Computerised storage system comprising replaceable units for managing testing of replacement units

#The present invention relates to a method, apparatus or software for use in a computerized storage system, comprising one or more replaceable units, for managing testing of one or more replacement units.

#Computer systems are commonly built from an interconnected set of distinct components. When units become faulty they need to be replaced with a correctly functioning unit. Some computer systems, such as high availability storage systems, have built-in redundancy in which critical units are duplicated. Critical units may be units such as disk arrays, interface cards and connection cables, any of which may become faulty and require replacement. In such systems, the redundancy provided by duplicated units can be used to provide continuous service while a faulty unit is replaced. For example, if faults or errors are detected that suggest a connection cable needs to be replaced, user input/output (I/O) is diverted to a duplicate unit while the replacement is performed. Once the replacement unit is in place it is made available for use with user I/O. However, faults or errors may recur if they were misdiagnosed, a replacement unit is faulty or if more that one unit is contributing to the fault. In addition, if a replaced unit does not correct the detected faults or errors because user I/O may be disrupted.

#The present invention accordingly provides, in a first aspect, a method for use in a computerized storage system, comprising one or more replaceable units, for managing testing of one or more replacement units, said method comprising the steps of: a) identifying a replaceable unit in a storage system for replacement and switching said storage system into a service mode; b) switching said storage system into a test mode automatically in response to the replacement of said replaceable unit with a replacement unit; c) testing said replacement unit in accordance with a predetermined testing policy so as to verify the functioning of said replacement unit; and d) returning said storage system to an online mode if said testing of said selected elements is successful.

#Embodiments of the invention will now be described, by way of example only, with reference to the accompanying drawings in which:

#FIG. 1 is a schematic illustration of a computer network comprising a computerized storage system;

#FIG. 2 is a schematic illustration showing the computerized storage system of FIG. 1 in further detail;

#With reference to FIG. 1, a computer system 101 comprises a first computer 102 loaded with a storage server application program 103 and connected to a storage system 104. The first computer 102 is connected to two second computers 105 via a network in the form of a local area network (LAN) 106. The second computers 105 are each loaded with a data processing application program 107 arranged to store and access data on the storage system 104 via the storage server 103. In other words, the data processing application programs 107 provide user input/output (I/O) to the storage server 103 and storage system 104.

#With reference to FIG. 2, the storage server application 103 comprises a storage controller 201, arranged to provide a user interface and to handle user I/O, and duplicated storage adapters 202, 203 arranged to cooperatively manage the provision, service and testing of the storage system 104 as described in further detail below. The storage system 104 comprises three disk arrays 204, 205, 206 each comprising four disks 207 and two interface cards 208, 209. Each interface card 208, 209 provides a connection port for connecting to the adapters 202, 203. A set of links 210, 211 are provided for connecting each of the adaptors 202, 203 to both interface cards 208, 209 of each disk array 204, 205, 206 in series. In FIG. 2, labels for each occurrence of some duplicated items, such as the links 210, 211, are omitted for clarity. The provision of multiple duplicate adaptors 202, 203, links 210, 211, disks 207 and interface cards 208, 209 provides the storage system 104 with multiple redundancy. In the present embodiment, each unit 207, 208, 209, 210, 211 is duplicated at least once so as to provide the redundancy by, in effect, providing two duplicate storage systems. The storage controller 201 is arranged to use the duplicate storage systems to provide high availability since, if one duplicate storage system is offline, its duplicate storage system is arranged to provide the required service.

#In the present embodiment, each of the multiple links 210, 211, disks 207 and interface cards 208, 209 are provided as distinct units that may be replaced individually, commonly referred to as Field Replaceable Units (FRUs). The storage controller 201 comprises functionality to detect a range of faults and errors in the links 210, 211, disks 207 and interface cards 208, 209 and once detected an engineer is alerted. In order to replace a faulty FRU the relevant duplicate storage system is put from an online mode into a service mode with availability being maintained by the redundant duplicate storage system. In the present embodiment, when a given duplicate storage system is in service mode, all user I/O is diverted to the redundant duplicate storage system. The online or service mode switching is performed cooperatively between the respective storage adaptors 202, 203 so as to ensure that both are not simultaneously in service mode and thus availability is maintained.