RE: 10046, Unaccounted for Time, Aix 5.3 -- How to Determine Process Pre-Emption
Date: Wed, 2 Dec 2009 15:35:26 -0600
Message-ID: <006e01ca7397$5e8179c0$1b846d40$_at_net>
Profiler (and other similar products) are already used. Mrnl is not, though it would most certainly be useful. Aggregated unaccounted for time is known.
What is being looked for by the "SA's" is something along the lines of strace, truss, ptrace, trace, whatever, and how those could be used by the SA's to see if pre-emption is occurring. They aren't sure of the best way/tool to use, and I don't have a recommendation for them, though plenty of examples, especially tied into Oracle, for flavors other than AiX.
Larry G. Elkins
Larry,
See pages 170-173 if you have access to a copy of Optimizing Oracle Performance (O'Reilly). Bottom line: unaccounted-for time in trace data on Oracle 9.2 or beyond is almost certainly time spent preempted.
Your analysis is sweet-spot for two of the software tools we make. Check out these if you're interested:
- Method R Profiler (http://method-r.com/software/profiler), which reports on properly aggregated unaccounted-for time.
- And mrnl (http://method-r.com/component/content/article/116) in the MR Tools suite, which allows you to drill in on exactly which specific lines of trace data are contributing the most unaccounted-for time. The final example on the mrnl web page I pointed to at the beginning of this paragraph is your situation on the nose.
Let me know if there's anything we can do for you...
-- http://www.freelists.org/webpage/oracle-lReceived on Wed Dec 02 2009 - 15:35:26 CST