Virtualized file server data sharing
In one embodiment, a system for managing a virtualization environment includes a set of host machines, each of which includes a hypervisor, virtual machines, and a virtual machine controller, and a first virtualized file server configured to receive a request to access a storage item located at a second virtualized file server, determine that the storage item is designated as being accessible by other virtualized file servers, identify an FSVM of the second virtualized file server at which the storage item is located, and forward the request to the FSVM of the second virtualized file server. The storage item may be designated as being accessible by other virtualized file servers when the storage item is associated with a predetermined tag value indicating that the storage item is shared among virtualized file servers. The predetermined tag value may be stored in a sharding map in association with the storage item.
1. At least one non-transitory computer readable medium encoded with instructions which, when executed, cause at least one processor to perform operations, the operations comprising:
detecting failure of a first file server virtual machine (FSVM) of a virtualized file server (VFS), the first FSVM hosted on a first computing node of a computing node cluster, the first FSVM configured to receive a first storage access operation sent to a first internet protocol (IP) address to facilitate the first storage access operation between a client and a virtualized file system of the VFS to access a first storage item of a namespace of storage items presented by the first FSVM, the namespace implemented using storage devices of a storage pool distributed across the computing node cluster; and
reassigning the first IP address corresponding to the first FSVM to a second FSVM of the VFS, the second FSVM hosted on a second computing node of the computing node cluster, the second FSVM configured to receive a second storage access operation sent to a second IP address to access a second storage item of the virtualized file system of the VFS, the second storage item being of the namespace of storage items, wherein the reassigning of the first IP address comprises configuring the second FSVM to receive a third storage access operation sent to the first IP address to access the first storage item of the namespace.
2. The computer readable medium of claim 1 , wherein the operations further comprise processing the first storage access operation to determine at least one storage location of the first storage item in the storage pool using a map.
3. The computer readable medium of claim 2 , wherein the reassigning of the first IP address further comprises configuring the second computing node to process the third storage access operation to determine the at least one storage location of the first storage item in the storage pool using the map.
4. The computer readable medium of claim 3 , wherein the reassigning of the first IP address further comprises configuring the second computing node to translate the third storage access operation to a block request at a filesystem hosted on the second computing node, the block request configured to access one or more data blocks of the first storage item at the at least one storage location in the storage pool.
5. The computer readable medium of claim 2 , wherein the operations further comprise translating the first storage access operation to a block request at a filesystem hosted on the first computing node, the block request configured to access one or more data blocks of the first storage item at the at least one storage location in the storage pool.
6. The computer readable medium of claim 1 , wherein the first storage access operation is issued by a user virtual machine running client software hosted on the first computing node.
7. The computer readable medium of claim 1 , wherein the operations further comprise selecting the second FSVM from a plurality of FSVMs hosted on computing nodes of the computing node cluster for reassignment of the first IP address.
8. The computer readable medium of claim 7 , wherein the operations further comprise selecting the second FSVM based on a failover path for the first FSVM identifying the second FSVM.
9. The computer readable medium of claim 7 , wherein the operations further comprise selecting the second FSVM based on a comparison between a number of outstanding or pending requests to the second FSVM and a number of outstanding or pending requests to at least another of the plurality of FSVMs.
10. The at least one non-transitory computer readable medium of claim 2 , wherein the namespace is implemented using storage devices of the storage pool, including at least one local storage device of the first computing node and at least one local storage device of the second computing node.
11. The at least one non-transitory computer readable medium of claim 2 , wherein the second FSVM is further configured to present the namespace of storage items.
12. The at least one non-transitory computer readable medium of claim 2 , wherein the first FSVM is configured to interact with a hypervisor to facilitate the first storage access operation.
13. The computer readable medium of claim 1 , wherein said detecting comprises detecting failure of the first FSVM or a first storage controller, the first FSVM and the first storage controller hosted on the first computing node; and
wherein the second FSVM and a second storage controller are hosted on the second computing node.
14. A distributed computing system comprising:
a computing node cluster including a first computing node and a second computing node;
a storage pool including storage devices distributed across the computing node cluster;
a first file server virtual machine (FSVM) of a virtualized file server (VFS), the first FSVM hosted on the first computing node, the first FSVM configured to receive a first storage access operation sent to a first internet protocol (IP) address to facilitate the first storage access operation between a client and a virtualized file system of the VFS to access a first storage item of a namespace of storage items presented by the first FSVM, the namespace implemented using the storage pool;
a second FSVM of the VFS, the second FSVM hosted on the second computing node, the second FSVM configured to receive a second storage access operation sent to a second IP address to access a second storage item of the virtualized file system of the VFS, the second storage item being of the namespace of storage items; and
a health monitoring service running on the first computing node, the health monitoring service configured to reassign, responsive to detection of failure of the first FSVM, the first IP address corresponding to the first FSVM to the second FSVM, wherein the reassigning of the first IP address comprises configuring the second FSVM to receive a third storage access operation sent to the first IP address to access the first storage item of the namespace.
15. The distributed computing system of claim 14 , wherein the first computing node is configured to process the first storage access operation to determine at least one storage location of the first storage item in the storage pool using a map.
16. The distributed computing system of claim 15 , wherein the reassigning of the first IP address comprises configuring the second computing node to process the third storage access operation to determine the at least one storage location of the first storage item in the storage pool using the map.
17. The distributed computing system of claim 16 , wherein the reassigning of the first IP address further comprises configuring the second computing node to translate the third storage access operation to a block request at a filesystem hosted on the second computing node, the block request configured to access one or more data blocks of the first storage item at the at least one storage location in the storage pool.
18. The distributed computing system of claim 15 , wherein the first computing node is further configured to translate the first storage access operation to a block request at a filesystem hosted on the first computing node, the block request configured to access one or more data blocks of the first storage item at the at least one storage location in the storage pool.
19. The distributed computing system of claim 15 , wherein the first storage access operation is issued by a user virtual machine running client software hosted on the first computing node.
20. The distributed computing system of claim 15 , wherein the health monitoring service is further configured to select the second FSVM from a plurality of FSVMs hosted on computing nodes of the computing node cluster for reassignment of the first IP address.
21. The distributed computing system of claim 20 , wherein the second FSVM is selected based on a failover path for the first FSVM identifying the second FSVM.
22. The distributed computing system of claim 20 , wherein the second FSVM is selected based on a comparison between a number of outstanding or pending requests to the second FSVM and a number of outstanding or pending requests to at least another of the plurality of FSVMs.
23. The distributed computing system of claim 14 , wherein the namespace is implemented using storage devices of the storage pool, including at least one local storage device of the first computing node and at least one local storage device of the second computing node.
24. The distributed computing system of claim 14 , wherein the second FSVM is further configured to present the namespace of storage items.
25. The distributed computing system of claim 14 , wherein the first FSVM is configured to interact with a hypervisor to facilitate the first storage access operation.
26. The distributed computing system of claim 14 , further comprising a first storage controller hosted on the first computing node and a second storage controller hosted on the second computing node.
27. A method comprising:
detecting failure of a first file server virtual machine (FSVM) of a virtualized file server (VFS), the first FSVM hosted on a first computing node of a computing node cluster, the first FSVM configured to receive a first storage access operation sent to a first internet protocol (IP) address to facilitate the first storage access operation between a client and a virtualized file system of the VFS to access a first storage item of a namespace of storage items presented by the first FSVM, the namespace implemented using storage devices of a storage pool distributed across the computing node cluster; and
reassigning the first IP address to a second FSVM, the second FSVM of the VFS, the second FSVM hosted on a second computing node of the computing node cluster, the second FSVM configured to receive a second storage access operation sent to a second IP address to access a second storage item of the virtualized file system of the VFS, the second storage item being of the namespace of storage items, wherein the reassigning of the first IP address comprises configuring the second FSVM to receive a third storage access operation sent to the first IP address to access the first storage item of the namespace.
28. The method of claim 27 , further comprising processing the first storage access operation to determine at least one storage location of the first storage item in the storage pool using a map.
29. The method of claim 28 , wherein the reassigning of the first IP address comprises configuring the second computing node to process the third storage access operation to determine the at least one storage location of the first storage item in the storage pool using the map.
30. The method of claim 29 , wherein the reassigning of the first IP address further comprises configuring the second computing node to translate the third storage access operation to a block request at a filesystem hosted on the second computing node, the block request configured to access one or more data blocks of the first storage item at the at least one storage location in the storage pool.
31. The method of claim 28 , further comprising translating the first storage access operation to a block request at a filesystem hosted on the first computing node, the block request configured to access one or more data blocks of the first storage item at the at least one storage location in the storage pool.
32. The method of claim 27 , wherein the first storage access operation is received from a user virtual machine running client software hosted on the first computing node.
33. The method of claim 27 , further comprising selecting the second FSVM from a plurality of FSVMs hosted on computing nodes of the computing node cluster for reassignment of the first IP address.
34. The method of claim 33 , further comprising selecting the second FSVM based on a failover path for the first FSVM identifying the second FSVM.
35. The method of claim 33 , further comprising selecting the second FSVM based on a comparison between a number of outstanding or pending requests to the second FSVM and a number of outstanding or pending requests to at least another of the plurality of FSVMs.
36. The method of claim 27 , wherein the namespace is implemented using storage devices of the storage pool, including at least one local storage device of the first computing node and at least one local storage device of the second computing node.
37. The method of claim 27 , wherein the second FSVM is further configured to present the namespace of storage items.
38. The method of claim 27 , wherein the first FSVM is configured to interact with a hypervisor to facilitate the first storage access operation.
39. The method of claim 27 , wherein said detecting comprises detecting failure of the first FSVM or a first storage controller, the first FSVM and the first storage controller hosted on the first computing node; and
wherein the second FSVM and a second storage controller are hosted on the second computing node.