Enterprise Bridge is designed for migration of very large data volumes. In case of large projects, migration process may run continuously for days, weeks and even months. Fault tolerance and automatic recovery are essential functions for successful completion of such challenging tasks.
Enterprise Bridge implements elaborate techniques to recover from migration errors. It has special parameter “Number of times to retry on failure”, which defines the number of attempts, which migration process will take before reporting conversion failure. Since network connectivity and responsiveness of source and target servers are always a factor of risk, reservations are made to retry any read and write operation certain number of times to avoid migration failure on isolated occasional errors.
Number can be any integer greater than zero. Repeated attempts are continued on progressively increasing time intervals until upper limit of attempts is reached. Intervals between attempts increase on quadratic law. For instance, limit of 5 attempts will yield a sequence of delays in 1, 4, 9, 16 and 25 seconds. Default number of attempts is 3. On very big conversions it is advised to set higher values, which might allow recovering network connectivity and even restarting servers without interruptions in conversion process.
