Strange Elapsed In Raw Trace

Listers,

9.2.0.5 EE on Windows 2000 Server. Checking out a process (doing a recursive insert in a loop building on prior results, a graph explosion of sort) and tracing it to get a feel for what's going on. Note on the 3rd EXEC below the much larger value for E. But you can see from the timestamps from the trace that the value for E is really out of whack. And the differences between tim between two operations should be the time between operations, so when you subtract tim between exec 2 and 3, it is roughly the same as the value for e. So my initial thinking is that this is probably related to tim and how it is derived/gotten. Anyone seen this before?

PARSE #5:c=0,e=3328,p=0,cr=0,cu=0,mis=1,r=0,dep=1,og=0,tim=3401329811 *** 2004-10-13 16:06:45.632
EXEC
#5:c=4562500,e=207030037,p=11799,cr=36578,cu=6669,mis=0,r=15004,dep=1,og=4,t
im=3608386193
*** 2004-10-13 16:12:35.476
EXEC
#5:c=6484375,e=349792572,p=22658,cr=90642,cu=3842,mis=0,r=15801,dep=1,og=4,t
im=3958240955
*** 2004-10-13 16:21:02.632
EXEC
#5:c=9546875,e=18446744065825247453,p=32226,cr=92572,cu=15058,mis=0,r=14816,
dep=1,og=4,tim=18446744069784377492
*** 2004-10-13 16:26:53.070
EXEC
#5:c=10937500,e=349867455,p=29844,cr=94029,cu=14211,mis=0,r=19047,dep=1,og=4
,tim=18446744070134816920
*** 2004-10-13 16:35:24.898
EXEC
#5:c=15000000,e=511765036,p=42621,cr=124698,cu=25651,mis=0,r=19900,dep=1,og=
4,tim=18446744070646657016

Regards,

Larry G. Elkins
elkinsl_at_flash.net
214.954.1781

--
http://www.freelists.org/webpage/oracle-l