Last updated on MAY 02, 2017
Applies to:
Oracle Database - Enterprise Edition - Version 10.2.0.3 to 10.2.0.5.0 [Release 10.2]Linux x86-64
Linux x86
Symptoms
You detect poor performance after rolling out a new OS and RDBMS
- The OS is LINUX 64 or 32-bit
- The problem appears on 10.2.0.x : the problem may not be version specific
You discover CPU Utiltization is saturated which is resulting in poor performance
The problem can be serious enough for users to be bumped off the database
Post investigation: You find CPU is for gettimeofday calls is taking most of the time
/
.
root> strace -c -p 16180
Process 16180 attached - interrupt to quit
Process 16180 detached
%time seconds usecs/call calls errors syscall
------ -------------------- ----------- ----------
99.93 75.396823 12 6431816 gettimeofday << 99+% for gettimeofday
0.03 0.023839 8 3103 poll
0.01 0.010994 10994 1 shmdt
0.01 0.005379 22 243 mmap
0.01 0.004376 337 13 munmap
...
Process 16180 attached - interrupt to quit
Process 16180 detached
%time seconds usecs/call calls errors syscall
------ -------------------- ----------- ----------
99.93 75.396823 12 6431816 gettimeofday << 99+% for gettimeofday
0.03 0.023839 8 3103 poll
0.01 0.010994 10994 1 shmdt
0.01 0.005379 22 243 mmap
0.01 0.004376 337 13 munmap
...
Changes
The new OS is LINUX 64-bit, however, this can occur in 32bit as well.
The new Oracle version is 10.2.0.x
STATISTICS_LEVEL=ALL ( default=TYPICAL)
Cause
Sign In with your My Oracle Support account |
|
Don't have a My Oracle Support account? Click to get started |
My Oracle Support provides customers with access to over a
Million Knowledge Articles and hundreds of Community platforms