Method and apparatus for restoring an instance of a storage server

Patent metadata

Field	Value
Publication number	US-9361192-B2
Application number	US-201213656060-A
Country	US
Kind code	B2
Filing date	Oct 19, 2012
Priority date	Oct 19, 2012
Publication date	Jun 7, 2016
Grant date	Jun 7, 2016

Abstract

Implementations of the present disclosure involve a system and/or method for gracelessly rebooting a storage appliance. The method and system includes a storage appliance in association with an event that will result in the loss of a state table from volatile memory that halts changes to at least one state table of the storage appliance. The state tables describe a plurality of file system states of one or more clients connected to the first storage appliance. The state information is written to a persistent memory of the storage appliance. The state table may then be repopulated using the state table information stored in persistent memory.

First claim

What is claimed is:

1. A computer-implemented method for restoring an instance of a first storage appliance comprising:
in association with an event that will result in a state table loss from volatile memory, halting changes to at least one state table of the first storage appliance, the at least one state table describing a plurality of file system states of one or more clients of the first storage appliance;
writing state table information for the first storage appliance from the at least one state table to a persistent memory of the first storage appliance;
transferring an Internet Protocol address of the first storage appliance to a second storage appliance;
transferring control of the persistent memory from the first storage appliance to the second storage appliance;
populating a second state table on the second storage appliance with the state table information for the first storage appliance, enabling the second storage appliance to take over the one or more clients of the first storage appliance;
halting changes to one or more entries in the second state table that were previously associated with the first storage appliance and describe a plurality of operating states of the one or more clients in communication with the Internet Protocol address;
writing information from the one or more entries of the second state table to the persistent memory;
returning the Internet Protocol address of the first storage appliance to the first storage appliance;
returning control of the persistent memory to the first storage appliance; and
repopulating the at least one state table using the state table information and the information from the one or more entries of the second state table stored in the persistent memory.

2. The computer-implemented method of claim 1 , wherein the event comprises a reboot command.

3. The computer-implemented method of claim 1 , wherein halting changes to the at least one state table of the first storage appliance comprises:
halting changes to the at least one state table for completed file system states;
completing at least one in-progress state; and
halting changes to the at least one state table corresponding to the at least one in-progress state when the in-progress state is completed.

4. The computer-implemented method of claim 1 , wherein the state table information comprises information required by a communication protocol for communications between the first storage appliance and the one or more clients.

5. The computer-implemented method of claim 1 , further comprising:
indicating in the persistent memory a successful write of the state table information when the state table information written to the persistent memory is complete for repopulating the at least one state table; and
aborting the repopulation of the at least one state table when the successful write of the state table information is missing.

6. A computer-implemented method for eliminating a grace period in a first storage appliance comprising:
starting a planned event that will result in a loss of state table information in volatile memory;
storing the state table information in a persistent memory, wherein the state information comprises information for recreating a state table;
transferring an Internet Protocol address from the first storage appliance to a second storage appliance;
transferring control of the persistent memory from the first storage appliance to the second storage appliance;
recreating the state table on the second storage appliance using the state information in the persistent memory, enabling the second storage appliance to take over states of the first storage appliance;
returning the Internet Protocol address back to the first storage appliance;
returning control of the persistent memory back to the first storage appliance;
disabling a storage appliance grace period that would otherwise allow the state table to be rebuilt from a plurality of clients of the first storage appliance; and
rebuilding the state table on the first storage appliance using the state information in the persistent memory.

7. The computer-implemented method of claim 6 , wherein the planned event comprises a reboot.

8. The computer-implemented method of claim 6 , wherein storing the state table information in the persistent memory comprises halting the modification of the state table.

9. The computer-implemented method of claim 8 , wherein halting the modification of the state table comprises:
evaluating whether a state in the state table may be halted; and
at least one of halting the state in the state table or waiting until the state may be halted and then halting the state in the state table.

10. The computer-implemented method of claim 6 , wherein the state table information comprises information required by a communication protocol for communications between the first storage appliance and at least one of the plurality of clients.

11. The computer-implemented method of claim 6 , further comprising:
indicating in the persistent memory a successful write of the state table information when the state table information written to the persistent is complete for repopulating the state table; and
aborting the repopulation of the state table when the successful write of the state table information is missing.

12. A system for eliminating a grace period comprising:
a first storage appliance comprising:
a first processor;
a first network interface with a first Internet Protocol (IP) address that allows for one or more clients to connect to the first storage appliance using a network;
a first volatile memory configured for storing a first state table comprising a state for each of the one or more clients connected to the first storage appliance; and
a first persistent memory configured to store data; and

a second storage appliance configured as a cluster capable of splitting operations of the one or more clients, the second storage appliance comprising:
a second processor,
a second network interface with a second Internet Protocol (IP) address that allows for the one or more clients to connect to the second storage appliance using the network;
a second volatile memory configured for storing a second state table comprising a state for each of the one or more clients connected to the second storage appliance; and
a second persistent memory configured to store data,
wherein the first volatile memory stores instructions for operation on the first processor, in association with an event that will result in the loss of the first state table, the first processor configured to:
store first state table information for the first state table in the first persistent memory;
transfer the first IP address from the first storage appliance to the second storage appliance;
transfer control of the first persistent memory from the first storage appliance to the second storage appliance;
rebuild the first state table on the second persistent memory using the state table information from the first persistent memory;
halt changes to one or more entries in the second state table that were previously a part of the first state table;
write information from the one or more entries in the second state table to the first state table information in the first persistent memory;
return the first IP address to the first storage appliance;
return control of the first persistent memory to the first storage appliance; and
rebuild the first state table on the first storage appliance with the first state table information; and
disable a storage appliance grace period.

13. The system for eliminating a grace period of claim 12 , wherein t

Assignees

• Oracle Int Corp

Inventors

• Smith Jeffrey Anderson
• Shivam Piyush
• Baker William Eugene

Classifications

• G06F11/1441 Primary

  Resetting or repowering · CPC title

• H04L49/65

  Re-configuration of fast packet switches · CPC title

• G06F3/06

  Digital input from, or digital output to, record carriers {, e.g. RAID, emulated record carriers or networked record carriers} · CPC title

• G06F9/4401

  Bootstrapping (security arrangements therefor G06F21/57) · CPC title

• H04L45/28

  using route fault recovery · CPC title