System and method for takeover of partner resources related to core dump

摘要

(For example, the transfer of the work memory of filer failed) system and method capable of more rapid takeover of the filer failed by partners clustered filer in a situation where there is a core dump procedure. In order to save time, at the same time as the takeover of the active file service disk of normal filer failed by partners, it is not necessary for it to run a core dump is allowed to start the takeover, wait for the completion of the core dump. In a nutshell, this is accomplished in the following manner. To be able without writing coredump on a single disk that is not involved in regular file service, and receive interference from coredump to advance the takeover of regular file services. For both filer two to form a cluster, is provided a reliable means to identify the coredump disk, thereby removing the dependency of the take-over to a communication means unreliable.