Saturday, August 15, 2020

AHF Cheatsheet


Do You Speak AHF ?


Oracle Autonomous Health Framework (AHF) presents the next generation of All in one solution that includes tools which work together autonomously 24x7 to keep database systems healthy and running. It does so while minimizing human reaction time using existing components like orachk, TFA and many more.
I started digging about this tool ever since I attended  Bill Burton‘s (lead dev for TFA  Collector feature at Oracle) presentation at RMOUG2020. As most of you, I had been asked numerous times to collect diagnostics and myriads of logs (IPS,RDA..etc) from MOS when I opened SRs. It just didn’t get better with years, so when I heard about that new thing that can merge Health and Diagnostics in one tool, I wanted to give it a try. I Hope you’ll find this AHF Cheat Sheet (blabla free) useful .

Advantages:

  • User friendly & real-time health monitoring, fault detection & diagnosis via a single interface

  • Secure consolidation of distributed diagnostic collections

  • Continuous availability

  • Machine learning driven, autonomous degradation detection, reduces your overheads (on both customer and oracle support Tiers)

  • TFA is still used for diagnostic collection and management and ORAchk/EXAchk for compliance check.

  • ORAchk/EXAchk now use the TFA secure socket and TFA scheduler for automatic checks (less overhead).


Direct links  :

 - AHF Config checks

 - Orachk

 - TFA

 - What's New in AHF 20.2

Installation Instructions

  1. As ROOT (recommended) : download the file,unzip and run the setup file (see AHF - (Doc ID 2550798.1)

    Requirement: AHF DATA directory : /5GB-10GB

$ unzip  AHF-LINUX_v20.1.1.BETA.zip -d AHF
$ ./AHF/ahf_setup 
common locations of ahf_loc is /opt/oracle/dcs/oracle.ahf or /u01/app/grid/oracle.ahf
  • Windows
    Download & unzip the Windows installer then run the following command (adapt the perlhome) :

C:> installahf.bat -perlhome D:\oracle\product\12.2.0\dbhome_1\perl
  • Syntax

$ ahf_setup [-ahf_loc AHF Location]
            [-data_dir AHF Repository]  -- Stores collections, metadata,etc (mandatory on 
Silent install). Location is either under ahf_loc
(5G)or TFA repo if outside Grid_Home,or above Grid
Base, or different. [-nodes node1,node2] -- default all [-extract[-notfasetup]] -- ahf_setup -extract [exachk|orachk] install
ORAchk/EXAchk without installing the rest of AHF.
-notfasetup option just extract TFA no install [-local] [-silent] [-tmp_loc directory] -- Default: /tmp [-debug [-level 1-6]] -- 1= FATAL, 2= ERROR, 3= WARNING, 4=INFO, 5=DEBUG,6= TRACE
  • Install with non default install or data location

[root$] ./ahf_setup [-ahf_loc install_dir] [-data_dir data_dir]
  • Install orachk/exachk without TFA

[root$] ahf_setup -extract orachk/exachk 
#OR ahf_setup -extract-notfasetup.
  • You can check the status of AHF after upgrade\installation using the below command

[root@nodedb1 ~]# tfactl status
.---------------------------------------------------------------------------------------------.
| Host   | Status of TFA | PID   | Port | Version   | Build ID             | Inventory Status |
+--------+---------------+-------+------+-----------+----------------------+------------------+
| Nodedb1| RUNNING       | 96324 | 5000 | 20.1.1.0.0| 20110020200220143104 | COMPLETE         |
| Nodedb0| RUNNING       | 66513 | 5000 | 20.1.1.0.0| 20110020200220143104 | COMPLETE         |
  • For advanced installation options, see User Guide .

  • You can manually uninstall AHF in case it got corruption   : 

    # rpm -e oracle-ahf 
    # tfactl uninstall

CONFIGURATION CHECKS

In this section we will first learn how to query current setting of AHF.
If you wish to synchronize your RAC nodes when a new node is added run the below :

 # tfactl syncnodes  
  • Tfactl print Commands

# tfactl  print 
  options [status|components|config|directories|hosts|actions|repository|suspendedips|protocols|smtp]

> example: Quick list scan listener logs directories when troubleshooting listener issues

# tfactl print directories -node node1 -comp tns -policy noexclusions
  • List AHF parameters

#  tfactl print config
-- you can also set or display a single parameter as shown below
# tfactl print version
# tfactl print hosts   
     Host Name : racnode1              
     Host Name : racnode2   
# tfactl get autodiagcollect       
# tfactl set autodiagcollect=ON  
  
  • List repositories

[root@nodedb1]# tfactl print repository    -- Shows the AHF data repositories .
[root@nodedb1]# tfactl showrepo -all       -- Same but with options(-TFA or -Compliance)
  .-----------------------------------------------------------------.
  |                             Myhost                              |
  +----------------------+------------------------------------------+
  | Repository Parameter | Value                                    | ====> -TFA
  +----------------------+------------------------------------------+
  | Location             | /u01/app/grid/oracle.ahf/data/repository |
  | Maximum Size (MB)    | 10240                                    |
  | Current Size (MB)    | 3                                        |
  | Free Size (MB)       | 10237                                    |
  | Status               | OPEN                                     |
  '----------------------+------------------------------------------'
 orachk repository: /u01/app/grid/oracle.ahf/data/nodedb1/orachk   ====> -complaince(Orachk)
- can also be checked here:
[root@nodedb1]# cat /opt/oracle/dcs/oracle.ahf/tfa/tfa.install.properties

Note : If the maximum size is exceeded or the file system free space gets to <=1 GB , then AHF suspends operations and closes the repository.
Use the tfactl purge command to clear collections from the repository.

[root@server] tfactl purge -older n[h|d] [-force]
[root@server] tfactl purge -older 30d
[root@server] tfactl purge -older 10h  
-- Change retention 
[root@server] tfactl set minagetopurge=48        -- default is 2 weeks
  • Directories

       You can see it during the installation or from the config file

[root@nodedb1]# cat /opt/oracle/dcs/oracle.ahf/tfa/tfa.install.properties

.-----------------------------------------------------------------------------------------------. | Summary of AHF Configuration | +-----------------+----------------------------------------------------+------------------------+ | Parameter | Value | Content | +-----------------+----------------------------------------------------+------------------------' | AHF Location | /opt/oracle/dcs/oracle.ahf | installation diretory | | TFA Location | /opt/oracle/dcs/oracle.ahf/tfa | | | Orachk Location | /opt/oracle/dcs/oracle.ahf/orachk | | | Data Directory | /opt/oracle/tfa/tfa_home/oracle.ahf/data |Logs from all components| | Repository | /opt/oracle/tfa/tfa_home/oracle.ahf/data/repository|Diagnostic collections | | Diag Directory | /opt/oracle/tfa/tfa_home/oracle.ahf/data/host/diag |Logs from all components| | Analyzer Dir | /opt/oracle/dcs/oracle.ahf/tfa/analyzer | | '-----------------+----------------------------------------------------'------------------------'

Tools

To view the status of Oracle TFA Support Tools (across nodes if RAC) hit the below command.

[oracle@server]  tfactl toolstatus 
.-------------------------------------------------------------------------------------------------------------------------------------------------------+.
|                  TOOLS STATUS - HOST : - My Host                                                                                                      | 
+---------------------+--------------+------------------------------------------------------------------------------------------------------------------| 
| Tool Type           | Tool         | Description                                                                                                      | 
+---------------------+--------------+------------------------------------------------------------------------------------------------------------------| 
| Development Tools   | orachk       | Oracle AHF installs either Oracle EXAchk for engineered systems or Oracle ORAchk for all non-engineered systems. | 
|                     | oratop       | Provides near real-time database monitoring                                                                      | 
+---------------------+--------------+------------------------------------------------------------------------------------------------------------------| 
|Support Tools Bundle | darda        | Diagnostic Assistant:Provides access to  RDA, ADR, OCM                                                           | 
|                     oswatcher(oswbb| Collects/archives OS metrics. metrics are useful for instance/node evictions and performance Issues.             | 
|                     | prw          | procwatcher (prw) Automates and captures database performance diagnostics and session level hang information     | 
+-------------------- +--------------+--------------+---------------------------------------------------------------------------------------------------| 
| TFA Utilities       | alertsummary |Provides summary of events for one (-database x) or more or Oracle ASM alert files from all nodes.                | 
|                     | calog        |Prints Oracle Clusterware activity logs.                                                                          | 
|                     | dbcheck      |                                                                                                                  | 
|                     | dbglevel     |Sets Oracle Clusterware log / trace levels using profiles.                                                        | 
|                     | grep         |grep for input string in logs                                                                                     | 
|                     | history      | Lists commands run in current Oracle Trace File Analyzer shell session.                                          | 
|                     | ls           | Searches files in Oracle Trace File Analyzer                                                                     | 
|                     | managelogs   | Purge slogs                                                                                                      | 
|                     | menu         | Oracle Trace File Analyzer Collector menu system.                                                                | 
|                     | param        | Prints parameter value.                                                                                          | 
|                     | ps           | Finds a process.                                                                                                 | 
|                     | pstack       | Runs pstack on a process                                                                                         | 
|                     | summary      | Prints system summary.                                                                                           | 
|                     | tail         | Tails log files.                                                                                                 | 
|                     | triage       | Summarize OSWatcher / ExaWatcher data. /tfactl.pl [run] triage -t <datetime> -d <duration> [-a] [-p <pid>] [-h]  | 
|                     | vi           | Searches and opens files in the vi editor.                                                                       | 
'---------------------+--------------+-----------------------------------------------------------------------------------------------------------------+ 

AHF(TFA) Main Menu

The Oracle TFA Menu provides an Interface to the TFA tools.

 # tfactl menu
 1. System Analysis
       Select one of the following categories:     
         1. Summary
         2. Events
         3. Analyze logs
         4. TFA Utilities
         5. Support Tool Bundle
         6. Tools status             
  2. Collections
      Select one of the following options:      
        1. Default Collection
        2. SRDC Collection
        3. Engineered System Collection
        4. Advanced Collection     
  3.Administration
    Select One of the following options:      
      1. Version & status
      2. Start, stop & auto start
      3. Hosts & ports
      4. Settings
      5. Actions submitted
      6. Manage Database logs
      7. Tracing Level
      8. Users
      9. Collect TFA Diagnostic Data

Components

ORAchk

Autostart for either ORAchk/Exachk will perform compliance tasks as described bellow:

  • 1 am daily : Orachk/Exachk daemon restarts everyday to discover any environment changes .

  • 2 am daily : Partial orachk run > The track most severe problems once a day ( oratier1/exatier1)

  • 3 am Sunday: Full Exa/Orachk > All known problems once a week .

  • Send an email to the notification address for each run with the health scores, check failures and changes since last run.

  • The daemon also automatically purges the repository every 2 weeks.

Check Orachk Daemon

--- Displays information about the running daemon.
$ orachk -d info                                       
--- Checks if the daemon is running. OR ()$ tfactl compliance -d status )
$ orachk -d status                 
--- check orachk daemon status 
$ tfactl statusahf -compliance orachk scheduler is running [PID: 99003] [Version: 20.1.0(BETA)]
--- Start/Stop the daemon
$ orachk -d start/stop 
$ tfactl stopahf/startahf -compliance

--- Displays the next automatic run time.  
$ orachk -d nextautorun
$ orachk –d ID autostart_client nextautorun   
-- specific profile id:(autostart_client/os/etc)    

Getting Options for the Daemon

SET/GET Options :
 - AUTORUN_SCHEDULE, NOTIFICATION_EMAIL,COLLECTION_RETENTION, 
    AUTORUN_FLAGS,PASSWORD_CHECK_INTERVAL
  • Orachk Profiles :
    Each compliance Task has a dedicated profile
    1- autostart_discovery : First time a demon started
    2- daemon_autostart : daily restart at 1am
    3- autostart_client_oratier1 : partial run 2am
    4- autostart_client : Full run on Sunday 3am
[root@nodedb1 ~]# orachk -get  NOTIFICATION_EMAIL,AUTORUN_SCHEDULE,COLLECTION_RETENTION
------------------------------------------------------------
ID: orachk.autostart_discovery
------------------------------------------------------------
ID: orachk.daemon_autostart
------------------------------------------------------------
NOTIFICATION_EMAIL  =   
AUTORUN_SCHEDULE  =  0 1 * * *
------------------------------------------------------------
ID: orachk.autostart_client_oratier1
------------------------------------------------------------
NOTIFICATION_EMAIL  =  user1@example.com
AUTORUN_SCHEDULE  =  0 2 * * *
COLLECTION_RETENTION  =  7
------------------------------------------------------------
ID: orachk.autostart_client
------------------------------------------------------------
NOTIFICATION_EMAIL  =  user1@example.com
AUTORUN_SCHEDULE  =  0 3 * * 0
COLLECTION_RETENTION  =  14
------------------------------------------------------------

You can also query the parameter per profile:

$ orachk –id autostart_client -get all   --- Query daemon options for weekly profile


Setting Orachk Options

  • Set autostart option (enabled by default after installation)

$ orachk –autostart
orachk daemon is started with PID : 99003
 Daemon log file location 
    is : /opt/oracle/tfa/tfa_home/oracle.ahf/data/nodedb1/orachk/orachk_daemon.log   
OR
$ tfactl startahf -compliance
  • Notification for Orachk/Exachk

=====================================
For both daily and weekly job profile 
======================================
$ orachk -id autostart_client_oratier1 -set  NOTIFICATION_EMAIL=user1@example.com
$ orachk -id autostart_client  -set  NOTIFICATION_EMAIL=user1@example.com
  • Schedule and Retention

$ orachk -id autostart_client_oratier1 –set "AUTORUN_SCHEDULE=0 3 * * *"  -> Time= 3 AM daily
$ orachk -id autostart_client  –set "collection_retention=7"
  • TEST the Notification

--- Sends a test email to validate email configuration. 
$ orachk -testemail notification_email=user1@example.com --- Send an orachek report by email to a test address
$ orachk –sendemail "notification_email=user1@example.com"
  • Run Orachk

$ orachk -h 
  -nopass           --- Does not show passed checks.
  -show_critical    --- Show Critical checks in the orachk report by default 
  -localonly        --- only on the local node.  
  -dbnames db_names --- Specify a list of DB names to run only on a subset of DBs.
  -dbnone           --- skips dbs
  -dball            --- all Dbs
  -b                --- Runs only the best practice checks.no recommended patch checks.
  -p                --- Runs only the patch checks.
  -m                --- Excludes Maximum Availability Architecture (MAA) checks.
  -diff             --- compare two orachk reports
  -clusternodes  dbadm01,dbadm02,dbadm03,dbadm04
  -nordbms          --- Runs in Grid environments with no Oracle Database.
  -cvuonly          --- Runs only Cluster Verification Utility (oposite= -nocvu)
  -failedchecks  <previous_run.html> ---Runs only checks from the presious_result,that failed 
        -profile profile/-excludeprofile peoplesoft,seibel,storage,dba,asm,control_VM,
preinstall,prepatch,hardware,goldengate,oratier1,
virtual_infra,obiee,ebs,control_VM,.. -showrepair <check_id> --- Show repair command for given check id.. -repair all| <check_id>,[<check_id>,..] | <file> -- Repair check(s). -nordbms --- Runs in Grid environments with no Oracle Database. -cvuonly --- Runs only Cluster Verification Utility -showrepair <check_id> --- Show repair command for given check id. -pre/postupgrade --- run pre/post-upgrade best practice checks for 11.2.0.4 databases
                            and above     

Exadata :   -ibswitches switches
             -cells cells

TFA

  • Trace File Analyzer will monitor for significant issues and If any is detected it will automatically:

    1. Take a diagnostic collection of everything needed to diagnose & resolve (cluster wide where necessary)

    2. Collocate distributed collections and Analyze collections for known problems

    3. Send an email notification, containing details of the issues detected, recommended solution where known and location of diagnostic collection(s)

  • Notification for TFA and Orachk/Exachk (Doesn’t always work)

$ tfactl set ahfnotificationaddress=another.body@example.com 
$ tfactl get ahfnotificationaddress
AHF Notification Address : another.body@example.com
  • Notification for TFA only

        Automatically diagcollect upon detected fault and send result to registered email address

$  tfactl set notificationaddress=another.body@example.com   --- for any ORACLE_HOME:
$  tfactl set notificationAddress=os_user:email     --- for the os owner of a specific home  
  • Smtp configuration

# tfactl set smtp.
    
   .---------------------------.
   | SMTP Server Configuration |
   +---------------+-----------+
   | Parameter     | Value     |
   +---------------+-----------+
   | smtp.auth     | false     |Set the Authentication flag to true or false.
   | smtp.user     | -         |
   | smtp.from     | tfa       |Specify the from mail ID.
   | smtp.cc       | -         |Specify the comma-delimited list of CC mail IDs
   | smtp.port     |        25 |
   | smtp.bcc      | -         |Specify the comma-delimited list of BCC mail IDs.
   | smtp.password | *******   |
   | smtp.host     | localhost |
   | smtp.debug    | true      |Set the Debug flag to true or false 
   | smtp.to       | -         |Specify the comma-delimited list of recipient mail IDs
   | smtp.ssl      | false     |Set the SSL flag to true or false.
   '---------------+-----------'   
i.e : chose smtp.from to specify the sender email ID
run : tfactl print smtp to view the current setting.                                

RUN TFA:

  • Diagnostic Repository

1- Check Automatic Diagnotic repository space usage (all homes are checked if -database is omitted)

$ tfactl managelogs -show usage -database Mydb     ---per database
$ tfactl managelogs -show usage -gi    -- Grid logs
--- Capacity management view 
$ tfactl managelogs -show variation -older 7d -database mydb 

2- Purge ADR directories for a database if the FS is filled

$ tfactl managelogs -purge -older 1d(day)/h(hour)/m(minute) -database mydb
--- Change the retention permanently 
$ tfactl set manageLogsAutoPurgePolicyAge=20d      --- default is 30 days 
  • USING SUPPORT TOOLS

      Analyze lists events summary and search strings in alert logs. Parsing the database, ASM,Grid alert logs, system message logs,
      OSWatcher Top, and OSWatcher Slabinfo files.

$ tfactl analyze -comp db -database pimsprd -type error -last *[h/d] 
–o analyze_output_pimpsrd.log options : [-comp <db|asm|crs|acfs|os|osw|oswslabinfo|oratop|all>] [-search "pattern"] [-node all| local| n1,n2,.., -from| -to| -for "MMM/DD/YYYY HH24:MI:SS"
--- Searches alert and system log files from the past two days for messages
that contain the case-insensitive string "error". $ tfactl analyze -search "error" -last 2d
--- all system log messages for July 1, 2016 at 11 am.
$ tfactl analyze -comp os -for "Jul/01/2016 11" -search "."    
--- database alert logs for the case-sensitive string "ORA-" from the past two days. 
$ tfactl analyze -search "/ORA-/c" -comp db -last 2d           
--- Summary of events collected from all alert logs and system messages
from the past five hours. (-type generic for all generic message)
$ tfactl analyze -last 5h   
--- OS summary of events from system messages from the past day.
$ tfactl analyze -comp os -last 1d
--- Top summary for the past six hours.                             
$ tfactl analyze -comp osw -last 6h                            

Summary overview, Alert log, Events and changes checks

---- view the summary 
$ tfactl summary [defaul= Complete Summary Collection] options :-crs,asm,acfs,database,exadata,patch,listener,network,os,tfa,
summary,-node <node(s)>
--- summary of important events in DB/ASM logs. $ tfactl alertsummary -database Mydb
--- view the events detected by TFA.has more options(–component asm|crs –database <db>)
$ tfactl events -database oidprd  -last 1d
--- Lists all changes in the system $ tfactl changes -last 1d |-from x -to y | -for t] $ tfactl grep "Starting oracle instance" alert $ tfactl grep "ORA-" alert $ tfactl tail -200 listener.log $ tfactl tail -1000 listener_scan2.log | grep error $ tfactl grep ORA- listener_scan1.log
  • Miscellaneous

$ tfactl oratop -database mydb 
$ tfactl prw start/stop   -- Procwatcher session                          
-- explore the stack of a process
$ tfactl pstack PID
--- Stop a running Oracle Trace File Analyzer collection. $ tfactl collection stop <collectionid> --- Add/remove directory to/from the list of directories to analyze their trace or log files.
$ tfactl directory add|remove|modify
--- Collect Automatic Diagnostic Repository diagnostic data.   
$ tfactl ips show problem/incident
--- Enable auto diagnostic collection
$ tfactl set autodiagcollect=ON           
--- To configure Oracle Cluster Health Advisor auto collection for ABNORMAL events:
$ tfactl set chaautocollect=ON
--- To enable Oracle Cluster Health Advisor notification through TFA:
$ tfactl set chanotification=on
--- To configure an email address for Cluster Health Advisor notifications  :
$ tfactl set notificationAddress=chatfa:john.doe@acompany.com
--- Set Oracle Grid Infrastructure trace levels. (-active database_resource_failure) $ tfactl dbglevel

IPS (Incident Packaging Services) functions

Photos credit : @markusdba

  • BLACKOUT ERROR

--- target type listene/asm/database/service/os/asmdg/crs/all 

$ tfactl blackout add -targettype database -target -db Mydb -event "ORA_0431"

--- Default  timeout 24h values can be h(hour)/d(day)
                                             
$ tfactl blackout add -targettype all -target all -event all -timeout 1h -reason "Patching"
--- Show all backed out items  
$ tfactl blackout print        
--- To remove the blackout after patching                                              
$ tfactl blackout remove -targettype all -target all -event all      
  • Data Redaction through masking and sanitizing before uploading in a SR

# orachk -profile asm -sanitize
# orachk -sanitize file_name
# tfactl diagcollect -srdc ORA-00600 -mask  (inner Data)
# tfactl diagcollect -srdc ORA-00600 -sanitize (Hardware environment)
--- Reduce the size of the collections
# tfactl set MaxFileCollectsize 500MB
  • DIAGCOLLECT

Collect logs from across nodes in a cluster

Usage : tfactl diagcollect [ [component_name1] [component_name2] ... [component_nameN] | 
[-srdc <srdc_profile>] | [-defips]]  [-node <all|lom <time> -to <time> | -for <time>] 
[-nocopy] [-notrim] [-silent] [-nocores][-collectalldirs][-collectdir <dir1,dir2..>][-sanitize|-mask][-examples]
    
components : -ips|-oda|-odalite|-dcs|-odabackup|-odapatching|-odadataguard|-odaprovisioning|
-odaconfig|-odasystem|-odastorage|-database|-asm|-crsclient|-dbclient|-dbms|
-ocm|-emplugins|-em|
-acfs|-install|-cfgtools|-os|-ashhtml|-ashtext|-awrhtml|-awrtext|-qos
-srdc Service Request Data Collection (SRDC). -defips Include in the default collection the IPS Packages for: ASM, CRS and Databases -node Specify comma separated list of host names for collection -tag <tagname> The files will be collected into tagname directory inside repository -z <zipname> The collection zip file will be given this name within the TFA collection repository -last <n><m|h|d> Files from last 'n' [m]inutes, 'n' [d]ays or 'n' [h]ours -since Same as -last. Kept for backward compatibility. -from "Mon/dd/yyyy hh:mm:ss" From <time> or "yyyy-mm-dd hh:mm:ss" or "yyyy-mm-ddThh:mm:ss" or "yyyy-mm-dd" -to "Mon/dd/yyyy hh:mm:ss" To <time> or "yyyy-mm-dd hh:mm:ss" or "yyyy-mm-ddThh:mm:ss" or "yyyy-mm-dd" -for "Mon/dd/yyyy" For <date>. or "yyyy-mm-dd" -nocopy Does not copy back the zip files to initiating node from all nodes -notrim Does not trim the files collected -silent This option is used to submit the diagcollection as a background process -nocores Do not collect Core files when it would normally have been collected -collectalldirs Collect all files from a directory marked "Collect All" flag to true -collectdir Specify comma separated list of directories and collection will include all files from these irrespective of type and time constraints in addition to components specified -sanitize Sanitize sensitive values in the collection using ACR -mask Mask sensitive values in the collection using ACR -examples Show diagcollect usage examples

components: 
-database|-asm|-dbms|-ocm|-em|-acfs|-install|-cfgtools|-os|-ashhtml|-ashtext|-awrhtml|-awrtext|-ips|-oda|-odalite|-dcs|-odabackup|-odapatching|-odadataguard|-odaprovisioning|-odaconfig|-odasystem|-odastorage|-crsclient|-dbclient|-emplugins|-qos

Photos credit : @markusdba

Examples

--- Trim/Zip all files updated in the last 12 hours along with chmos/osw data across the cluster & collect at the initiating node (on-demand diagcollection)  simplest way to capture diagnostics 
$ tfactl diagcollect

--- Collecting ASH data for all nodes $ tfactl diagcollect -ashhtml -database MYDB -last 2m

--- Collecting AWR data for all nodes
$ tfactl diagcollect -awrhtml -database MYDB -last 2m

--- Trim and Zip all files updated in the last 8 hours as well as chmos/osw data from across the cluster
$ tfactl diagcollect -last 8h 

--- Trim/Zip files from DBs hrdb & fdb
$ tfactl diagcollect -database hrdb,fdb -last 1d -z foo    

--- Trim/Zip all crs files, o/s logs and chmos/osw data from node1 & node2 updated in the last 6 hours
$ tfactl diagcollect -crs -os -node node1,node2 -last 6h

--- Trim and Zip all ASM logs from node1 updated between from and to time
$ tfactl diagcollect -asm -node node1 -from "Mar/29/2020" -to "Mar/29/2020 21:00:00"

--- Trim and Zip all log files updated on "Mar/29/2020" and collect at the collect at the initiating node
$ tfactl diagcollect -for "Mar/29/2020"

--- Trim / Zip all log files updated from 09:00 on "Mar/29/2020" to 09:00 on "Mar/30/2020"
(i.e. 12 hours before and after the time given)
$ tfactl diagcollect -for "Mar/29/2020 21:00:00"

--- Trim / Zip all crs files updated in the last 12 hours.Also collect all files from /tmp_dir1 and /tmp_dir2 at the initiating node
$ tfactl diagcollect -crs -collectdir /tmp_dir1,/tmp_dir2


SRDC
For certain types of problems Oracle Support will ask you to run a Service Request Data Collection (SRDC) instead of reading and gathering different data across your system. At the end of the collection the file is automatically uploaded to the SR.

- Syntax :
$ tfactl diagcollect -srdc “srdc_type” -sr SR#

  • Description of some srdc


-

Photos credit : @markusdba

Upload diagnostic information to My Oracle Support

Configuration 
====================
$ tfactl setupmos
   User Id: JohnDO@email.com
   Password:
  - CERTIMPORT - Successfully imported certificate --will generate generate a wallet
== Run diagcollect for a service request======
$ tfactl diagcollect -srdc ORA-00600 -sr 2-39797576867  -sanitize -mask
==  upload the collected data to MOS
$ tfactl upload -sr 2-39797576867 -wallet <list_of_files> 
$ tfactl upload -sr 2-39797576867  -user JohnDoe@mail.com <list_of_files>   

Note : A full list of SRDC profiles can be checked here.

What's New in AHF 20.2

  • Resource savings through fewer repeat collections. Run below command to limit or stop Oracle TFA collecting the same events in a given time frame.

    $ tfactl floodcontrol
  • Limit TFA to a maximum of 50% of a single CPU use (default is 1, max value can be 4 or 75% of available CPUs, whichever is lower )

    $ tfactl setresourcelimit -value 0.5
  • Easier to upload diagnostic collections. For example to configure ORAchk uploads to be stored in a database you can use:

    $ tfactl setupload -name mysqlnetconfig -type sqlnet
    Defaults to ORACHKCM user
    Enter mysqlnetconfig.sqlnet.password :
    Enter mysqlnetconfig.sqlnet.connectstring : (SAMPLE=(CONNECT=(STRING)))
    Enter mysqlnetconfig.sqlnet.uploadtable : MYtab
  • Run all database checks without root password (i.e oracle )

    $ tfactl access promote -user oracle
    Successfully promoted access for 'oracle'.
    $ tfactl access lsusers
    .--------------------------------.
    |      TFA Users in odadev01     |
    +-----------+---------+----------+
    | User Name | Status  | Promoted |
    +-----------+---------+----------+
    | oracle    | Allowed | true     |
    '-----------+---------+----------'
    @oracle$ tfactl analyze -last 5h  -runasroot

5 comments:

  1. That is a impressive list, thanks for sharing, will keep it close!!!

    ReplyDelete
    Replies
    1. Glad you liked it Victor your were the first to read the draft :).

      Delete
    2. Didn't know about new option about cpu restrictions, that is great, specially after the 19 version having cpu "leaks" issues, really great resource

      Delete
    3. Yes, that's a brand new option I noticed last time I updated AHF for a client. One more thing I didn't mention is that the scheduled orachck jobs may not work for some environments as it depends on TFA's scheduler now (not sure why but I will share more once MOS SR is solved) .

      Delete