Oracle RAC haldamine

Sissejuhatus

TODO

Klastri omadused

TODO

Võrk

RAC kasutab nn VIP aadresse, mille poole kliendid pöörduvad. VIP aadressid on seadistatud nn eth aliastena, eth0:0 jne. Klastrihalduse tarkvara tegeleb nende arvutite vahel liigutamisega.

Salvestusseadmed

# /etc/init.d/oracleasm listdisks
DATA1
FRA1
GRID
RLOG1
RLOG2

# multipath -ll | grep HITACHI
backup1 (360060e80102d1dc00511c564000003f1) dm-16 HITACHI,DF600F
360060e80102d1dc00511c56400000010 dm-1 HITACHI,DF600F
grid (360060e80102d1dc00511c564000003ec) dm-2 HITACHI,DF600F
fra1 (360060e80102d1dc00511c564000003f0) dm-14 HITACHI,DF600F
360060e80102d1dc00511c56400000002 dm-0 HITACHI,DF600F
rlog2 (360060e80102d1dc00511c564000003ee) dm-4 HITACHI,DF600F
rlog1 (360060e80102d1dc00511c564000003ed) dm-3 HITACHI,DF600F
data1 (360060e80102d1dc00511c564000003ef) dm-12 HITACHI,DF600

Protsessid

• grid kasutajana töötavad klastri protsessid
• oracle kasutajana töötavad andmebaasi protsessid
• Klastri õla peatamine, seejuures võtavad allesjäänud node'id vip aadressid üle

# /u01/app/11.2.0/grid/bin/crsctl stop cluster

• Klastri ressursside nimekiri

grid$ . system/env.sh
grid$ crsctl status resource -t

• RAC node'il instance'i seiskamiseks

oracle$ . system/env.sh
oracle$ srvctl stop instance -d dba -i dba2

• RAC node'il andmebaasi seiskamiseks

oracle$ srvctl stop database -d dba

• Käivitamiseks

oracle$ srvctl start database -d dba

• Teenuse peatamine node'il

oracle$ srvctl stop service -d dba -s loomad.test.loomaaed -n oradb2

• Kogu klastri peatamine, esimene korraldus kõigil klastri komponentidel, teine ühel suvalisel ja viimane kõigil

oracle$ emctl stop dbconsole
# /u01/app/11.2.0/grid/bin/crsctl stop cluster -all
# shutdown -h now

Logi

• Klastri peamine logifail on

/u01/app/11.2.0/grid/log/oradb2a/alertoradb2a.log

Kasutamine

• Instant Clientiga ja ilma tnsnames.ora seadistusfailita

$ LD_LIBRARY_PATH=. ./sqlplus kasutajanimi/parool@"(DESCRIPTION=(ADDRESS=(PROTOCOL=TCP)(HOST = 10.100.7.66)(PORT = 1521))(CONNECT_DATA=(SID = dba1)))"
$ LD_LIBRARY_PATH=. ./sqlplus kasutajanimi/parool@"(DESCRIPTION=(ADDRESS=(PROTOCOL=TCP)(HOST = 10.100.7.67)(PORT = 1521))(CONNECT_DATA=(SID = dba2)))"
$ LD_LIBRARY_PATH=. ./sqlplus kasutajanimi/parool@"(DESCRIPTION=(ADDRESS=(PROTOCOL=TCP)(HOST = 10.100.7.68)(PORT = 1521))(CONNECT_DATA=(SID = dba3)))"

Varundamine

Varundamise ajal peavad

• rakendused seisma
• andmebaas ja klaster töötama

oracle kasutajana tuleb viimasel klastri komponendil öelda

oracle$ . ./system/env.sh
oracle$ rman target / nocatalog

Recovery Manager: Release 11.2.0.1.0 - Production on Thu Sep 23 16:04:32 2010
 
Copyright (c) 1982, 2009, Oracle and/or its affiliates.  All rights reserved.
 
connected to target database: DBA (DBID=1175559531)
using target database control file instead of recovery catalog
 
RMAN> backup database; 
 
Starting backup at 23-SEP-10
allocated channel: ORA_DISK_1
channel ORA_DISK_1: SID=22 instance=dba2 device type=DISK
channel ORA_DISK_1: starting compressed full datafile backup set
channel ORA_DISK_1: specifying datafile(s) in backup set
input datafile file number=00022 name=+DATA/dba/datafile/webmdata.281.720112973
...
input datafile file number=00034 name=+DATA/dba/datafile/it_indx.290.726595429
channel ORA_DISK_1: starting piece 1 at 23-SEP-10
channel ORA_DISK_1: finished piece 1 at 23-SEP-10
piece handle=/u02/rman/mmlokipe_1_1 tag=TAG20100923T160402 comment=NONE
channel ORA_DISK_1: backup set complete, elapsed time: 00:43:56
Finished backup at 23-SEP-10

Starting Control File and SPFILE Autobackup at 23-SEP-10
piece handle=/u02/autobackup/c-1175559231-20100923-02 comment=NONE
Finished Control File and SPFILE Autobackup at 23-SEP-10

RMAN>

Tulemus tekib kataloogi /u02/rman.

Kasulikud lisamaterjalid

• http://kuutorvaja.eenet.ee/wiki/Oracle_andmebaas