| 摘要 |
The present invention relates to a method for recovering a communications failure of a parallel connected server system and, more particularly, to a method which recovers the communications failure of the sever system having a master server equipped as an active server and a plurality of slave servers connected in parallel. The method comprises: a master server failure determination step of determining whether a communications failure occurs or not in a master server; a temporary master server designating step of designating any one among the slave servers as a temporary master server when it is determined by the failure determination step that the failure occurs in the master server; a data traffic allocation step of allocating data traffic of the master server to the temporary master server; a master server recovering step of recovering the failure of the master server; a master server recovering determination step of determining, by the master server recovering step, whether the failure of the master server is recovered or not; a data traffic reallocating step of reallocating the data traffic of the temporary master server to the master server when it is determined by the master server recovering determination step that the master server is recovered; and a slave server designating step of designating the temporary master sever as the slave server after the data traffic is reallocated. Accordingly, the method of the present invention can effectively and easily recover the communications failure of the master server without a separate physical switching process during a server failure. |
