installed 11g, one of the processes eats up 100% CPU, and it doesn't stop. it started after i created database with DBCA. Recreated the database for a few times- always the same.
EM shows SQL that eats up CPU:

SELECT EXECUTION_ID, STATUS, STATUS_DETAIL
FROM MGMT_JOB_EXEC_SUMMARY
WHERE JOB_ID = :B3 AND TARGET_LIST_INDEX = :B2 AND EXPECTED_START_TIME = :B1

Any ideas what could be locked or stuck?

SQL> select count(*) from mgmt_job_exec_summary;

COUNT(*)

53398

- One session is consuming a lot of CPU (about 70% of 1 CPU and 55% of 2 CPU) in a sustained way. It also holds several locks on SYSMAN objetcs. That's session MODULE is OEM.CacheModeWaitPool. It seems to be looping with the following queries:

SELECT execution_id, status, status_detail
FROM mgmt_job_exec_summary
WHERE job_id = :b3 AND target_list_index = :b2 AND expected_start_time = :b1;

SELECT execution_id, status, status_detail
FROM mgmt_job_exec_summary
WHERE job_id = :b3 AND target_list_index = :b2 AND expected_start_time = :b1;

INSERT INTO mgmt_job_exec_summary
(job_id, job_type_id, execution_id, expected_start_time,
start_time, end_time, source_execution_id, target_list_index,
status, status_detail, triggering_severity
)
VALUES (:b11, :b10, :b6, :b9,
:b8, :b7, DECODE (:b5, NULL, :b6, :b5), :b4,
:b3, :b2, :b1
);

Essentially for the linux platform:
./emctl stop dbconsole
./sqlplus SYSMAN
SQL> execute MGMT_PAF_UTL.STOP_DAEMON
SQL> execute MGMT_PAF_UTL.START_DAEMON
./emctl start dbconsole

FROM:

http://forums.oracle.com/forums/thread.jspa?threadID=588715&tstart=0&messageID=2704390#2704390

[@more@]

ITPub博客

Oracle 11gR1 100% CPU Usage Overhead running EM DBConsole

最新文章