APPLIES TO:
Oracle Database - Enterprise Edition - Version 12.2.0.1 and later
Information in this document applies to any platform.
SYMPTOMS
- Alert log messages such as the following are seen:
Suspending MMON action ‘alert message cleanup‘ for 82800 seconds
Suspending MMON action ‘metrics monitoring‘ for 82800 seconds
Suspending MMON action ‘undo usage‘ for 82800 seconds
Suspending MMON slave action kewrmafsa_ for 82800 seconds
Suspending MMON slave action kewrmapsa_ for 82800 seconds
Suspending MMON slave action kttetslv_ for 82800 seconds
CAUSE
In certain circumstances, typically where failure to complete an activity of some sort has occurred and MMON has timed out (with something like ORA-12751), MMON actions can be temporarily suspended. This suspension can be triggered by a large number of factors and generates alert log messages depending on where the issue is detected. Typically the warnings indicate excessive resource usage of some nature. You may well see other messages such as ORA-12751 "cpu time or run time policy violation" errors. See the following article for assistance with these:
SOLUTION
While MMON activity is important in maintaining information in the database, if the activity is excessive such that it exceeds cpu or run time policies, then MMON activity can be temporarily suspended. This is done so that more important activity can continue. The "Suspending MMON ... for 82800 seconds" warning messages indicate that you should look at the overall performance of your system to check if there is anything that could be causing MMON activities to exceed their resource allocation such that they are suspended. If not, then investigate the specific area mentioned in the warning.
If the suspension of MMON starts to cause other problems, then contact support for further advice.
结果很显然：这个告警只有在导致其他问题时才去关注，其他情况都可以忽略。