High Availbility

OS & Virtualization

Monday, February 17, 2014

Diagnose Oracle RAC



Goal

To document the logs that should be uploaded for diagnosing Oracle Clusterware issue.
For more information about diagcollection, check out "diagcollection.sh -help"

This note will be obsolete in future, it's strongly recommended to use TFA to prune and collect files from all nodes:
note 1513912.1 - TFA Collector - Tool for Enhanced Diagnostic Gathering

Solution


Linux/UNIX 11gR2/12cR1

1. Execute the following as root user:
# script /tmp/diag.log
# id
# env
# cd
# $GRID_HOME/bin/diagcollection.sh
# exit
The following .gz files will be generated in the current directory and need to be uploaded along with /tmp/diag.log:

crsData_.tar.gz,
ocrData_.tar.gz,
oraData_.tar.gz,
coreData_.tar.gz (only --core option specified)
os_.tar.gz
Please ensure all above information are provided from all the nodes.

Linux/UNIX 10gR2/11gR1

1. Execute the following as root user:
# script /tmp/diag.log
# id
# env
# cd
# export OCH=
# export ORACLE_HOME=
# export HOSTNAME=
# $OCH/bin/diagcollection.pl -crshome=$OCH --collect

# exit


The following .gz files will be generated in the current directory and need to be uploaded along with /tmp/diag.log:
crsData_.tar.gz,
ocrData_.tar.gz,
oraData_.tar.gz,
coreData_.tar.gz (only --core option specified)

2. For 10gR2 and 11gR1, if getting an error while running root.sh, please collect /tmp/crsctl.*
Please ensure all above information are provided from all the nodes.

Windows 11gR2/12cR1:

set GRID_HOME=
%GRID_HOME%\perl\bin\perl %GRID_HOME%\bin\diagcollection.pl --collect

The following .zip files will be generated in the current directory and need to be uploaded:
crsData_.zip,
ocrData_.zip,
oraData_.zip,
coreData_.zip (only --core option specified)


Windows 10gR2/11gR1

set ORACLE_HOME=
set OCH=
set ORACLE_BASE=
$OCH%\perl\bin\perl %OCH%\bin\diagcollection.pl --collect

No comments: