Placeholder for AMDU binaries and using with ASM 10g (Doc ID 553639.1)

In this Document

Purpose

Requirements

Configuring

KNOWN ISSUES

Instructions

Script

Sample Output

References

---

APPLIES TO:

Oracle Database - Enterprise Edition - Version 10.1.0.3 to 10.2.0.4 [Release 10.1 to 10.2]
Information in this document applies to any platform.
Oracle Server Enterprise Edition - Version: 10.1.0.3 to 10.2.0.4

PURPOSE

On Oracle 10g the content of a diskgroup can be reviewed if the diskgroup is mounted and there are internal views that display the specific allocation of the files (ASM files and Database files) inside of the disks.

When the diskgroup is not mounted, this information is not available, which makes difficult to diagnose the errors avoiding the diskgroup to be mounted.   This problem has been resolved with AMDU.

AMDU is a tool introduced in 11g where it is posible to extract all the available metadata from one or more ASM disks, generate formatted block printouts from the dump output, extract one or more files from a diskgroup (mounted/unmounted) and write them to the OS file system.

This tool is very important when dealing with internal errors related to the ASM metadata. 

Although this tool was released with 11g, it can be used with ASM 10g.

This document provides the links to download the files required to run AMDU with an ASM 10g environment.

Platform
Linux X86 (Platform 46)	amdu_lnx_32.zip
Linux X86-64 AMD64/EM64T (Platform 226)	amdu_lnx_X86-64.zip
Solaris 10 Sparc 64bit (Platform 23)	amdu_solaris10_64.zip
Solaris 9 Sparc 64bit (Platform 23)	amdu_solaris9_64.zip
HP-UX PA-RISC 64bits (Platform 59)	amdu_HP-PARISC.zip
HP-UX Itanium (Platform 197)	amdu_hp_itanium.zip
AIX	amdu_aix.zip

* Additional platforms will be added

Table 1.  List of platforms

The scope of this document is not to explain AMDU usability and characteristics.  The main goal is to provide a placeholder where AMDU can be downloaded and use on Oracle 10g environments.  For Oracle 11g, this tool is part of the distribution and will be present on the $ORACLE_HOME/bin directory.

REQUIREMENTS

For specific details, please check the step-by-step section.

NOTE:  If a core dump or any error is reported during the execution, please submit a remark, so the binary can be replaced. 

CONFIGURING

1. Download the zip file referenced in Table 1 and transfer to the server.
2. Unzip the content

There are 4 files included in the zip file:

1. amdu -- amdu binary
2. libclntsh.so.11.1
3. libnnz11.so
4. libskgxp11.so

Note: The extension of the lib* files could be different for some platforms.

3. If using different ORACLE_HOMEs for ASM and DB, use the ASM location. Modify variable LD_LIBRARY_PATH or LIBPATH (on AIX) or equivalent, to include the directory where amdu was downloaded.  Also modify PATH variable:

$cd <your directory>
$export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:`pwd`   
or $setenv LD_LIBRARY_PATH ${LD_LIBRARY_PATH}:`pwd`
$export PATH=$PATH:`pwd`  or setenv PATH ${PATH}:`pwd`

or 

$cd <your directory>
$export LIBPATH=$LIBPATH:`pwd` 
or $setenv LIBPATH ${LIBPATH}:`pwd`
$export PATH=$PATH:`pwd` or setenv PATH ${PATH}:`pwd`

KNOWN ISSUES

* Execution of amdu could failed with a core dump. 

Reviewing the callstack in the coredump:

#0 0x0806235d in slfipn ()
#1 0x0805e88c in lfimknam ()
#2 0x081123bd in kfmuCreateDirectory ()
#3 0x0810a753 in kfmu_main ()
#4 0x08091a13 in lpmcall ()
#5 0x08065678 in lpmpmai ()
#6 0x08057fb9 in main ()

This is due to bug 5997071. The workaround is to unset any NLS* variable, starting with NLS_LANG.

INSTRUCTIONS

 

$  amdu -diskstring '/dev/asm1_disk*' -dump 'DATA1'

There are two main parameters:

-diskstring    Identify the location where the devices used by diskgroup are located.
-dump          Define the diskgroup whose content will be dumped by AMDU

Below is an example of how to use amdu, particularly to find the location of the DISK DIRECTORY.

This example scans disks under '/dev/asmdisk* and will dump details for diskgroup DATA.

Executing amdu:

$ ./amdu -diskstring '/dev/asmdisk*' -dump 'DATA' 

Change to the directory created which the name will be printed in the screen 
during execution of amdu.

Two files will be created:

* report.txt which provides details of disks identified during 
  the amdu execution
* <diskgroup>.map, Map files are ASCII files that describe the data in the 
  image files for a particular disk group

In the map file (*.map) we look for F00000002: (Disk Directory is always asm file# 2)

N0014 D0002 R00  A00000002 F00000002 I0 E00000000 U00 C00256 S0001 B0035663872

Now, read report.txt to identify the disk using N0014 and D0002:

N0014 -- DISK REPORT N0014 
D0002 -- Disk # 2 
A00000002 -- Allocation Unit 2
F00000002 -- File#2

The  file report.txt will have the path for this disk, using the values 
referenced before:

----------------------------- DISK REPORT  N0014 ----------------------
Disk Path: /dev/asmdisk17
Unique Disk ID:
Disk Label:
Physical Sector Size: 512 bytes
Disk Size: 954 megabytes
Group Name: DATA
Disk Name: DATA_0002
Failure Group Name: DATA_0002 Disk Number: 2 Header Status: 3

The AMDU information indicates that Disk Directory is located at AU 2 from Disk# 2 identified with path /dev/asmdisk17. 

Use kfed to get that specific AU:

$ kfed read /dev/asmdisk17 aunum=2 blknum=0 text=disk17_dd.txt

Note: Start reviewing from block 0. The entry of the added disk may be in a different block.

CAUTION

This sample code is provided for educational purposes only, and is not supported by Oracle Support. It has been tested internally, however, we do not guarantee that it will work for you. Ensure that you run it in your test environment before using.

SCRIPT

None

SAMPLE OUTPUT

Every time AMDU is executed a directory is created on the current location, unless it is override by parameter -directory.  The directory will have the format amdu_YYYY_MM_DD_HH24_MM_SS

$amdu -diskstring '/dev/asm1*' -dump 'DATA1'
amdu_2008_02_14_15_04_11/

The default command referenced in the step-by-step section, will generate following files:

DATA1_0001.img   -- could be more than one. size is limited to 2gb.
DATA1.map
report.txt

Here are some details about the files:

1. report.txt

Like the name, it include details about the disks scanned (name, diskgroup name, failgroup name, size, creation time, etc).  

2. *.map

For every diskgroup referenced on command -dump, will be a *.map file, which can be used to find the exact location of the ASM metadata on the disks. This file is a key component when reviewing specific areas of the metadata.

3. *.img

For each diskgroup, could be many image files.  The size is limited to 2gb and will be a exact dump of the content of the diskgroup.


Compress all the files under the directory and ask the customer to upload them to the Service Request.