Efficiency by design when creating backup strategies

by Michael Dinh on Sep 21, 2018 12:00:00 AM

Most backup implementations to not take log management into consideration. Example: Have you seen thousands of backup logs or backup log dating back to 2017? Is it necessary? Worse, paged in the middle of the night for disk filling up. I will share backup implementation and log management for database emrep.

Recovery Point Objective (RPO) is RECOVERY WINDOW OF 1 DAYS.
 RMAN catalog is being used.

Spool error to a different file for crontab.

 $ crontab -l # Backup database daily at 1am 0 1 * * * /u01/scripts/rman/rman_db.ksh emrep 0 100 > /tmp/emrep_rman_db.out 2>/tmp/emrep_rman_db.err # Backup archivelog every 4 hour 0 */4 * * * /u01/scripts/rman/rman_arch.ksh env_emrep > /tmp/emrep_rman_arch.out 2>/tmp/emrep_rman_arch.err

Zero byte from .err file indicates there are no errors.

 $ ll /tmp/*rman* -rw-r--r-- 1 oracle oinstall 0 Sep 17 09:39 /tmp/emrep_rman_arch.err -rw-r--r-- 1 oracle oinstall 897 Sep 17 09:39 /tmp/emrep_rman_arch.out -rw-r--r-- 1 oracle oinstall 0 Sep 17 01:00 /tmp/emrep_rman_db.err -rw-r--r-- 1 oracle oinstall 640 Sep 17 01:00 /tmp/emrep_rman_db.out

Manage log by overwriting.

Log indicates exact script ran prefixed with ORACLE_SID (passed to script)

 $ ls -lrt /u01/app/oracle/admin/emrep/scripts/log total 292 -rw-r--r-- 1 oracle oinstall 16039 Sep 12 12:44 emrep.rman_db.ksh.log.Wed -rw-r--r-- 1 oracle oinstall 9248 Sep 12 13:18 emrep.rman_arch.ksh.log.Wed13 -rw-r--r-- 1 oracle oinstall 13609 Sep 14 10:41 emrep.rman_arch.ksh.log.Fri10 -rw-r--r-- 1 oracle oinstall 8033 Sep 14 16:00 emrep.rman_arch.ksh.log.Fri16 -rw-r--r-- 1 oracle oinstall 7891 Sep 14 20:00 emrep.rman_arch.ksh.log.Fri20 -rw-r--r-- 1 oracle oinstall 8321 Sep 15 00:00 emrep.rman_arch.ksh.log.Sat00 -rw-r--r-- 1 oracle oinstall 11280 Sep 15 01:01 emrep.rman_db.ksh.log.Sat -rw-r--r-- 1 oracle oinstall 10525 Sep 15 04:00 emrep.rman_arch.ksh.log.Sat04 -rw-r--r-- 1 oracle oinstall 10738 Sep 15 08:00 emrep.rman_arch.ksh.log.Sat08 -rw-r--r-- 1 oracle oinstall 11217 Sep 15 12:00 emrep.rman_arch.ksh.log.Sat12 -rw-r--r-- 1 oracle oinstall 11437 Sep 15 16:00 emrep.rman_arch.ksh.log.Sat16 -rw-r--r-- 1 oracle oinstall 11656 Sep 15 20:00 emrep.rman_arch.ksh.log.Sat20 -rw-r--r-- 1 oracle oinstall 11813 Sep 16 00:00 emrep.rman_arch.ksh.log.Sun00 -rw-r--r-- 1 oracle oinstall 19085 Sep 16 01:01 emrep.rman_db.ksh.log.Sun -rw-r--r-- 1 oracle oinstall 11284 Sep 16 04:00 emrep.rman_arch.ksh.log.Sun04 -rw-r--r-- 1 oracle oinstall 11285 Sep 16 08:00 emrep.rman_arch.ksh.log.Sun08 -rw-r--r-- 1 oracle oinstall 10937 Sep 16 12:00 emrep.rman_arch.ksh.log.Sun12 -rw-r--r-- 1 oracle oinstall 11224 Sep 16 16:00 emrep.rman_arch.ksh.log.Sun16 -rw-r--r-- 1 oracle oinstall 11440 Sep 16 20:00 emrep.rman_arch.ksh.log.Sun20 -rw-r--r-- 1 oracle oinstall 11597 Sep 17 00:00 emrep.rman_arch.ksh.log.Mon00 -rw-r--r-- 1 oracle oinstall 18469 Sep 17 01:01 emrep.rman_db.ksh.log.Mon -rw-r--r-- 1 oracle oinstall 11284 Sep 17 04:00 emrep.rman_arch.ksh.log.Mon04 -rw-r--r-- 1 oracle oinstall 11289 Sep 17 08:00 emrep.rman_arch.ksh.log.Mon08

Based on RPO:

Keep 7 logs for daily database backup. Keep 42 logs for achivelog backup (6 daily x 7 days)

Echo RMAN Backup Log to output as this will make troubleshooting more efficient.

Have you ever had to dig through a backup script to file logfile?

 $ cat /tmp/emrep_rman_arch.out The Oracle base remains unchanged with value /u01/app/oracle LD_LIBRARY_PATH=/u01/app/oracle/product/12.1.0.2/dbhome_1/lib:/lib:/usr/lib NLS_DATE_FORMAT=DD-MON-YYYY HH24:MI:SS ORACLE_BASE=/u01/app/oracle ORACLE_HOME=/u01/app/oracle/product/12.1.0.2/dbhome_1 ORACLE_SID=emrep PATH=/usr/lib64/qt-3.3/bin:/usr/local/bin:/bin:/usr/bin:/usr/local/sbin:/usr/sbin:/sbin:/home/oracle/bin:/usr/local/bin:/home/oracle/bin:/usr/local/bin:/u01/app/oracle/product/12.1.0.2/dbhome_1/bin Oracle Instance alive for sid "emrep"  ============================== RMAN Backup Log /u01/app/oracle/admin/emrep/scripts/log/emrep.rman_arch.ksh.log.Mon09 ==============================  RMAN> 2> 3> 4> 5> 6> 7> 8> 9> 10> 11> 12> 13> 14> 15> 16> 17> 18> 19> 20> 21> 22> 23> 24> 25> 26> 27> 28> 29> 30> 31> 32> 33> 34>

In conclusion, when designing backup strategies, be sure to take into consideration log usability, scalability and efficiency.

Topics: Technical Track Cassandra Cqlsstablewriter Sstableloader Sstablesimpleunsortedwriter Thrift Data

No Comments Yet

Let us know what you think

See all

Share this