Search icon CANCEL
Subscription
0
Cart icon
Your Cart (0 item)
Close icon
You have no products in your basket yet
Arrow left icon
Explore Products
Best Sellers
New Releases
Books
Videos
Audiobooks
Learning Hub
Conferences
Free Learning
Arrow right icon
Arrow up icon
GO TO TOP
Oracle 11g R1/R2 Real Application Clusters Essentials

You're reading from   Oracle 11g R1/R2 Real Application Clusters Essentials Design, implement, and support complex Oracle 11g RAC environments for real world deployments

Arrow left icon
Product type Paperback
Published in May 2011
Publisher Packt
ISBN-13 9781849682664
Length 552 pages
Edition 1st Edition
Languages
Arrow right icon
Toc

Table of Contents (20) Chapters Close

Oracle 11g R1/R2 Real Application Clusters Essentials
Credits
About the Authors
About the Reviewers
www.PacktPub.com
Preface
1. High Availability FREE CHAPTER 2. Oracle 11g RAC Architecture 3. Clusterware Installation 4. Automatic Storage Management 5. Managing and Troubleshooting Oracle 11g Clusterware 6. RAC Database Administration and Workload Management 7. Backup and Recovery 8. Performance Tuning 9. Oracle 11g Clusterware Upgrade 10. Real-world Scenarios 11. Enabling RAC for EBS 12. Maximum Availability Additional Resources and Tools for the Oracle RAC Professional Index

Operating system-level commands for tuning and diagnosis


While the previous Oracle database commands prove useful for managing and troubleshooting Oracle RAC and Oracle Clusterware issues, there is a critical need to use operating system commands with the Linux and Unix environment to diagnose and tune difficult issues that may arise in the Oracle RAC ecosystem. We will now provide you with some useful details on tools that can benefit you in these situations.

Strace

Linux provides a useful tracing tool for system calls that can be used for diagnosis and debugging issues with Oracle 11g RAC environments. As DBAs, we often have problems with our database environments when a runaway Oracle database process causes a performance problem in consuming excessive memory from the operating system. By understanding how to use a trace tool, such as strace for your Linux environment, you can pinpoint the root cause of such a problem and find a solution to the failure more effectively.

To display the options for strace, enter the strace at a Linux or Unix shell prompt session as follows:

[oracle@raclinux1 ~]$ strace
usage: strace [-dffhiqrtttTvVxx] [-a column] [-e expr] ... [-o file]
              [-p pid] ... [-s strsize] [-u username] [-E var=val] ...
              [command [arg ...]]
   or: strace -c [-e expr] ... [-O overhead] [-S sortby] [-E var=val] ...
              [command [arg ...]]
-c -- count time, calls, and errors for each syscall and report summary
-f -- follow forks, -ff -- with output into separate files
-F -- attempt to follow vforks, -h -- print help message
-i -- print instruction pointer at time of syscall
-q -- suppress messages about attaching, detaching, etc.
-r -- print relative timestamp, -t -- absolute timestamp, -tt -- with usecs
-T -- print time spent in each syscall, -V -- print version
-v -- verbose mode: print unabbreviated argv, stat, termio[s], etc. args
-x -- print non-ascii strings in hex, -xx -- print all strings in hex
-a column -- alignment COLUMN for printing syscall results (default 40)
-e expr -- a qualifying expression: option=[!]all or option=[!]val1[,val2]...
   options: trace, abbrev, verbose, raw, signal, read, or write
-o file -- send trace output to FILE instead of stderr
-O overhead -- set overhead for tracing syscalls to OVERHEAD usecs
-p pid -- trace process with process id PID, may be repeated
-s strsize -- limit length of print strings to STRSIZE chars (default 32)
-S sortby -- sort syscall counts by: time, calls, name, nothing (default time)
-u username -- run command as username handling setuid and/or setgid
-E var=val -- put var=val in the environment for command
-E var -- remove var from the environment for command
[oracle@raclinux1 ~]$

Now, let's take a look at how to use STRACE to trace an Oracle 11g RAC process:

[oracle@raclinux1 ~]$ strace oracle
execve("/u01/app/oracle/product/11.1.0/11g/bin/oracle", ["oracle"], [/* 41 vars */]) = 0
uname({sys="Linux", node="raclinux1.us.oracle.com", ...}) = 0
brk(0)                                  = 0xf56b000
access("/etc/ld.so.preload", R_OK)      = -1 ENOENT (No such file or directory)
open("/u01/app/oracle/product/11.1.0/11g/lib/tls/i686/sse2/libskg

We can also use STRACE to examine system calls for an Oracle 11g database instance, in an Oracle 11g RAC environment, using the following example:

[oracle@raclinux1 bin]$ strace -c oracle
% time     seconds  usecs/call     calls    errors syscall
------ ----------- ----------- --------- --------- ----------------
  4.19    0.865232       41202        21           write
  4.02    0.831352      207838         4           lstat64
  4.02    0.831329      207832         4           _llseek
  4.02    0.831327      207832         4           times
  3.42    0.707903      707903         1           io_setup
  3.42    0.707867      707867         1           _sysctl
  3.42    0.707864      707864         1           socket
  3.42    0.707843      707843         1           bind
  3.42    0.707839      707839         1           set_thread_area
  3.42    0.707837      707837         1           futex
  3.42    0.707836      707836         1           dup
  3.42    0.707836      707836         1           set_tid_address
  3.42    0.707835      707835         1           gettid
  3.42    0.707833      707833         1           fcntl64
  3.42    0.707831      707831         1           getuid32
  3.42    0.707831      707831         1           geteuid32
  3.21    0.662962       82870         8           mprotect
  3.21    0.662671       82834         8           gettimeofday
  2.77    0.572299       26014        22           fstat64
  2.71    0.560835       12192        46           old_mmap
  2.66    0.549201      109840         5           shmdt
  2.61    0.539206      107841         5           shmat
  2.61    0.539194      107839         5         5 access
  2.61    0.539158      107832         5           rt_sigprocmask
  2.46    0.509131        2204       231           close
  2.39    0.494136       41178        12         9 shmget
  2.01    0.415677      207839         2           uname
  1.57    0.325381       20336        16        16 mkdir
  1.30    0.268287        5708        47        17 stat64
  1.27    0.263489         909       290        61 open
  1.19    0.246993       41166         6           getrlimit
  1.19    0.246990       41165         6           rt_sigaction
  0.87    0.179247        2938        61           read
  0.60    0.123500       41167         3           brk
  0.60    0.123500       41167         3           setrlimit
  0.60    0.123494       41165         3           lseek
  2.70   -1.1557161        2786       200           mmap2
  1.52   -1.1313491        2986       105           munmap
------ ----------- ----------- --------- --------- ----------------
100.00   20.669398                  1134       108 total

Truss

Truss is a useful utility available for the Sun Solaris Unix operating system that can be used for tracing Oracle 11g background processes.

Truss provides the ability to walk through trace, step by step, for Oracle 11g internal processes, by using the –p parameter to the truss command, followed by the Solaris process id (PID). You can also trace child processes for Oracle background processes by using the truss command with the –f parameter. The parameters available for the truss command are listed next:

truss [-fcaeil] [-[tvx] [!]syscall...] [-s [!]signal...] [-m [!]fault...] [-[rw] [!]fd...] [-o outfile] command | -p pid 

Some typical options for truss are as follows:

  • -o: output to file

  • -f: trace child processes

  • -c: count system calls

  • -p: trace calls based on Unix PID

Now, let's take a look at how to trace the Oracle 11g database instance process by using the truss utility with Solaris:

solaris06$ truss -cp 7612 
^C 
Syscall	seconds	calls	errors
read 	.900	55 
Write	.500	1
times	.100 	93
yield 	.500	426

GDB

GDB is a system debugger available for the Linux platform that can be used to discover problems with Oracle system processes. The man page available in Unix and Linux platforms for GDB will provide you with all of the command-line syntax options. To display the syntax for GDB, enter the man GDB command at a Unix or Linux shell prompt. Further details on GDB are also available online at http://linux.die.net/man/1/gdb.

GDB provides you with the ability to trace system memory usage for background processes, as well as the ability to dump memory values for these background processes. This is useful in troubleshooting difficult problems, such as when a memory leak occurs with an Oracle 11g database process. Before you can use the GDB utility to trace the Oracle 11g database instance process, you will first need to determine the specific Process ID (PID) in Linux by using the ps and grep Linux/Unix operating system commands. In the following example, we will use GDB to trace an open SQL*PLUS session within the Linux operating system.

After finding the process id (4107) for the SQL*PLUS session, we can begin the debug session with GDB:

[oracle@raclinux1 ~]$ gdb $ORACLE_HOME/bin/oracle 4107 start the new session for GDB 
GNU gdb Red Hat Linux (6.1post-1.20040607.62rh)
Copyright 2004 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB.  Type "show warranty" for details.
This GDB was configured as "i386-redhat-linux-gnu"...(no debugging symbols found)...Using host libthread_db library "/lib/tls/libthread_db.so.1".

Attaching to program:   /u01/app/oracle/product/11.1.0/11g/bin/oracle, process 4107   The GDB dump starts by attaching to the PID we gave it earlier
Reading symbols from /u01/app/oracle/product/11.1.0/11g/lib/libskgxp11.so...(no debugging symbols found)...done.
Loaded symbols for /u01/app/oracle/product/11.1.0/11g/lib/libskgxp11.so
Reading symbols from /lib/tls/librt.so.1...(no debugging symbols found)...done.
Loaded symbols for /lib/tls/librt.so.1  GDB loads the shared libraries for Oracle to perform the memory dump trace
Reading symbols from /u01/app/oracle/product/11.1.0/11g/lib/libnnz11.so...(no debugging symbols found)...done.
Loaded symbols for /u01/app/oracle/product/11.1.0/11g/lib/libnnz11.so
Reading symbols from /u01/app/oracle/product/11.1.0/11g/lib/libclsra11.so...done.
Loaded symbols for /u01/app/oracle/product/11.1.0/11g/lib/libclsra11.so
Reading symbols from /u01/app/oracle/product/11.1.0/11g/lib/libdbcfg11.so...done.
Loaded symbols for /u01/app/oracle/product/11.1.0/11g/lib/libdbcfg11.so
Reading symbols from /u01/app/oracle/product/11.1.0/11g/lib/libhasgen11.so...done.
Loaded symbols for /u01/app/oracle/product/11.1.0/11g/lib/libhasgen11.so
Reading symbols from /u01/app/oracle/product/11.1.0/11g/lib/libskgxn2.so...done.Loaded symbols for /u01/app/oracle/product/11.1.0/11g/lib/libskgxn2.so
Reading symbols from /u01/app/oracle/product/11.1.0/11g/lib/libocr11.so...done.
Loaded symbols for /u01/app/oracle/product/11.1.0/11g/lib/libocr11.so
Reading symbols from /u01/app/oracle/product/11.1.0/11g/lib/libocrb11.so...done.Loaded symbols for /u01/app/oracle/product/11.1.0/11g/lib/libocrb11.so
Reading symbols from /u01/app/oracle/product/11.1.0/11g/lib/libocrutl11.so...done.
Loaded symbols for /u01/app/oracle/product/11.1.0/11g/lib/libocrutl11.so
Reading symbols from /usr/lib/libaio.so.1...done.
Loaded symbols for /usr/lib/libaio.so.1
Reading symbols from /lib/libdl.so.2...done.
Loaded symbols for /lib/libdl.so.2
Reading symbols from /lib/tls/libm.so.6...done.
Loaded symbols for /lib/tls/libm.so.6
Reading symbols from /lib/tls/libpthread.so.0...done.

(gdb) stepi we use the STEPI command for GDB to walk the dump through the memory
[Switching to Thread -1208035648 (LWP 4107)]
0x0090a7a2 in _dl_sysinfo_int80 () from /lib/ld-linux.so.2
(gdb) call ksudss(10)   We issue the CALL command to move through the memory stack
$1 = 0

(gdb) detach   we use the detach command for GDB to end our memory tracing session
Detaching from program: /u01/app/oracle/product/11.1.0/11g/bin/oracle, process 4107

The next step is to attach the Oracle process id (PID) 4107 found earlier to the GDB debugger. Before we can make use of the GDB functions, we must first use the GDB attach command to link the debugger to the Oracle process that will be traced. This will enable us to drill down further into the tracing for Oracle 11g by using the attach command with GDB, as well as show the memory dump information:

(gdb) attach 4107  we start a new memory dump trace by using the attach command for GDB
Attaching to program: /u01/app/oracle/product/11.1.0/11g/bin/oracle, process 4107
[New Thread -1208035648 (LWP 4107)]
Symbols already loaded for /u01/app/oracle/product/11.1.0/11g/lib/libskgxp11.so
Symbols already loaded for /lib/tls/librt.so.1
Symbols already loaded for /u01/app/oracle/product/11.1.0/11g/lib/libnnz11.so
Symbols already loaded for /u01/app/oracle/product/11.1.0/11g/lib/libclsra11.so
Symbols already loaded for /u01/app/oracle/product/11.1.0/11g/lib/libdbcfg11.so
Symbols already loaded for /u01/app/oracle/product/11.1.0/11g/lib/libhasgen11.soSymbols already loaded for /u01/app/oracle/product/11.1.0/11g/lib/libskgxn2.so
Symbols already loaded for /u01/app/oracle/product/11.1.0/11g/lib/libocr11.so
Symbols already loaded for /u01/app/oracle/product/11.1.0/11g/lib/libocrb11.so
Symbols already loaded for /u01/app/oracle/product/11.1.0/11g/lib/libocrutl11.soSymbols already loaded for /usr/lib/libaio.so.1
Symbols already loaded for /lib/libdl.so.2
Symbols already loaded for /lib/tls/libm.so.6
Symbols already loaded for /lib/tls/libpthread.so.0
Symbols already loaded for /lib/libnsl.so.1
Symbols already loaded for /lib/tls/libc.so.6
Symbols already loaded for /lib/ld-linux.so.2
Symbols already loaded for /usr/lib/libnuma.so
Symbols already loaded for /lib/libnss_files.so.2
Symbols already loaded for /u01/app/oracle/product/11.1.0/11g/lib/libnque11.so
[Switching to Thread -1208035648 (LWP 3107)]
0x0090a7a2 in _dl_sysinfo_int80 () from /lib/ld-linux.so.2

Now, to get the stack dump, we will need to issue the bt command for our GDB session:

(gdb) bt We issue the BT command for GDB to obtain a listing of the memory stack dump 
#0  0x0090a7a2 in _dl_sysinfo_int80 () from /lib/ld-linux.so.2
#1  0x003d14c3 in __read_nocancel () from /lib/tls/libpthread.so.0
#2  0x0e5b652e in sntpread ()
#3  0x0e5b64cf in ntpfprd ()
#4  0x0e59b3c7 in nsbasic_brc ()
#5  0x0e59e20e in nsbrecv ()
#6  0x0e5a2d20 in nioqrc ()
#7  0x0e39da65 in __PGOSF20_opikndf2 ()
lock icon The rest of the chapter is locked
Register for a free Packt account to unlock a world of extra content!
A free Packt account unlocks extra newsletters, articles, discounted offers, and much more. Start advancing your knowledge today.
Unlock this book and the full library FREE for 7 days
Get unlimited access to 7000+ expert-authored eBooks and videos courses covering every tech area you can think of
Renews at $19.99/month. Cancel anytime
Banner background image