| main - Diagnostic Assistant: Exadata Main Menu | |
The Oracle Diagnostic Assistant Exadata Main Menu enables you to collect
diagnostics based on a specific issue. Exadata system components have been
grouped by functional categories below and includes access to Exachk.
Select one of the following options:
(1) Database
(2) Exadata - Cell
(3) Exadata - Hardware
(4) Network
(5) Oracle Errors
(6) Processes
(7) RAC
----------Generic Diagnostic Tools ----------
(11) Exachk
(H)elp (E)xit
Showing Menu Help:
Oracle Diagnostic Assistant Exadata Main Menu Help
The Exadata Main Menu enables you to diagnose your Exadata environment.
Explanation of available options:
(1) Database
o Select this option if you suspect a database issue
(2) Exadata - Cell
o Select this option if you suspect a cell failure, performance
issue, or patching problem
(3) Exadata - Hardware
o Select this option if you suspect underperforming flash drives
or failed drives on Exadata Cell or Compute Node
(4) Network
o Select this option if you suspect issues with the listener, Network
cabling, IB Switch or any other Network related problem
(5) Oracle Errors
o Select this option to identify and package incidents
(6) Processes
o Currently the only option available in this section is to analyze
a process you suspect is spinning/hanging using procwatcher
o It is important to note that ProcWatcher is NOT delivered with
the Diagnostic Assistant
(7) RAC
o Select this option if you suspect issues with the RAC instance or
node evictions or you encounter clusterware startup issues
Generic Diagnostic Tools
This section gives direct access to diagnostic tools that are NOT
delivered with the Diagnsotic Assistant, but are commonly used to
troubleshoot issues related to the menu target.
(11) Exachk Collection
o Access to the Exachk diagnostic tool
|
|
| Menu Option: 1 - [+] Show Menu Details | |
| Menu Option: 2 - [+] Show Menu Details | |
| Menu Option: 3 - [+] Show Menu Details | |
| Menu Option: 4 - [+] Show Menu Details | |
| Menu Option: 5 - [+] Show Menu Details | |
| Menu Option: 6 - [+] Show Menu Details | |
| Menu Option: 7 - [+] Show Menu Details | |
| Menu Option: 11 - [+] Show Menu Details | |
| DaExaDB - Database Issues | |
Use this menu if you suspect you are having database problems.
Select one of the following options:
(1) Oracle Errors
(2) Database HANG
(3) Database / Instance Crash
(4) Lost Write
(5) ASM Issues and Errors
(M)ain (B)ack (H)elp (E)xit
Showing Menu Help:
Exadata: Database Issues Help
Explanation of available options:
Diagnosis is currently available for the following database issues:
(1) Oracle Errors
o Selecting this option will allow you to identify the following
Oracle Error incidents and package the incidents
o ORA-600 Internal Errors
o ORA-1578 ORACLE data block corrupted
o ORA-7445 exception encountered: core dump
o ORA-3113 end-of-file on communication channel
o ORA-4031 unable to allocate string bytes of shared memory
(2) Database HANG
o Selecting this option will execute the Exadata_DatabaseHang RDA
profile which includes the following modules:
o RACD (performs a database hang analysis)
o PERF (collects performance information)
o LOG (collects database trace and log information)
(3) Database / Instance Crash
o Selecting this option will execute the following RDA module:
o LOG (collects database trace and log information)
(4) Lost Write
o Select this option to collect the controlfile and redohdr dump data
(5) ASM Issues and Errors
o Selecting this option will execute the following RDA module:
o ASM (collects Automatic Storage Management Information)
|
|
| Menu Option: 1 - [+] Show Menu Details | |
| Menu Option: 2 - [+] Show Command Details | |
| Menu Option: 3 - [+] Show Menu Details | |
| Menu Option: 4 - [+] Show Command Details | |
| Menu Option: 5 - [+] Show Command Details | |
| Menu Option: M - [+] Show Menu Details | |
| Menu Option: B - [+] Show Menu Details | |
| DaExaDBOraErr - Database: Oracle Errors | |
For Oracle Errors, a 2 step process is needed to gather the incident
details. Option 1 will identify the incidents and option 2 will create
the incident package.
Select one of the following options:
(1) View incidents, limited by number
(2) Create incident package
(M)ain (B)ack (H)elp (E)xit
Showing Menu Help:
Database Issues: Oracle Errors Help
Select option 1 to find the incident number needed for option 2.
Explanation of available options:
(1) Show incident, limited by number
o You will be prompted for MODE
--> The 'MODE' of output includes BASIC, BRIEF or DETAIL where
each option provides a different level of detail
o you will be prompted for the number of incidents you wish to view
--> default is 50
(2) Create package
o incident id is required
o this option will create a logical incident package
|
|
| Menu Option: 1 - [+] Show Command Details | |
| Menu Option: 2 - [+] Show Command Details | |
| Menu Option: M - [+] Show Menu Details | |
| Menu Option: B - [+] Show Menu Details | |
| DaExaDBCrash - Database / Instance Crash | |
RDA will execute the following module=LOG to help diagnose your
database orinstance crash.
Select one of the following options:
(1) Run A New Collection
(2) View Existing Collection
(M)ain (B)ack (H)elp (E)xit
Showing Menu Help:
Database Issues: Database or Instance Crash Help
Explanation of available options:
(1) Run A New Collection
o RDA will execute the following module:
- LOG (collects database trace and log information)
(2) View Existing Collection
o Selecting this option will allow for the previous like
collection to be displayed
o If no previous collection exists, it will be executed and
results displayed and next time, this file will be available
for viewing
|
|
| Menu Option: 1 - [+] Show Command Details | |
| Menu Option: 2 - [+] Show Command Details | |
| Menu Option: M - [+] Show Menu Details | |
| Menu Option: B - [+] Show Menu Details | |
| DaExaCell - exadata - Cell Issues | |
Select one of the following options:
(1) Failed Drives on Exadata Cell or Compute Node
(2) Exadata Cell Failure >15 mins
(3) Cell Patching
(4) Sick or underperforming Cell / Performance Issue >15 mins
(5) Long Brownout Due to Cell Failure >15 mins
(6) ILOM issues
(M)ain (B)ack (H)elp (E)xit
Showing Menu Help:
Exadata - Cell Issues Help
Explanation of available options:
This option enables the following Exadata modules to be selected
for execution:
(1) Failed Drives on Exadata Cell or Compute Node
o The 'Exadata_FailedDrives' RDA profile will execute
(2) Exadata Cell Failure
o The 'Exadata_SickCell' RDA profile will execute
o This collection may take over 15 minutes to complete
(3) Cell Patching
o Collects important patching and validataion logs used for
identifying potential patching issues
(4) Sick or underperforming Cell / Performance Issue
o The 'Exadata_SickCell' RDA profile will execute
o You will be prompted to enter a specific cell name/IP or
hit enter to take the default of all cells
o This collection may take over 15 minutes to complete
(5) Long Brownout Due to Cell Failure
o The 'Exadata_Assessment' RDA profile will execute
o This collection may take over 15 minutes to complete
(6) ILOM issues
o Generates ILOM Snapshot
o Collects ILOM Data
|
|
| Menu Option: 1 - [+] Show Command Details | |
| Menu Option: 2 - [+] Show Command Details | |
| Menu Option: 3 - [+] Show Menu Details | |
| Menu Option: 4 - [+] Show Command Details | |
| Menu Option: 5 - [+] Show Command Details | |
| Menu Option: 6 - [+] Show Menu Details | |
| Menu Option: M - [+] Show Menu Details | |
| Menu Option: B - [+] Show Menu Details | |
| DaExaCellPatch - Cell Patching Issues | |
Required Cell Patching Validation logs.
Select one of the following options:
(1) Remote Cell/Compute Node Patching Logs
(2) Database Node Patching Logs
(3) Patching Logs using ILOM
(M)ain (B)ack (H)elp (E)xit
Showing Menu Help:
Exadata - Cell Issues: Cell Patching Issues Help
This menu includes options to collect important patching and validation
logs. Options 1 and 2 will collect cell-side and database-side data and
and generally run as a 2 step process. Option 3 - Collect Patching Logs
using ILOM, can be used when the cell is not accessible but the ILOM can
be accessed.
Explanation of available options:
(1) Collect Remote Cell/Compute Node Patching Logs
o This option requires "root" user credentials and captures
these files from the cell side:
/var/log/cellos/CheckHWnFWProfile.log
/var/log/cellos/validations.log
/var/log/cellos/checkconfigs_configuration_summaries.txt
/var/log/cellos/validations/upgradecbusb.log
/var/log/cellos/patch.out.place.log
/root/_patch_hctap_/_p_/wait_out
/root/_patch_hctap_/ all .log and .txt files
/boot/ all .log and .txt files
imageinfo
ipmitool event logs
/opt/oracle.SupportTools/CheckHWnFWProfile hardware and
firmware checks
/var/log/messages
/SP/console/history
/root/_patch_hctap_/ all .log and .txt files
/boot/ all .log and .txt files
imageinfo
ipmitool event logs
/opt/oracle.SupportTools/CheckHWnFWProfile hardware and
firmware checks
/var/log/messages
/SP/console/history
(2) Database Node Patching Logs
o This option captures patching logs from the node location
of where the patchmgr was last run
o Root user credentials are required
o Files collected:
{PATCHMGR_DIR}/patchmgr.stderr
{PATCHMGR_DIR}/patchmgr.stderr
{PATCHMGR_DIR}/patchmgr_output
{PATCHMGR_DIR}/.../...logs
(3) Collect Patching Logs using ILOM
o Select this option when the cell is not accessible but
the ILOM can be accessed
o Root user credentials are required
|
|
| Menu Option: 1 - [+] Show Command Details | |
| Menu Option: 2 - [+] Show Command Details | |
| Menu Option: 3 - [+] Show Command Details | |
| Menu Option: M - [+] Show Menu Details | |
| Menu Option: B - [+] Show Menu Details | |
| DaExaCellILOM - ILOM Issues | |
This menu will help diagnose your ILOM issues. Select option 1
to generate the snapshot followed by option 2 to collect option
1 data.
Select one of the following options:
(1) Generate ILOM Snapshot
(2) Collect ILOM Data
(M)ain (B)ack (H)elp (E)xit
Showing Menu Help:
Exadata - Cell Issues: ILOM Issues Help
Explanation of available options:
(1) Generate ILOM Snapshot
o Select this option to view instructions to manually generate
the ILOM Snapshot for a cell
o This process requires root access to the cell
(2) Collect ILOM Data
o Select this option to collect the ILOM Snapshot data generated
by step 1
o User will be prompted for the directory path and the node name
o The snapshot data will be included with all the other data
captured in the session zip file
|
|
| Menu Option: 1 - [+] Show Menu Details | |
| Menu Option: 2 - [+] Show Command Details | |
| Menu Option: M - [+] Show Menu Details | |
| Menu Option: B - [+] Show Menu Details | |
| DAexaCellILOMGen - Generate ILOM Snapshot | |
You must run the ILOM Snapshot outside the Diagnostic Assistant
at this time.
Please perform the following steps in order given.
Detailed instructions for each step are available in (H)elp.
1. ssh to [node name]-ilom and enter the password
2. Confirm header details
3. Run command: set /SP/diag/snapshot dataset=normal
4. Confirm response
5. Run set command to set output directory
6. Run command: cd /SP/diag/snapshot
7. Type command: show
8. Verify the status under properties is 'Running'
9. When Snapshot Complete, end ILOM session
(M)ain (B)ack (H)elp (E)xit
Showing Menu Help:
ILOM Issues: Generate ILOM Snapshot Help
Explanation of available options:
1 ssh to the [node name]-ilom and enter the password at the prompt
example: ssh cell01-ilom
password
2 You should see a header similar to the following output:
Sun(TM) Integrated Lights Out Manager
Version 3.0.9.19.a r55943
Copyright 2009 Sun Microsystems, Inc. All rights reserved.
Use is subject to license terms.
3 After the prompt,type this command:
set /SP/diag/snapshot dataset=normal
4 You will get a response of:
Set 'dataset' to 'normal'
5 After the prompt, type this command:
set /SP/diag/snapshot dump_uri=sftp://root:[password]@
[IP address of host]/[directory to upload snapshot to]
ex: set /SP/diag/snapshot dump_uri=sftp://root:
passwd@192.168.1.7/tmp
6 After the prompt, type this command:
cd /SP/diag/snapshot
7 Type command: show
8 Verify that the status shown under properties is 'Running'
example: show
/SP/diag/snapshot
Targets:
Properties:
dataset = normal
dump_uri = (Cannot show property)
encrypt_output = false
result = Running
9 Continue to enter the "show" command until these lines
appear under Properties, indicating the process is complete:
result = Collecting data into [sftp location from step 5]
Snapshot Complete.
Done.
10 To end the ILOM session, type command: exit
See My Oracle Support Knowledge Article 1062544.1: Diagnostic
information for ILOM, ILO , LO100 issues (https://support.oracle.com/
CSP/main/article?cmd=show&type=NOT&id=1062544.1) for additional
information.
|
|
| Menu Option: M - [+] Show Menu Details | |
| Menu Option: B - [+] Show Menu Details | |
| DaExaHW - exadata - Hardware Issues | |
Select one of the following options:
(1) Underperforming flash drive
(2) Failed Drives on Exadata Cell or Compute Node
(3) LSI MegaRAID Diagnostics
(M)ain (B)ack (H)elp (E)xit
Showing Menu Help:
Exadata: Exadata - Hardware Issues Help
Explanation of available options:
(1) Underperforming flash drive
o The 'Exadata_FlashDrives' RDA profile will execute
(2) Failed Drives on Exadata Cell or Compute Node
o The 'Exadata_FailedDrives' RDA profile will execute
(3) LSI MegaRAID Diagnostics
o The following diagnostics run various MegaCli64 Commands:
/opt/MegaRAID/MegaCli/MegaCli64 -h
|
|
| Menu Option: 1 - [+] Show Command Details | |
| Menu Option: 2 - [+] Show Command Details | |
| Menu Option: 3 - [+] Show Menu Details | |
| Menu Option: M - [+] Show Menu Details | |
| Menu Option: B - [+] Show Menu Details | |
| DaExaHWMegaRAID - LSI MegaRAID | |
Your LSI MegaRAID is about to be diagnosed.
Select one of the following options:
(1) Status of the battery(s)
(2) Status of the Virtual Drive(s)
(3) Status of the cache in the dbnodes
(M)ain (B)ack (H)elp (E)xit
Showing Menu Help:
Exadata - Hardware Issues LSI MegaRAID Help
The following diagnostics run various MegaCli64 Commands:
/opt/MegaRAID/MegaCli/MegaCli64 -h
Explanation of available options:
(1) Provides information about LSI battery state
MegaCli64 -AdpBbuCmd ?aALL
(2) Provides information about Virtual Drive(s)
MegaCli64 -LDInfo -Lall -aAll
(3) Verify the state of cache in the dbnodes
MegaCli64 -CfgDsply -a0 |grep -i writethrough
|
|
| Menu Option: 1 - [+] Show Command Details | |
| Menu Option: 2 - [+] Show Command Details | |
| Menu Option: 3 - [+] Show Command Details | |
| Menu Option: M - [+] Show Menu Details | |
| Menu Option: B - [+] Show Menu Details | |
| DaExaNetwork - Network Issues | |
Select one of the following options:
(1) Listener Hang/Performance
(2) Network cabling issue
(3) IB Switch Issues
(4) General Network Issues
(M)ain (B)ack (H)elp (E)xit
Showing Menu Help:
Exadata: Network Issues Help
Explanation of available options:
(1) Listener Hang/Performance
o The 'ONET' RDA module will execute
(2) Network cabling issue
o The 'Exadata_NetworkCabling' RDA profile will execute
(3) IB Switch Issues
o The 'Exadata_IbSwitch' RDA profile will be presented to
execute with or without root level data
(4) General Network Issues
o The 'NET' RDA module will execute
|
|
| Menu Option: 1 - [+] Show Command Details | |
| Menu Option: 2 - [+] Show Command Details | |
| Menu Option: 3 - [+] Show Menu Details | |
| Menu Option: 4 - [+] Show Command Details | |
| Menu Option: M - [+] Show Menu Details | |
| Menu Option: B - [+] Show Menu Details | |
| NetworkIBS - IB Switch | |
Your IB Switch is about to be diagnosed.
Select one of the following options:
(1) Run A New Collection
(2) Add root level data to the collection
(M)ain (B)ack (H)elp (E)xit
Showing Menu Help:
Network: IB Switch Help
Explanation of available options:
(1) Run A New Collection
o The 'Exadata_IbSwitch' RDA profile will execute without
root level data
(2) Add root level data to the collection
o The 'Exadata_IbSwitch' RDA profile will execute
o The following root level data will be included:
- root access data must be provided
- root data can only be added to an existing collection
- option 1 must be run at least once before this command
will execute successfully
|
|
| Menu Option: 1 - [+] Show Command Details | |
| Menu Option: 2 - [+] Show Command Details | |
| Menu Option: M - [+] Show Menu Details | |
| Menu Option: B - [+] Show Menu Details | |
| DAexaOraErr - Oracle Errors | |
For Oracle Errors, a 2 step process is needed to gather the incident
details. Option 1 will identify the incidents and option 2 will create
the incident package.
Select one of the following options:
(1) View incidents, limited by number
(2) Create incident package
(M)ain (B)ack (H)elp (E)xit
Showing Menu Help:
Exadata Detailed Diagnostics: Oracle Errors Help
Select option 1 to find the incident number needed for option 2.
Explanation of available options:
(1) Show incident, limited by number
o You will be prompted for MODE
- The 'MODE' of output includes BASIC, BRIEF or DETAIL where
each option provides a different level of detail
o you will be prompted for the number of incidents you wish to view
- default is 50
(2) Create package
o incident id is required
o this option will create a logical incident package
|
|
| Menu Option: 1 - [+] Show Command Details | |
| Menu Option: 2 - [+] Show Command Details | |
| Menu Option: M - [+] Show Menu Details | |
| Menu Option: B - [+] Show Menu Details | |
| DAexaProcess - Process Issues | |
Process issues will be diagnosed here.
Select one of the following options:
(1) Process Spins and Hangs with Procwatcher
(M)ain (B)ack (H)elp (E)xit
Showing Menu Help:
Exadata Detailed Diagnostics: Process Issues Help
Explanation of available options:
(1) Process Spins and Hangs with Procwatcher
o Procwatcher is a tool to examine and monitor Oracle database
and/or clusterware processes at an interval
o for issues or questions encountered while using this option,
please refer to My Oracle Support Knowledge Article 459694.1:
Script to Monitor and Examine Oracle DB and Clusterware Processes
(https://support.oracle.com/CSP/main/article?cmd=show&type=
NOT&id=459694.1) for assistance
|
|
| Menu Option: 1 - [+] Show Menu Details | |
| Menu Option: M - [+] Show Menu Details | |
| Menu Option: B - [+] Show Menu Details | |
| Procwatch - Procwatcher | |
Procwatcher is an external tool to examine and monitor Oracle
database and/or clusterware processes at an interval.
If you experience problems with Procwatcher, please contact the owner
of My Oracle Support Knowledge Article 459694.1: Script to Monitor and
Examine Oracle DB and Clusterware Processes (https://support.oracle.com/
CSP/main/article?cmd=show&type=NOT&id=459694.1) for assistance.
Select one of the following options:
(1) Procwatcher status
(2) Start Procwatcher
(3) Stop Procwatcher
(4) Package Procwatcher files
(M)ain (B)ack (H)elp (E)xit
Showing Menu Help:
Exadata Process Issues: Procwatcher Help
Procwatcher is an external tool to examine and monitor Oracle
database and/or clusterware processes at an interval.
If you experience problems with Procwatcher, please contact the owner
of My Oracle Support Knowledge Article 459694.1 - Procwatcher: Script
to Monitor and Examine Oracle DB and Clusterware Processes
(https://support.oracle.com/CSP/main/article?cmd=show&type=
NOT&id=1070954.1)for assistance.
The following options will prompt for the directory path where
Procwatcher is installed. Once prompted, the value is stored in
config/user.properties: DA_PROCWATCHER_DIR and you will not be
prompted again.
Explanation of available options:
(1) Procwatcher status
(2) Start Procwatcher
(3) Stop Procwatcher
(4) Package Procwatcher files
Additionally, if Procwatcher is deployed, all output will go to
GRID_HOME/log/procwatcher but it is possible to run prw.sh standalone
from any directory if it is not deployed.
If installed as standalone, and if prw finds clusterware active,
the output will go to GRID_HOME/log/procwatcher
|
|
| Menu Option: 1 - [+] Show Command Details | |
| Menu Option: 2 - [+] Show Command Details | |
| Menu Option: 3 - [+] Show Command Details | |
| Menu Option: 4 - [+] Show Command Details | |
| Menu Option: M - [+] Show Menu Details | |
| Menu Option: B - [+] Show Menu Details | |
| DAexaRAC - RAC Issues | |
This menu targets RCA issues.
Select one of the following options:
(1) RAC Instance or node Evictions
(2) Clusterware Startup Problems
(M)ain (B)ack (H)elp (E)xit
Showing Menu Help:
Exadata Detailed Diagnostics: RAC Issues Help
Explanation of available options:
(1) RAC Instance or node Evictions
o The 'Exadata_RacInstance' RDA profile will execute
This profile includes the following modules:
o OS
o LOG
o PERF
o RAC
o ASM
o EXA
(2) Clusterware Startup Problems
o This command collects Clusterware related log files
o You will be prompted for
- GRID_HOME path
- Date you are concerned about in YYY-MM-DD format
o This command collects Clusterware related log files:
- Current crs resource state
\$GRID_HOME/bin/crsctl stat res -t -init
- Clusterware log files by specific date:
$GRID_HOME/log/{hostname}/*.log
\$GRID_HOME/log/{hostname}/ohasd/ohasd.log
\$GRID_HOME/log/{hostname}/agent/ohasd/oraagent_\$USER/
oraagent_\$USER.log
\$GRID_HOME/log/{hostname}/agent/ohasd/orarootagent_root/
orarootagent_root.log
\$GRID_HOME/log/{hostname}/cssd/ocssd.log
\$GRID_HOME/log/{hostname}/crsd/crsd.log
\$GRID_HOME/log/{hostname}/agent/crsd/oraagent_\$USER/
oraagent_\$USER.log
\$GRID_HOME/log/{hostname}/agent/crsd/orarootagent_root/
orarootagent_root.log
\$GRID_HOME/log/{hostname}/diskmon/diskmon.log
\$GRID_HOME/log/{hostname}/gpnpd/gpnpd.log
|
|
| Menu Option: 1 - [+] Show Command Details | |
| Menu Option: 2 - [+] Show Command Details | |
| Menu Option: M - [+] Show Menu Details | |
| Menu Option: B - [+] Show Menu Details | |
| DAexaChk - exachk Collection | |
Exachk is designed to audit various important configuration settings
within an Exadata System - Database Servers, Storage Servers, etc.
Exachk MUST already be installed as this tool is NOT delivered
with the Diagnostic Assistant.
Select from the following options:
(1) Verify Exachk Configuration
(2) Run Exachk (see Help for pre-requisites)
(3) Run Exachk Without Root Privilege (see Help for pre-requisites)
(4) Collect Data After Manual DAexaChk Run
(M)ain (B)ack (H)elp (E)xit
Showing Menu Help:
Exadata Detailed Diagnostics: Exachk Help
Explanation of available options:
(1) Verify Exachk Configuration
o Select this option to see the pre-requisite configuration steps
that have to be performed to run the Exachk collection from the
Diagnostic Assistant
(2) Run Exachk
o Select this option to run Exachk in non-interactive mode
including root privileged checks
- root privileged checks on database servers will be executed
- root privileged checks on storage servers will be executed
- NO Infiniband Switch checks will be performed
Pre-requisite configuration steps 1-5 must have been performed
and verified before selecting this collection
(3) Run Exachk Without Root Privileges
o Select this option to run Exachk in non-interactive mode without
root privileged checks
- No root privileged checks on database servers will be executed
- No root privileged checks on storage servers will be executed
- NO Infiniband Switch checks will be performed
Pre-requisite configuration steps 1-3 must have been performed
and verified before selecting option 4 collection.
(4) Collect Data After ManualDAexaChk Run
o Select this option to collect data from an Exachk run
performed outside of the Diagnostic Assistant (i.e. manually
or from another tool)
o If you have performed option 2 or 3, you do not need to select
this option since the Exachk data will have been automatically
collected
|
|
| Menu Option: 1 - [+] Show Menu Details | |
| Menu Option: 2 - [+] Show Command Details | |
| Menu Option: 3 - [+] Show Command Details | |
| Menu Option: 4 - [+] Show Command Details | |
| Menu Option: M - [+] Show Menu Details | |
| Menu Option: B - [+] Show Menu Details | |
| DAexaChkVerify - Verify Exachk Configuration | |
To run the Exachk collection from the Diagnostic Assistant,
please verify the following pre-requisite configuration steps
have been performed.
Detailed instructions for each step are available in (H)elp.
1 configure ssh user equivalence for rdbms software owner (database)
2 configure ssh user equivalence for rdbms software owner (storage)
3 Set the environment variable RAT_CELL_SSH_USER
4 update the sudoers file on each database server
5 update the sudoers file on each storage server
(M)ain (B)ack (H)elp (E)xit
Showing Menu Help:
Exachk: Steps to Verify Exachk Configuration Help
Instructions and usage documented in the following:
o Knowledge Article 1070954.1: Oracle Exadata Database Machine Exachk
or HealthCheck (https://support.oracle.com/CSP/main/
article?cmd=show&type=NOT&id=1070954.1)
(1) configure ssh user equivalence for rdbms software owner
Example: oracle on execution database server to oracle user on
all other database servers.
o To validate proper ssh user equivalence configuration, login
as the RDBMS software owner (eg. oracle) and execute the
following commands:
> ssh -o NumberOfPasswordPrompts=0 -o StrictHostKeyChecking=no
-l oracle dbServerName "echo \"oracle user equivalence is
setup correctly\""
where oracle is the RDBMS software owner and dbServerName
is the database server hostname
o Repeat for each database server
o If "Permission denied (publickey,gssapi-with-mic,password)."
is returned then the ssh user equivalence is not properly
configured
(2) configure ssh user equivalence for rdbms software owner
(eg., oracle) from execution node to cellmonitor user on all
storage servers.
o To validate proper ssh user equivalence configuration login as
the RDBMS software owner (eg., oracle) and execute the
following commands:
>\$ ssh -o NumberOfPasswordPrompts=0 -o StrictHostKeyChecking=no
-l cellmonitor storageServerName "echo \"cellmonitor user
equivalence is setup correctly\""
where storageServerName is the storage server hostname.
o Repeat for each storage server
o If "Permission denied (publickey,gssapi-with-mic,password)."
is returned then the ssh user equivalence is not properly
configured
(3) Set the environment variable RAT_CELL_SSH_USER to cellmonitor in
the execution environment of the rdbms software owner (eg oracle)
# export RAT_CELL_SSH_USER=cellmonitor
(4) update the sudoers file on each database server
using visudo command:
>oracle ALL=(root) NOPASSWD:/tmp/root_Exachk.sh
(5) update the sudoers file on each storage server
using visudo command:
>cellmonitor ALL=(root) NOPASSWD:/tmp/Exachk_cells.sh
|
|
| Menu Option: M - [+] Show Menu Details | |
| Menu Option: B - [+] Show Menu Details | |