Crash management of host computing systems in a cluster
AssigneeVMware, Inc.
InventorJinto Antony
Life of this patent
Estimating the term…
Abstract
Techniques for managing crash in a host computing system in a cluster are disclosed. In one embodiment, a core dump partition may be configured in memory associated with a remote access controller (RAC) within the host computing system. When the host computing system is crashed, a screen shot associated with the host computing system is obtained through the RAC. Further, a first occurrence of crash in the host computing system is determined based on the obtained screen shot. Furthermore, the core dump file associated with the first occurrence of crash is copied from the core dump partition to a shared datastore. Also, the host computing system is reset through the RAC upon copying the core dump file to the shared datastore.
Claims20
A method for managing crash in a host computing system in a cluster, comprising:
- configuring, by a management server, a core dump partition in memory associated with a remote access controller (RAC) within the host computing system, wherein the RAC enables remote management of the host computing system to perform configuration or maintenance of the host computing system;
- obtaining, by the management server, a screen shot associated with the host computing system through the RAC when the host computing system is crashed;
- determining, by the management server, a first occurrence of crash in the host computing system based on the obtained screen shot;
- copying, by the management server, core dump file associated with the first occurrence of crash from the core dump partition to a shared datastore;
- and
- resetting the host computing system, by the management server, through the RAC upon copying the core dump file.
A system for managing crash, the system comprising:
- a host computing system comprising
- a remote access controller (RAC);
- and
- memory associated with the RAC;
- a shared datastore;
- and
- a management server comprising a crash management module to:
- configure a core dump partition in the memory associated with the RAC within the host computing system, wherein the RAC enables remote management of the host computing system to perform configuration or maintenance of the host computing system;
- obtain a screen shot associated with the host computing system through the RAC when the host computing system is crashed;
- determine a first occurrence of crash in the host computing system based on the obtained screen shot;
- copy core dump file associated with the first occurrence of crash from the core dump partition to the shared datastore;
- and
- reset the host computing system through the RAC upon copying the core dump file.
A non-transitory computer-readable storage medium including instructions that are configured, when executed by a management server, to perform a method for managing crash in a host computing system in a cluster, the method comprising:
- configuring a core dump partition in memory associated with a remote access controller (RAC) within the host computing system, wherein the RAC enables remote management of the host computing system to perform configuration or maintenance of the host computing system;
- obtaining a screen shot associated with the host computing system through the RAC when the host computing system is crashed;
- determining a first occurrence of crash in the host computing system based on the obtained screen shot;
- copying core dump file associated with the first occurrence of crash from the core dump partition to a shared datastore;
- and
- resetting the host computing system through the RAC upon copying the core dump file.
Dependencies are inferred from the claim text.