Node discovery and cluster formation for a secondary storage appliance
View Patent ↗Techniques are disclosed for identifying “free” compute nodes that have not been configured as a member of a secondary storage system. As disclosed, a node may automatically detect that it is a free node (i.e., one available to join a distributed computing cluster which provides the secondary storage system or to form a new cluster). When a node detects that it is not part of a cluster, the node may build an initial network configuration and begin advertising itself as a free node. Once the initial network state is available, the node may broadcast advertisements observed by a management tool and joined to the secondary storage system.
1. A method for indicating a presence of a node in a distributed computing cluster, the method comprising:
determining, by operation of one or more computer processors on the node, that the node has not been configured as a member of an existing distributed computing cluster; and
in response:
configuring a first network state on the node,
determining a node identifier (ID) for the node, and
broadcasting one or more messages advertising the node ID and a network address of a configuration interface hosted by the node, wherein the one or more messages indicate that the node has not been configured as a member of the existing distributed computing cluster and is available to join the existing distributed computing cluster, and wherein the network address is used to configure the node using zero configuration messages.
2. The method of claim 1 , wherein the existing distributed computing cluster is a secondary storage system providing backup services for a primary storage system.
3. The method of claim 2 , wherein the first network state is configured using a DHCP service to obtain a temporary network address for broadcasting the one or more messages.
4. The method of claim 2 , wherein the first network state is configured using a link-local network address assigned to a network interface on the node.
5. The method of claim 2 , further comprising:
receiving, over the configuration interface, configuration data used to configure a second network state on the node;
configuring the second network state on the node; and
joining the existing distributed computing cluster.
6. The method of claim 5 , wherein the configuration data includes a static IP address to assign to a network interface on the node.
7. The method of claim 2 , wherein the one or more messages are broadcast according to a zero configuration protocol, and wherein the node ID and the network address of the configuration interface are store in a TXT record of the messages.
8. A non-transitory computer-readable storage medium storing instructions that, when executed by a processor, cause the processor to perform an operation for indicating a presence of a compute node in an existing distributed computing cluster, the operation comprising:
determining that the node has not been configured as a member of the existing distributed computing cluster; and in response:
configuring a first network state on the node,
determining a node identifier (D) for the node, and
broadcasting one or more messages advertising the node ID and a network address of a configuration interface hosted by the node, wherein the one or more messages indicate that the node has not been configured as a member of the existing distributed computing cluster and is available to join the existing distributed computing cluster, and wherein the network address is used to configure the node using zero configuration messages.
9. The non-transitory computer-readable storage medium of claim 8 , wherein the existing distributed computing cluster is a secondary storage system providing backup services for a primary storage system.
10. The non-transitory computer-readable storage medium of claim 9 , wherein the first network state is configured using a DHCP service to obtain a temporary network address for broadcasting the one or more messages.
11. The non-transitory computer-readable storage medium of claim 9 , wherein the first network state is configured using a link-local network address assigned to a network interface on the node.
12. The non-transitory computer-readable storage medium of claim 9 , wherein the operation further comprises:
receiving, over the configuration interface, configuration data used to configure a second network state on the node;
configuring the second network state on the node; and
joining the existing distributed computing cluster.
13. The non-transitory computer-readable storage medium of claim 12 , wherein the configuration data includes a static IP address to assign to a network interface on the node.
14. The non-transitory computer-readable storage medium of claim 9 , wherein the one or more messages are broadcast according to a zero configuration protocol, and wherein the node ID and the network address of the configuration interface are store in a TXT record of the messages.
15. A system comprising:
a processor; and
a memory storing one or more applications, which, when executed on the processor perform an operation for indicating a presence of a compute node in an existing distributed computing duster, the operation comprising:
determining that the node has not been configured as a member of the existing distributed computing duster, and in response:
configuring a first network state on the node;
determining a node identifier (ID) for the node; and
broadcasting one or more messages advertising the node ID and a network address of a configuration interface hosted by the node, wherein the one or more messages indicate that the node has not been configured as a member of the existing distributed computing duster and is available to join the existing distributed computing duster, and wherein the network address is used to configure the node using zero configuration messages.
16. The system of claim 15 , wherein the existing distributed computing cluster is a secondary storage system providing backup services for a primary storage system.
17. The computer-readable storage medium of claim 15 , wherein the first network state is configured using a DHCP service to obtain a temporary network address for broadcasting the one or more messages.
18. The system of claim 15 , wherein the first network state is configured using a link-local network address assigned to a network interface on the node.
19. The system of claim 15 , wherein the operation further comprises:
receiving, over the configuration interface, configuration data used to configure a second network state on the node;
configuring the second network state on the node; and
joining the existing distributed computing cluster.
20. The system of claim 19 , wherein the configuration data includes a static IP address to assign to a network interface on the node.
21. The system of claim 15 , wherein the one or more messages are broadcast according to a zero configuration protocol, and wherein the node ID and the network address of the configuration interface are store in a TXT record of the messages.
22. A method for managing duster membership in an existing distributed computing duster, the method comprising:
detecting a message broadcast by a compute node, wherein the message indicates that the compute node has not been configured as a member of the existing distributed computing cluster and wherein the compute node broadcasts the message using a first configured network state on the compute node;
determining from the message, a node ID and a network address of a configuration interface hosted by the compute node; and
sending the compute node, over the configuration interface, configuration data, wherein the compute node applies the configuration data on the compute node to configure a second network state on the compute node, wherein the compute joins the existing distributed computing duster after configuring the second network state, and wherein the existing distributed computing duster is a secondary storage system providing backup services for a primary storage system, and wherein the network address is used to configure the compute node using zero configuration messages.