Error log consolidation

1. A server computer, comprising:

a plurality of system processors configurable to form isolated execution partitions;

error log consolidation logic configured to, based on detection of a fault in the server, retrieve error logs from the system processors, and to consolidate the retrieved logs with server computer information not available to the system processors to generate a consolidated error log;

wherein the consolidated error log includes a comprehensive set of server information relevant to identifying a cause of the detected fault.

2. The server computer of claim 1 , further including a fault analysis engine;

wherein the fault analysis engine is configured to determine a root cause of the detected fault based on the consolidated error log;

wherein the error log consolidation logic is configured to delay provision of the consolidated error log to the analysis engine until a recovery operation responsive to the detected fault is complete.

3. The server computer of claim 2 , wherein the delay comprises a selectable time interval determined based on the detected fault.

4. The server computer of claim 2 , wherein the detected fault indicates a correctable error, and failure of the recovery operation indicates an uncorrectable error, and the consolidated error log includes error logs generated for both the correctable and uncorrectable errors.

5. The server computer of claim 2 , wherein the detected fault indicates a correctable error, and the consolidated error log includes a record of the results of, and the steps performed as part of, the recovery operation.

6. The server computer of claim 1 , wherein the error log consolidation logic is configured to include server environmental information in the consolidated error log.

7. The server computer of claim 1 , further comprising a plurality of distributed management processors and queues configured to store error logs written by the system processors for retrieval by the distributed management processors while the system processors execute a recovery operation based on the detected error.

8. The server computer of claim 1 , wherein the error log consolidation logic retrieves a given error log generated by a given system processor via a management processor located in a partition with the given system processor.

9. The server computer of claim 1 , wherein error information generated by a server computer hardware resource shared by multiple partitions is retrieved directly from the hardware resources by a management processor coupled to the hardware resource and to the error log consolidation logic.

10. A method, comprising:

detecting, by a server computer, a fault in the computer;

retrieving, by an administration processor of the computer, error logs generated by a plurality of system processors of the computer; and

combining the retrieved error logs with computer information not available to the system processors to construct a consolidated error log comprising a comprehensive set of server information relevant to identifying a cause of the detected fault.

11. The method of claim 10 , wherein the combining comprises eliminating redundant error information from the consolidated error log.

12. The method of claim 10 , further comprising:

providing the consolidated error log to a fault analysis engine executed by the administration processor; and

determining a root cause of the detected fault based on the consolidated error log.

13. The method of claim 12 , further comprising: delaying the providing, by the administration processor, until a recovery operation responsive to the detected fault is complete, wherein a time of the delay comprises a time interval selected based on the detected fault.

14. The method of claim 10 , further comprising:

detecting, wherein the detected fault is a correctable error, failure of a recovery operation performed to correct the detected fault, wherein failure of the recovery operation indicates an uncorrectable error; and

including in the consolidated error report, error logs generated for both the correctable and uncorrectable error.

15. The method of claim 10 , wherein including error logs for the uncorrectable error comprises including in the consolidated error log, a record of the results and the steps performed as part of a recovery operation initiated to correct the detected fault.

16. A computer-readable medium encoded with a computer program comprising:

instructions that when executed cause a processor in a computer to:

retrieve error logs generated by a plurality of system processors of the computer responsive to a detected fault in the computer;

combine the retrieved error logs with computer information not available to the system processors to construct a consolidated error log comprising a comprehensive set of information relevant to identifying a cause of the detected fault; and

determine a root cause of the detected fault based on the consolidated error log.

17. The computer-readable medium of claim 16 , further comprising instructions that when executed cause the processor to eliminate redundant error information, provided in the retrieved error logs, as part of constructing the consolidated error log.

18. The computer-readable medium of claim 16 , further comprising instructions that when executed cause the processor to delay construction of the consolidated error log until a recovery operation responsive to the detected fault is complete, wherein a time of the delay comprises a time interval selected based on the detected fault.

19. The computer-readable medium of claim 16 , further comprising instructions that when executed cause the processor to:

determine whether the detected fault is a correctable error or an uncorrectable error;

determine, based on the fault being a correctable error, whether a recovery operation performed to correct the fault has failed, wherein failure of the recovery operation indicates an uncorrectable error;

include in the consolidated error report, error logs generated for both the correctable and uncorrectable error; and

include in the consolidated error log, a record of the results and the steps performed as part of a recovery operation initiated to correct the detected fault.

20. The computer-readable medium of claim 16 , further comprising instructions that when executed cause the processor to include computer environmental information in the consolidated error log.