International Business Machines Corporation

Globalfoundries Inc.

Ashwani Kumar

Steven Edward Klein

Matthew G Borlick

Matthew J Kalos

Roger Gregory Hathorn

Todd Charles Sorenson

Micah Robison

Charles Scott Cardinell

Beth Ann Peterson

Bret Wayne Holley

John Norbert McCauley

Mark Steven Bayus

Daniel W Sherman

James Lamar Hood

Tiep Quoc Pham

Takashi Suzuki

Various embodiments for troubleshooting a network device in a computing storage environment by a processor. A determination is made if a parity error in a specific port is either an instruction cache parity error or a recoverable data cache type parity error. If the parity error is determined not to be the instruction cache parity error, and determined not to be the recoverable data cache type parity error, a full recovery operation is initiated.

Alternative port error recovery with limited system impact

Various embodiments for troubleshooting a network device in a computing storage environment by a processor. In response to an error in a specific port, an alternative error recovery operation is initiated on the port by performing at least one of initiating a silent recovery operation by reloading a failed instruction, taking the port offline, cleaning up any active transactions associated with the port, performing a hardware reset operation port, and bringing the port online.

Provided are a computer program product, system, and method for injecting congestion in a link between adaptors in a network. A congestion request is sent to a selected adaptor in a containing network component comprising one of a plurality of network components. The selected adaptor is in communication with a linked adaptor in a linked network component comprising one of the network components. The congestion request causes a delay in servicing the selected adaptor to introduce congestion on a link between the selected adaptor and the linked adaptor.

Injecting congestion in a link between adaptors in a network

Provided are a computer program product, system, and method for injecting congestion in a link between adaptors in a network. A congestion request is received for the selected adaptor at a containing network component comprising one of the network components. In response to the received congestion request, servicing the selected adaptor is delayed to introduce congestion on a link between the selected adaptor and the linked adaptor.

A method for recovering from an error in a multi-core storage-system component is disclosed. In one embodiment, such a method includes detecting an error in a first core of a multi-core component. The method determines whether the error was one of (1) detected by the first core; and (2) detected by a core other than the first core. In the event the error was detected by the first core and the error is recoverable, the first core recovers from the error without substantially impacting operation of other cores in the multi-core component. In the event the error was detected by a core other than the first core and the error is recoverable, a core other than the first core recovers from the error without substantially impacting operation of other cores in the multi-core component. A corresponding apparatus and computer program product are also disclosed.

Reduced-impact error recovery in multi-core storage-system components

Systems and methods for detecting supported small form-factor pluggable (SFP) devices in an adapter are provided. One system includes multiple ports, each port configured to be coupled to a SFP device, a tracking device configured to store data representing a list of supported SFP devices for the adapter, and a processor coupled to each of the plurality of ports and the tracking device. The processor is configured to scan each port at start-up, determine if any of the ports is coupled to a non-supported SFP device, and disable any ports that are coupled to a non-supported SFP device. One method includes scanning each port at start-up, determining if any of the ports is coupled to a non-supported SFP device, and disabling any ports that are coupled to a non-supported SFP device. Also provided are computer storage mediums including computer code for performing the above method.

Systems and methods for detecting supported small form-factor pluggable (SFP) devices

Provided are a method, system, and computer program product for managing monitored conditions in adaptors. An adaptor detects a condition comprising one of a plurality of monitored conditions and determines whether the detected condition is indicated on a recent condition list indicating conditions detected at the adaptors. The adaptor communicates the detected condition to cause the detected condition to be included in the recent condition list available at the adaptors in response to determining that the detected condition is not indicated in the recent condition list. Further, a system receives from one of the adaptors indication of a detected condition comprising one of a plurality of monitored conditions that is not indicated on a recent condition list. The received indicated condition is added to the recent condition list to produce an updated recent condition list. Transmission is caused of the updated recent condition list to the adaptors.

Managing monitored conditions in adaptors in a multi-adaptor system

Disclosed are a method, apparatus, and computer program product to reduce the detrimental effects of taking resources offline. A resource is taken offline in a storage network by quiescing activity to the resource providing notification that the resource will become unavailable, and taking the resource offline.

Taking a resource offline in a storage network

A method is provided to protect against ill-behaved microcode by balancing between an actual occurrence of a hardware problem and a microcode bug setting a flag appearing as a hardware problem. In this method, the error recovery is performed only on a single piece of hardware and no further error recovery action is taken on other pieces of similar hardware. The approach addresses the problem by treating a hit on one card as a hardware problem, but as a bug on subsequent cards. The invention keeps track of whether or not the same event has occurred on the same type of hardware so not to take action on more than one instance of the hardware. Subsequent hits on another instance of the hardware will not trigger the hardware related recovery on additional hardware.

Growing community of inventors

Tucson, AZ, United States of America

Ashwani Kumar