Detecting and handling errors in a bus structure

A system for detecting and handling errors in a bus structure, the system configure to perform operations comprising:

• determining a location of an error in the bus structure by obtaining error information from a plurality of hardware registers associated with a bus;
• determining whether an amount of errors in one or more hardware registers of the plurality of hardware registers exceeds a predetermined threshold;
• monitoring an interrupt message reporting the error information from the one or more hardware registers to obtain the error information;
• disabling the interrupt message when the amount of errors in the one or more hardware registers exceeds the predetermined threshold;
• detecting a performance deterioration of a hardware device among a plurality of hardware devices corresponding to the one or more hardware registers;
• in response to determining performance deterioration of the hardware device, inferring that an error occurs in the hardware device;
• determining whether a backup device exists for the hardware device;
• in response to determining existence of the backup device, performing a failover on the hardware device using the backup device;
• resetting the hardware device;
• and
• after resetting the hardware device, deciding whether to switch back to the hardware device from the backup device at least partially based on an error recovery history of the hardware device, wherein the error recovery history of the hardware device is indicative of the error in the bus structure being temporary if the hardware device is reset for a first time.

A computer program product comprising:

• a non-transitory computer readable medium encoded with computer executable program code, wherein the code enables execution across one or more processors for detecting and handling errors in a bus structure, comprising:
• determining a location of an error in the bus structure by obtaining error information from a plurality of hardware registers associated with a bus by performing at least one of:
• periodically polling the plurality of hardware registers to obtain the error information by dynamically adjusting a periodicity for periodically polling the plurality of hardware registers based on the amount of errors in one or more hardware registers of the plurality of hardware registers;
• and
• monitoring an interrupt message reporting the error information from the one or more hardware registers to obtain the error information, wherein when the amount of errors is greater than a predetermined threshold, disabling the interrupt message;
• and
• wherein the error information is obtained by periodically polling the plurality of hardware registers when the amount of errors is greater than a predetermined threshold;
• determining whether the amount of errors in the one or more hardware registers exceeds the predetermined threshold;
• detecting a performance deterioration of a hardware device among a plurality of hardware devices corresponding to the one or more hardware registers;
• in response to determining the performance deterioration of the hardware device, inferring that an error occurs in the hardware device;
• determining whether a backup device exists for the hardware device;
• in response to determining existence of the backup device, performing a failover on the hardware device using the backup device;
• resetting the hardware device;
• and
• after resetting the hardware device, deciding whether to switch back to the hardware device from the backup device at least partially based on an error recovery history of the hardware device, wherein the error recovery history of the hardware device is indicative of the error in the bus structure being temporary if the hardware device is reset for a first time.