Indicating an error within a dispersed storage network

• US 9,092,386 B2
• Filed: 06/18/2013
• Issued: 07/28/2015
• Est. Priority Date: 04/26/2010
• Status: Active Grant

• Current Assignee: Pure Storage, Inc.
• Original Assignee: Cleversafe Incorporated (International Business Machines Corporation)
• Inventors: Baptist, Andrew, Leggette, Wesley, Resch, Jason K.
• Primary Examiner(s): Leibovich, Yair
• Application Number:
  US13/920,708
• Publication Number:
  US 20130283094A1
• Time in Patent Office: 770 Days
• Field of Search: 714/6.2, 714/47.2
• US Class Current: 1/1
• CPC Class Codes: G06F 11/1666 where the redundant component is memory or memory area
• CPC Class Codes: H04L 1/24 Testing correct operation
• CPC Class Codes: H04L 63/08 for authentication of entities cryptographic mechanisms or cryptographic arrangements for entity authentication H04L9/32
• CPC Class Codes: H04L 63/123 received data contents, e.g. message integrity
• CPC Class Codes: H04L 67/1097 for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
• CPC Class Codes: H04L 67/51 Discovery or management thereof, e.g. service location protocol [SLP] or web services
• CPC Class Codes: H04L 69/40 for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection management of faults, events, alarms or notifications in data switching networks H04L41/06

Abstract

A method begins by a processing sending access requests to a threshold number of storage units of a dispersed storage network (DSN) and receiving, from each of at least some of the threshold number of storage units, an access response or an error response. When one error response is received from a storage unit of the at least some of the threshold number of storage units in response to one of the access requests, the method continues with the processing module ascertaining a likely cause for an error corresponding to the error response. The method continues with the processing module determining, based on the likely cause for the error, whether to resend the one of the access requests to the storage unit, issue a modified access request to the storage unit, or send the one of the access requests to another storage unit.

Claims

1. A method for execution by one or more processing modules of one or more requesting devices of a dispersed storage network (DSN), the method comprises:

sending access requests to a threshold number of storage units of the DSN;

receiving, from each of at least some of the threshold number of storage units, an access response or an error response; and

when one error response is received from a storage unit of the at least some of the threshold number of storage units in response to one of the access requests;

ascertaining a likely cause for an error corresponding to the error response; and

determining, based on the likely cause for the error, whether to;

resend the one of the access requests to the storage unit;

issue a modified access request to the storage unit;

or

send the one of the access requests to another storage unit.

2. The method of claim 1 further comprises:

when more than one error responses are received from more than one storage units of the at least some of the threshold number of storage units in response to more than one of the access requests;

for each of the more than one error responses, ascertaining the likely cause for the error corresponding to the error response to produce a set of likely causes;

interpreting the set of likely causes to determine whether the set of likely causes includes a common cause or different causes; and

when the set of likely causes includes the common cause, determining, based on the common cause, whether to;

resend the more than one of the access requests to the more than one storage units;

issue more than one modified access requests to the more than one storage units;

or

send the more than one of the access requests to more than one other storage units.

3. The method of claim 2 further comprises one or more of:

when the set of likely causes includes the different causes, issuing a system maintenance error;

when the set of likely causes includes the different causes;

determining whether the different causes are time dependent errors, where, by waiting for synchronization of information to occur across the DSN resolves the time dependent errors; and

when the different causes are the time dependent errors, waiting a predetermined period of time and then resending the more than one of the access requests to the more than one storage units; and

when the set of likely causes includes the different causes;

individually resolving the different causes; and

when the different causes are resolved, resending the more than one of the access requests to the more than one storage units.

4. The method of claim 1, wherein the ascertaining the likely cause for the error comprises:

interpreting the error response to identify an error class, wherein the error class includes one or more of;

an unknown operation error condition;

an invalid name error condition;

an operation conflict error condition;

a check error condition;

an internal storage unit error condition;

an unauthorized request error condition; and

an invalid address range error condition.

5. The method of claim 4 further comprises:

interpreting the error response to identify corresponding error detail, wherein the corresponding error detail includes one or more;

when the error class is the unknown operation error condition, identity of the unknown operation;

when the error class is the invalid name error condition, a slice name of an encoded data slice, wherein the slice name is invalid with respect to the storage unit;

when the error class is the operation conflict error condition, identity of a conflicting operation and identity of an issuer of the conflicting operation;

when the error class is the check error condition, an indication that a revision level of the one of the access requests does not correspond to an expected revision level by the storage unit;

when the error class is the unauthorized request error condition, an indication that one or more requesting devices is not authorized; and

when the error class is the invalid address range error condition, an indication that a requested DSN address is outside of a range of DSN addresses allocated to the storage unit.

6. The method of claim 1 further comprises:

determining the likely cause to be a timing delay between the one or more requesting devices and the storage unit updating DSN memory addressing allocation information;

waiting for the timing delay to pass; and

resending the one of the access requests to the storage unit.

7. The method of claim 1 further comprises:

determining the likely cause to be an assigned DSN address of the one of the access requests to be outside an allocated range of DSN addresses for the storage unit;

modifying the one of the access requests to include a DSN address within the allocated range of DSN addresses to produce the modified access request; and

sending the modified access request to the storage unit.

8. The method of claim 1 further comprises:

determining that the likely cause to be a conflict between DSN memory allocation information of the one or more requesting devices and the DSN memory allocation information of the storage unit;

selecting the other storage unit; and

sending the one of the access requests to the other storage unit.

9. The method of claim 1, wherein the threshold number of storage units comprises one of:

a write threshold number of storage units when the access requests are write access requests;

a read threshold number of storage units when the access requests are read access requests; and

a decode threshold number of storage units when the access requests are the read access requests.

10. A computer of one or more computers of a dispersed storage network (DSN), the computer comprises:

an interface;

a memory; and

a processing module operably coupled to the interface and to the memory, where the processing module is operable to;

send, via the interface, access requests to a threshold number of storage units of the DSN;

receive, from each of at least some of the threshold number of storage units via the interface, an access response or an error response; and

when one error response is received from a storage unit of the at least some of the threshold number of storage units in response to one of the access requests;

ascertain a likely cause for an error corresponding to the error response; and

determine, based on the likely cause for the error, whether to;

resend the one of the access requests to the storage unit;

issue a modified access request to the storage unit;

or

send the one of the access requests to another storage unit.

11. The computer of claim 10 further comprises:

the processing module is further operable to;

when more than one error responses are received from more than one storage units of the at least some of the threshold number of storage units in response to more than one of the access requests;

for each of the more than one error responses, ascertain the likely cause for the error corresponding to the error response to produce a set of likely causes;

interpret the set of likely causes to determine whether the set of likely causes includes a common cause or different causes; and

when the set of likely causes includes the common cause, determine, based on the common cause, whether to;

resend the more than one of the access requests to the more than one storage units;

issue more than one modified access requests to the more than one storage units;

or

send the more than one of the access requests to more than one other storage units.

12. The computer of claim 11 further comprises one or more of:

when the set of likely causes includes the different causes, the processing module is further operable to issue a system maintenance error;

when the set of likely causes includes the different causes, the processing module is further operable to;

determine whether the different causes are time dependent errors, where, by waiting for synchronization of information to occur across the DSN resolves the time dependent errors; and

when the different causes are the time dependent errors, wait a predetermined period of time and then resend, via the interface, the more than one of the access requests to the more than one storage units; and

when the set of likely causes includes the different causes, the processing module is further operable to;

individually resolve the different causes; and

when the different causes are resolved, resend, via the interface, the more than one of the access requests to the more than one storage units.

13. The computer of claim 10, wherein the processing module functions to ascertain the likely cause for the error by:

interpreting the error response to identify an error class, wherein the error class includes one or more of;

an unknown operation error condition;

an invalid name error condition;

an operation conflict error condition;

a check error condition;

an internal storage unit error condition;

an unauthorized request error condition; and

an invalid address range error condition.

14. The computer of claim 13 further comprises:

the processing module further functions to interpret the error response to identify corresponding error detail, wherein the corresponding error detail includes one or more;

when the error class is the unknown operation error condition, identity of the unknown operation;

when the error class is the invalid name error condition, a slice name of an encoded data slice, wherein the slice name is invalid with respect to the storage unit;

when the error class is the operation conflict error condition, identity of a conflicting operation and identity of an issuer of the conflicting operation;

when the error class is the check error condition, an indication that a revision level of the one of the access requests does not correspond to an expected revision level by the storage unit;

when the error class is the unauthorized request error condition, an indication that one or more requesting devices is not authorized; and

when the error class is the invalid address range error condition, an indication that a requested DSN address is outside of a range of DSN addresses allocated to the storage unit.

15. The computer of claim 10 further comprises:

the processing module is further operable to;

determine the likely cause to be a timing delay between the one or more requesting devices and the storage unit updating DSN memory addressing allocation information;

wait for the timing delay to pass; and

resend, via the interface, the one of the access requests to the storage unit.

16. The computer of claim 10 further comprises:

the processing module is further operable to;

determine the likely cause to be an assigned DSN address of the one of the access requests to be outside an allocated range of DSN addresses for the storage unit;

modify the one of the access requests to include a DSN address within the allocated range of DSN addresses to produce the modified access request; and

send, via the interface, the modified access request to the storage unit.

17. The computer of claim 10 further comprises:

the processing module is further operable to;

determine that the likely cause to be a conflict between DSN memory allocation information of the one or more requesting devices and the DSN memory allocation information of the storage unit;

select the other storage unit; and

send, via the interface, the one of the access requests to the other storage unit.

18. The computer of claim 10, wherein the threshold number of storage units comprises one of:

a write threshold number of storage units when the access requests are write access requests;

a read threshold number of storage units when the access requests are read access requests; and

a decode threshold number of storage units when the access requests are the read access requests.