Process OS id : 2265390 alive after kill

alert警告文件中报：
Tue Jul 21 02:50:30 2009
Immediate Kill Session#: 2186, Serial#: 17945
Immediate Kill Session: sess: 70000042a6faa70  OS pid: 2265390
Tue Jul 21 02:50:30 2009
Process OS id : 2265390 alive after kill
Errors in file
Tue Jul 21 02:50:49 2009
Immediate Kill Session#: 2187, Serial#: 48042
Immediate Kill Session: sess: 7000004296fecf0  OS pid: 2760774
Tue Jul 21 02:50:49 2009
Process OS id : 2760774 alive after kill
Errors in file /oracle/admin/nrms/udump/nrms_ora_2507006.trc
Tue Jul 21 03:11:48 2009
Immediate Kill Session#: 2145, Serial#: 57134
Immediate Kill Session: sess: 7000004296e2b68  OS pid: 2404696
Tue Jul 21 03:11:49 2009
Process OS id : 2404696 alive after kill
Errors in file
Tue Jul 21 03:14:32 2009
Immediate Kill Session#: 2649, Serial#: 34742
Immediate Kill Session: sess: 700000429833dc8  OS pid: 2338998
Tue Jul 21 03:14:32 2009
Process OS id : 2338998 alive after kill
Errors in file /oracle/admin/nrms/udump/nrms_ora_1749396.trc




Metalink中的说明：


Subject:     Getting Trace Files With "Short stack dump: unable to dump stack due to error 72".
      Doc ID:     790393.1     Type:     PROBLEM
      Modified Date :     16-MAR-2009     Status:     MODERATED

In this Document
  Symptoms
  Cause
  Solution
  References

This document is being delivered to you via Oracle Support's Rapid Visibility (RaV) process, and therefore has not been subject to an independent technical review.

Applies to:
Oracle Server Enterprise Edition - Version: 10.2.0.1
This problem can occur on any platform.
Symptoms

When doing a startup or shutdown, you might see the following in the alert log :

Wed Mar 11 05:18:14 2009
Process OS id : 974992 alive after kill
Errors in file /u01/app/oracle/admin/kroprd/udump/kroprd_ora_721366.trc
Wed Mar 11 05:18:15 2009
Process OS id : 1188106 alive after kill
Errors in file /u01/app/oracle/admin/kroprd/udump/kroprd_ora_721366.trc
Wed Mar 11 05:18:17 2009
ALTER DATABASE CLOSE NORMAL

And the trace files will contain the following message :

Process Group: DEFAULT, pseudo proc: 700000057b078f0
O/S info: user: oracle, term: UNKNOWN, ospid: 840160
OSD pid info: Unix process pid: 840160, image: oraclekroprd@hemlock
Short stack dump: unable to dump stack due to error 72
Cause
The "alive after kill" messages in the Job Queue slaves can sometimes be seen if a shutdown immediate is issued during a startup or at least before the Job Queue slaves have completed initializing. It is not a problem and turning off the Job Queue Processes will simply avoid getting these trace files but that would be a severe workaround just to avoid getting 2K trace files.

The Job Queue Process subsystem is separate from the rest of the DB startup/ shutdown process and it's common for the Job Queue Processes to be alive after kill or not to startup for a while after database startup. The "alive after kill" message was removed from the alert log in 11g because too many people wrongly thought this was a big problem.
Solution

1. Ignore the messages.

2. Upgrade to 11g where this message is removed from the alert log.
References
Bug 7233028 - * TRACE FILE BEING GENERATED WHILE SHUTDOWN DATABASE

记录一下