Non-disruptive controller replacement in a cross-cluster redundancy configuration

Patent metadata

Field	Value
Publication number	US-2017075783-A1
Application number	US-201615361625-A
Country	US
Kind code	A1
Filing date	Nov 28, 2016
Priority date	Nov 13, 2014
Publication date	Mar 16, 2017
Grant date	—

Abstract

During a storage redundancy giveback from a first node to a second node following a storage redundancy takeover from the second node by the first node, the second node is initialized in part by receiving a node identification indicator from the second node. The node identification indicator is included in a node advertisement message sent by the second node during a giveback wait phase of the storage redundancy giveback. The node identification indicator includes an intra-cluster node connectivity identifier that is used by the first node to determine whether the second node is an intra-cluster takeover partner. In response to determining that the second node is an intra-cluster takeover partner, the first node completes the giveback of storage resources to the second node.

First claim

What is claimed is:

1 . A method, comprising:
during a storage redundancy giveback from a first node to a second node, initializing the second node following a storage redundancy takeover from the second node by the first node, wherein the initializing the second node comprises:
receiving from the second node a node identification indicator that includes an intra-cluster node connectivity identifier, and
determining whether the second node is an intra-cluster takeover partner for the first node based on the intra-cluster node connectivity identifier, and
in response to determining that the second node is an intra-cluster takeover partner for the first node, performing a giveback of storage resources from the first node to the second node.

2 . The method of claim 1 , wherein the determining whether the second node is an intra-cluster takeover partner comprises:
comparing the intra-cluster node connectivity identifier with intra-cluster connectivity information of the first node.

3 . The method of claim 1 , comprising:
receiving, during a giveback wait phase of the storage redundancy giveback, a node advertisement message sent by the second node, wherein the node advertisement message comprises the node identification indicator.

4 . The method of claim 1 , wherein the first node and the second node are coupled to storage devices via one or more switches, and wherein the intra-cluster node connectivity identifier comprises an identifier of a switch of the one or more switches.

5 . The method of claim 1 , wherein the initializing the second node comprises:
booting the second node, wherein the node identification indicator is received during a giveback phase of the booting.

6 . The method of claim 1 , wherein the first node and the second node are paired as partners and a third node and a fourth node are paired as partners, wherein the method comprises:
reading a copy of cross-cluster configuration information stored locally by the second node, wherein the first node, the third node, and the fourth node individually store local copies of the cross-cluster configuration information; determining that a storage controller device identifier specified by the copy of the cross-cluster configuration information is different than an identifier associated with a storage controller device within the second node; and sending a storage controller replacement message to the first node, the third node, and the fourth node to indicate that a storage controller device replacement was performed for the second node.

7 . The method of claim 6 , comprising:
determining that the first node, the third node, and the fourth node have modified locally stored copies of the cross-cluster configuration information based on the storage controller replacement message; and modifying the copy of the cross-cluster configuration information locally stored by the second node.

8 . The method of claim 6 , comprising:
instructing the first node, the third node, and the fourth node to replace controller device identifiers within locally stored copy of the cross-cluster configuration information using a controller device identifier comprised within the storage controller replacement message.

9 . A non-transitory machine readable medium having stored thereon instructions for performing a method comprising program code which when executed by at least one machine, causes a machine to:
during a storage redundancy giveback from a first node to a second node, initialize the second node following a storage redundancy takeover from the second node by the first node, wherein the initializing the second node comprises:
receive, during a giveback wait phase of the storage redundancy giveback, a node advertisement message, comprising a node identification indicator that includes an intra-cluster node connectivity identifier, from the second node, and
determine whether the second node is an intra-cluster takeover partner for the first node based on the intra-cluster node connectivity identifier, and
in response to determining that the second node is an intra-cluster takeover partner for the first node, perform a giveback of storage resources from the first node to the second node.

10 . The non-transitory machine readable medium of claim 9 , wherein the program code causes the machine to compare the intra-cluster node connectivity identifier received with intra-cluster connectivity information of the first node.

11 . The non-transitory machine readable medium of claim 9 , wherein the program code causes the machine to boot the second node, wherein the node advertisement message is received during a giveback phase of the booting.

12 . The non-transitory machine readable medium of claim 9 , wherein the first node and the second node are paired as partners and a third node and a fourth node are paired as partners, wherein the program code causes the machine to:
read a copy of cross-cluster configuration information stored locally by the second node, wherein the first node, the third node, and the fourth node individually store local copies of the cross-cluster configuration information; determine that a storage controller device identifier specified by the copy of the cross-cluster configuration information is different than an identifier associated with a storage controller device within the second node; and send a storage controller replacement message to the first node, the third node, and the fourth node to indicate that a storage controller device replacement was performed for the second node.

13 . The non-transitory machine readable medium of claim 12 , wherein the program code causes the machine to:
determine that the first node, the third node, and the fourth node have modified locally stored copies of the cross-cluster configuration information based on the storage controller replacement message; and modify the copy of the cross-cluster configuration information locally stored by the second node.

14 . The non-transitory machine readable medium of claim 12 , wherein the program code causes the machine to:
instruct the first node, the third node, and the fourth node to replace controller device identifiers within locally stored copy of the cross-cluster configuration information using a controller device identifier comprised within the storage controller replacement message.

15 . A computing device comprising:
a memory having stored thereon instructions for performing a method; and a processor coupled to the memory, the processor configured to execute the instructions to cause the processor to:
during a storage redundancy giveback from a first node to a second node, initialize the second node following a storage redundancy takeover from the second node by the first node, wherein the initializing the second node comprises:
receive, during a giveback phase of the second node being booted, a node identification indicator from the second node, wherein the node identification indicator includes an intra-cluster node connectivity identifier, and
determine whether the second node is an intra-cluster takeover partner for the first node based on the intra-cluster node connectivity identifier, and

in response to determining that the second node is an intra-cluster takeover partner for the first node, perform a giveback of storage resources from the first node to the second node

16 . The computing device of claim 15 , wherein the instructions to cause the processor to compare the intra-cluster node connectivity identifier with intra-cluster connectivity information of the first node.

Assignees

• Netapp Inc

Inventors

• Panara Amit V
• Patel Chaitanya V
• Keremane Hrishikesh
• Majmudar Pankti Vinay
• Unnikrishnan Santhosh
• Elpula Sravan Kumar
• Coatney Susan M

Classifications

• G06F3/067

  Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS] · CPC title

• G06F11/2071

  using a plurality of controllers · CPC title

• G06F11/2092 Primary

  Techniques of failing over between control units · CPC title

• G06F3/0632

  by initialisation or re-initialisation of storage systems · CPC title

• G06F11/2097

  maintaining the standby controller/processing unit updated (initialisation or re-synchronisation thereof G06F11/1658 and subgroups) · CPC title