Tales From A Lazy Fat DBA

Den of an Oracle DB Performance freak & a fan of Cassandra, MySQL and PostgreSQL …

  • Likes

    • 138,101
  • Archives

  • Categories

  • Subscribe

  • Advertisements
  • Prashant Dixit is the FatDBA

  • Follow Tales From A Lazy Fat DBA on WordPress.com
  • My Twitter Feeds

  • Disclaimer!

    FatDBA or Oracle ‘Ant’ is an independent web-blog/site.The experiences, Test cases, views, and opinions expressed in this website are my own and does not reflect the views or opinions of my employer.

    This site is independent of and does not represent Oracle Corporation in any way. Oracle does not officially sponsor, approve, or endorse this site or its content.
    Product and company names mentioned in this website may be the trademarks of their respective owners.

Online Patching & Bug Fixing — > : Bug 16342845 : EXCESSIVE CPU IN DBW PROCESSES FOR FAST OBJECT CHECKPOINTSDB

Posted by FatDBA on March 2, 2015

During further investigation we found that all the Database Writers started consuming high resources and waits on ‘DB FILE ASYNC I/O SUBMIT’ event and all rest of the database sessions started waiting on ‘enq: KO – Fast Object Checkpoint’

We have one most critical production system running on oracle 11.2.0.3 on the top of RHEL.
Recently we started experiencing a very bizarre and weird issue where applications timeouts happens intermittently and during that specific period extremely slow IO operations happens at the database level.

While investigation found that during the issue DB started throwing huge and frequent waits on ‘enq: KO – fast object checkpoint’ and ‘db file async I/O submit’ events and all of the queries in the database during the period starts snailing.
At the same time all the 8 available DBWR processes started waiting on “db file async I/O Submit” wait event and seems takes time to write blocks to the disk and this leads to late check-pointing in the system which in turn slow down database.

DB_Waits_Stats-duringSlowness-onetmproduction___mmma

From AWR
—————
foreground wait event
enq: KO – fast object checkpoint — 70% of the DB time.

background wait event
db file async I/O submit — 90% of bg time

Wait Event Histogram
———————-
db file async I/O submit — 100% >1s (% of waits)
enq: CR – block range reuse ckpt — 100% >1s (% of waits)
enq: KO – fast object checkpoint — 100% >1s (% of waits)
enq: RO – fast object reuse — 100% >1s (% of waits)

After weeks of probing and hours of analyzing system state dumps, Hang analyze and writer process traces we reached to a conclusion that we have strike by a Bug 16342845 : EXCESSIVE CPU IN DBW PROCESSES FOR FAST OBJECT CHECKPOINTSDB

Below mentioned are steps performed to apply the recommended patch.
PRIMARY: disable log shipping
STANDBY: stop MRP

1. Disable Log Shipping on Primary Database.
2. Stop MRP process on Physical Standby Database.
3. Apply the patch first in to Physical Standby (online)
4. Monitor the performance on the Phy STDBY db.
5. Will apply the patch on Logical Standby Database.
6. Monitor the logical database performance.
7. Will apply the patch on Primary Database.
8. Monitor the system performance.
9. Start Log shipping on Primary DB end.
10.Start MRP process on Standby database.

Step 1.
Disable log shipping on Primary DB.

SQL> show parameter LOG_ARCHIVE_DEST_

NAME TYPE VALUE
———————————— ——————————– ——————————
log_archive_dest_1 string location=”/vol5/oracle/dixit_db_
arc/dixitdb/arch”, valid_for=(A
LL_LOGFILES,ALL_ROLES)

log_archive_dest_2 string service=”dixitdb_al”, LGWR ASYNC
NOAFFIRM delay=0 optional com
pression=disable max_failure=0
max_connections=1 reopen=300
db_unique_name=”dixitdb_al” net_
timeout=30, valid_for=(all_log
files,primary_role)

log_archive_dest_3 string service=”dixitdb_r”, NOAFFIRM AS
YNC VALID_FOR=(ALL_LOGFILES,PR
IMARY_ROLE) DB_UNIQUE_NAME=one
tm_r

alter system set log_archive_dest_state_2=defer scope=both;

Step 2:
Stop MRP process on Physical Standby Database.

Before
PROCESS STATUS DELAY_MINS
——— ———— ———-
MRP0 WAIT_FOR_LOG 0

SQL> alter database recover managed standby database cancel;

After cancelling recover process the MRP automatically stops.

PROCESS STATUS DELAY_MINS
——— ———— ———-

Step 3:
Opatch online apply – common to all environments.

Recommendations
1. Take binaries backup (ORACLE_HOME).
2. Take Full DB Backup.
3. Check Opatch Prerequisites in order to identify and resolve any patch conflicts.

-bash-3.2$ export PATH=$PATH:$ORACLE_HOME/OPatch

-bash-3.2$ opatch prereq CheckConflictAgainstOHWithDetail -ph ./
Oracle Interim Patch Installer version 11.2.0.3.0
Copyright (c) 2012, Oracle Corporation. All rights reserved.

PREREQ session

Oracle Home : /opt/app/oracle/product/11gR2
Central Inventory : /opt/app/oracle/oraInventory
from : /opt/app/oracle/product/11gR2/oraInst.loc
OPatch version : 11.2.0.3.0
OUI version : 11.2.0.3.0
Log file location : /opt/app/oracle/product/11gR2/cfgtoollogs/opatch/opatch2015-03-01_05-34-42AM_1.log

Invoking prereq “checkconflictagainstohwithdetail”

Prereq “checkConflictAgainstOHWithDetail” passed.

-bash-3.2$ opatch apply online -connectString dixitdb_al:sys:xxxxxxx
Oracle Interim Patch Installer version 11.2.0.3.0
Copyright (c) 2012, Oracle Corporation. All rights reserved.

Oracle Home : /opt/app/oracle/product/11gR2
Central Inventory : /opt/app/oracle/oraInventory
from : /opt/app/oracle/product/11gR2/oraInst.loc
OPatch version : 11.2.0.3.0
OUI version : 11.2.0.3.0
Log file location : /opt/app/oracle/product/11gR2/cfgtoollogs/opatch/20505138_Mar_01_2015_05_35_29/apply2015-03-01_05-35-29AM_1.log

The patch should be applied/rolled back in ‘-all_nodes’ mode only.
Converting the RAC mode to ‘-all_nodes’ mode.
Applying interim patch ‘20505138’ to OH ‘/opt/app/oracle/product/11gR2’
Verifying environment and performing prerequisite checks…
All checks passed.
Provide your email address to be informed of security issues, install and
initiate Oracle Configuration Manager. Easier for you if you use your My
Oracle Support Email address/User Name.
Visit http://www.oracle.com/support/policies.html for details.
Email address/User Name:

You have not provided an email address for notification of security issues.
Do you wish to remain uninformed of security issues ([Y]es, [N]o) [N]: Y
Backing up files…

Patching component oracle.rdbms, 11.2.0.3.0…
Installing and enabling the online patch ‘bug20505138.pch’, on database ‘dixitdb_al’.

Verifying the update…
Patch 20505138 successfully applied
Log file location: /opt/app/oracle/product/11gR2/cfgtoollogs/opatch/20505138_Mar_01_2015_05_35_29/apply2015-03-01_05-35-29AM_1.log

OPatch succeeded.


From Alert log during the patch apply:

Sun Mar 01 06:17:13 2015
Patch file bug20505138.pch is out of sync with oracle binary; performing fixup
Patch file bug20505138.pch has been synced with oracle binary
Patch bug20505138.pch Installed – Update #1
Patch bug20505138.pch Enabled – Update #2
Sun Mar 01 06:17:15 2015
Online patch bug20505138.pch has been installed
Online patch bug20505138.pch has been enabled

-bash-3.2$ opatch lsinventory
Oracle Interim Patch Installer version 11.2.0.3.0
Copyright (c) 2012, Oracle Corporation. All rights reserved.

Oracle Home : /opt/app/oracle/product/11gR2
Central Inventory : /opt/app/oracle/oraInventory
from : /opt/app/oracle/product/11gR2/oraInst.loc
OPatch version : 11.2.0.3.0
OUI version : 11.2.0.3.0
Log file location : /opt/app/oracle/product/11gR2/cfgtoollogs/opatch/opatch2015-03-01_05-35-53AM_1.log

Lsinventory Output file location : /opt/app/oracle/product/11gR2/cfgtoollogs/opatch/lsinv/lsinventory2015-03-01_05-35-53AM.txt

——————————————————————————–
Installed Top-level Products (1):

Oracle Database 11g 11.2.0.3.0
There are 1 products installed in this Oracle Home.

Interim patches (5) :

Patch (online) 20505138: applied on Sun Mar 01 05:35:42 CET 2015
Unique Patch ID: 18568953
Created on 9 Feb 2015, 19:34:29 hrs PST8PDT
Bugs fixed:
16367081, 16463153, 16342845


‘Hope That Helps’
Thanks
Prashant Dixit

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

w

Connecting to %s

 
%d bloggers like this: