What Is Exachk In Exadata

Exachk and TFA are now a part of the autonomous health framework (AHF) and its report provides a holistic view of the Exadata machine. The report also contains the verification, explanation and manual verification commands so that we can evaluate the risk and self-correct conditions. Exachk reports includes:

  (1)  Configuration check for database server, storage server and network fabric switches which includes:

                         (A) Firmware

                            (B) Operating System

                            (C) Storage Server

                            (D) Grid Infrastructure and ASM

                            (E) Database


       (2) Maximum Availability Architecture Scorecard (MAA Score card)

                     (A) MAA configuration review

                        (B) Exadata Software Planner

                        (C) Exadata Critical Issue Alert


       (3) Automation Correction if applicable:

                     (A) Configuration Correction

                        (B) Critical issue avoidance


       (4) Pre-requisite check for Database and Grid Infrastructure Update


       (5) Pre-requisite check for database and grid upgrade


       (6) Pre-requisite check for application continuity check



                        Exachk Report Schedule

 

Exachk run in two modes one in full mode which run once in a week and other one is in exatier1 mode or we can also say cumulative mode which run on daily basis. By default, the Exachk full report runs at 03:03 AM on every Sunday and Exatier 1 reports run every day at 02:03 AM.

 

                                  Need to run full Exachk Report

 

Sine Exachk full report contain the entire Exadata machine information, alerts and recommendation. So, it takes a lot of time and if somehow, we’ve to run it then run it on off hour. For e.g.  during Patching we’ve to run the full exachk report 1 week prior to start of the patching activity and followed by Exatier 1 report during the day of the activity.

                                   Exatier 1 Report

 

By default, Exatier1 report runs every day but we should also run it in the following cases as well:

 (1)   Immediately after deployment

 (2)  After a configuration change

 (3)  Before or after any planned maintenance activity like Patching

 (4)  If oracle support need exachk report to troubleshoot the issue

          

                             How To Check Exachk Version

 

Since Exachk is now a part of AHF. So, we can check the version of ahf and it also shows the version of TFA and Exachk.

 

[root@exatest ~] #  ahfctl version -all

AHF version: 22.1.5

TFA version: 22.1.5

EXACHK  VERSION: 22.1.5_20220712

 

                                    OR

[root@exatest ~]#   exachk -v

EXACHK VERSION: 20.1.0(BETA)_20200220


                     How To Check The Location Of Exachk

 

[root@exatest ~]#  cat /etc/oracle.ahf.loc

/opt/oracle.ahf

                     Validate Exachk Schedule

 

[root@exatest ~]#  exachk -get all

 ------------------------------------------------------------

ID: exachk.autostart_client_exatier1

------------------------------------------------------------

AUTORUN_FLAGS  =  -usediscovery -profile exatier1 -dball -showpass -tag autostart_client_exatier1 -readenvconfig

COLLECTION_RETENTION  =  7

AUTORUN_SCHEDULE  =  3 2 * * 1,2,3,4,5,6

------------------------------------------------------------

------------------------------------------------------------

ID: exachk.autostart_client

------------------------------------------------------------

AUTORUN_FLAGS  =  -usediscovery -tag autostart_client -readenvconfig

COLLECTION_RETENTION  =  14

AUTORUN_SCHEDULE  =  3 3 * * 0

------------------------------------------------------------

 

                 How To Check Exachk Repository

 

[root@exatest ~]# ahfctl showrepo -compliance

Exachk repository: /u02/oracle.ahf/data/exacs1/exachk/user_root/output

[root@exatest ~]# ls -ltr /u02/oracle.ahf/data/exacs1/exachk/user_root/output
...
drwxr-x--- 8 root dba     307200 Sep  8 18:21 exachk_exacs1_psclone_ps_090822_175129_1938XL206N
-r--r----- 1 root root 121019389 Sep  8 18:22 exachk_exacs1_psclone_ps_090822_175129_1938XL206N.zip
drwxr-x--- 8 root dba      69632 Sep  8 18:50 exachk_exacs1_psclone_ps_090822_184356
-r--r----- 1 root root  99308675 Sep  8 18:51 exachk_exacs1_psclone_ps_090822_184356.zip
drwxr-x--- 8 root dba      69632 Sep  8 18:52 exachk_exacs1_psclone_ps_090822_184413
-r--r----- 1 root root  99145563 Sep  8 18:52 exachk_exacs1_psclone_ps_090822_184413.zip
drwxr-x--- 8 root dba      73728 Sep  8 18:53 exachk_exacs1_psclone_ps_090822_184517
-r--r----- 1 root root  99269484 Sep  8 18:53 exachk_exacs1_psclone_ps_090822_184517.zip
drwxr-x--- 4 root dba       4096 Sep  9 02:05 exachk_090922_02039_discovery
-rw-r----- 1 root root       997 Sep  9 02:11 exachk_daemon.log

                   

                    How To Check Exachk Report Retention

 

The report that Exachk generates has a retention period and the by default retention period of Exachk report is as follows

 

[root@exatest ~]#  exachk -get collection_retention

------------------------------------------------------------

ID: exachk.autostart_client_exatier1

------------------------------------------------------------

COLLECTION_RETENTION  =  7

------------------------------------------------------------

------------------------------------------------------------

ID: exachk.autostart_client

------------------------------------------------------------

COLLECTION_RETENTION  =  14

------------------------------------------------------------

 

NOTE: We can change the Exachk retention

 

                       Command To Run Exachk Manually

 

[root@exatest ~]#  exachk   ## To Execute full exachk report

 [root@exatest ~]#  exachk  -profile exatier1  ## To Execute Tier1 exachk report

  

Execute Exachk in a partitioned cluster

 In a large Exadata OVM is logically divided into smaller subsystem for e.g. A full RACK is divided into 2 half RACK then in that case if we’ve to run exachk report against any of the specific cluster then we’ve to specify cluster node,cells and the network switches because logically the cluster is divided but physically that cluster is one large Exadata cluster and if we don’t specify the cluster node,cells and the network switches then by default Exachk report collect entire Exadata cluster report.

 [root@exatest ~]#  exachk -profile -exatier1 -clusternodes exatest01,exatest02

-         Cells exatestcel01, exatestcel02, exatestcel03

-         Ibswitches exatestsw-ibb0, exatestsw-iba0

 

                              Exachk Report Criteria

 

Exachk report criteria is basically divides into 3 parts inside Exachk report such as warning, fail and critical

 Warning:  A warning check show the recommendation that improves system but doesn’t jeopardize the system.

 Fail: Following thing are reported under Fail checks

 

(1)   Config issue that impacts RTO/RPO

(2)  Config check that should be addressed in next quarterly update

(3)  MAA config check that improves availability, data integrity and disaster recovery.

(4)  Configuration software mismatch

 Critical: Following things are report under critical checks:

 

(1)   Disk corruption or data loss

(2)  System wide availability impact

(3)  Compromised redundancy or inability to restore redundancy

(4)  Inability to update software in a rolling manner

(5)  Intermittent wrong result with Exadata feature such as smart scan





Post a Comment

Previous Post Next Post