arrays of the one or more nodes.