Data Guard Failover after primary site network failure or disconnect.

Hello Experts:
I'll try to be clear and specific with my issue:
Environment:
Two nodes with NO shared storage (I don't have an Observer running).
Veritas Cluser Server (VCS) with Data Guar Agent. (I don't use the Broker. Data Guard agent "takes care" of the switchover and failover).
Two single instance databases, one per node. NO RAC.
What I'm being able to perform with no issues:
Manual switch(over) of the primary database by running VCS command "hagrp -switch oraDG_group -to standby_node"
Automatic fail(over) when primary node is rebooted with "reboot" or "init"
Automatic fail(over) when primary node is shut down with "shutdown".
What I'm NOT being able to perform:
If I manually unplug the network cables from the primary site (all the network, not only the link between primary and standby node so, it's like a server unplug from the energy source).
Same situation happens if I manually disconnect the server from the power.
This is the alert logs I have:
This is the portion of the alert log at Standby site when Real Time Replication is working fine:
Recovery of Online Redo Log: Thread 1 Group 4 Seq 7 Reading mem 0
Mem# 0: /u02/oracle/fast_recovery_area/standby_db/onlinelog/o1_mf_4_9c3tk3dy_.log
At this moment, node1 (Primary) is completely disconnected from the network. SEE at the end when the database (standby which should be converted to PRIMARY) is not getting all the archived logs from the Primary due to the abnormal disconnect from the network:
Identified End-Of-Redo (failover) for thread 1 sequence 7 at SCN 0xffff.ffffffff
Incomplete Recovery applied until change 15922544 time 12/23/2013 17:12:48
Media Recovery Complete (primary_db)
Terminal Recovery: successful completion
Forcing ARSCN to IRSCN for TR 0:15922544
Mon Dec 23 17:13:22 2013
ARCH: Archival stopped, error occurred. Will continue retrying
ORACLE Instance primary_db - Archival ErrorAttempt to set limbo arscn 0:15922544 irscn 0:15922544
ORA-16014: log 4 sequence# 7 not archived, no available destinations
ORA-00312: online log 4 thread 1: '/u02/oracle/fast_recovery_area/standby_db/onlinelog/o1_mf_4_9c3tk3dy_.log'
Resetting standby activation ID 2071848820 (0x7b7de774)
Completed: ALTER DATABASE RECOVER MANAGED STANDBY DATABASE FINISH
Mon Dec 23 17:13:33 2013
ALTER DATABASE RECOVER MANAGED STANDBY DATABASE FINISH
Terminal Recovery: applying standby redo logs.
Terminal Recovery: thread 1 seq# 7 redo required
Terminal Recovery:
Recovery of Online Redo Log: Thread 1 Group 4 Seq 7 Reading mem 0
Mem# 0: /u02/oracle/fast_recovery_area/standby_db/onlinelog/o1_mf_4_9c3tk3dy_.log
Identified End-Of-Redo (failover) for thread 1 sequence 7 at SCN 0xffff.ffffffff
Incomplete Recovery applied until change 15922544 time 12/23/2013 17:12:48
Media Recovery Complete (primary_db)
Terminal Recovery: successful completion
Forcing ARSCN to IRSCN for TR 0:15922544
Mon Dec 23 17:13:22 2013
ARCH: Archival stopped, error occurred. Will continue retrying
ORACLE Instance primary_db - Archival ErrorAttempt to set limbo arscn 0:15922544 irscn 0:15922544
ORA-16014: log 4 sequence# 7 not archived, no available destinations
ORA-00312: online log 4 thread 1: '/u02/oracle/fast_recovery_area/standby_db/onlinelog/o1_mf_4_9c3tk3dy_.log'
Resetting standby activation ID 2071848820 (0x7b7de774)
Completed: ALTER DATABASE RECOVER MANAGED STANDBY DATABASE FINISH
Mon Dec 23 17:13:33 2013
ALTER DATABASE RECOVER MANAGED STANDBY DATABASE FINISH
Attempt to do a Terminal Recovery (primary_db)
Media Recovery Start: Managed Standby Recovery (primary_db)
started logmerger process
Mon Dec 23 17:13:33 2013
Managed Standby Recovery not using Real Time Apply
Media Recovery failed with error 16157
Recovery Slave PR00 previously exited with exception 283
ORA-283 signalled during: ALTER DATABASE RECOVER MANAGED STANDBY DATABASE FINISH...
Mon Dec 23 17:13:34 2013
Shutting down instance (immediate)
Shutting down instance: further logons disabled
Stopping background process MMNL
Stopping background process MMON
License high water mark = 38
All dispatchers and shared servers shutdown
ALTER DATABASE CLOSE NORMAL
ORA-1109 signalled during: ALTER DATABASE CLOSE NORMAL...
ALTER DATABASE DISMOUNT
Shutting down archive processes
Archiving is disabled
Mon Dec 23 17:13:38 2013
Mon Dec 23 17:13:38 2013
Mon Dec 23 17:13:38 2013
ARCH shutting downARCH shutting down
ARCH shutting down
ARC0: Relinquishing active heartbeat ARCH role
ARC2: Archival stopped
ARC0: Archival stopped
ARC1: Archival stopped
Completed: ALTER DATABASE DISMOUNT
ARCH: Archival disabled due to shutdown: 1089
Shutting down archive processes
Archiving is disabled
Mon Dec 23 17:13:40 2013
Stopping background process VKTM
ARCH: Archival disabled due to shutdown: 1089
Shutting down archive processes
Archiving is disabled
Mon Dec 23 17:13:43 2013
Instance shutdown complete
Mon Dec 23 17:13:44 2013
Adjusting the default value of parameter parallel_max_servers
from 1280 to 470 due to the value of parameter processes (500)
Starting ORACLE instance (normal)
************************ Large Pages Information *******************
Per process system memlock (soft) limit = 64 KB
Total Shared Global Region in Large Pages = 0 KB (0%)
Large Pages used by this instance: 0 (0 KB)
Large Pages unused system wide = 0 (0 KB)
Large Pages configured system wide = 0 (0 KB)
Large Page size = 2048 KB
RECOMMENDATION:
Total System Global Area size is 3762 MB. For optimal performance,
prior to the next instance restart:
1. Increase the number of unused large pages by
at least 1881 (page size 2048 KB, total size 3762 MB) system wide to
get 100% of the System Global Area allocated with large pages
2. Large pages are automatically locked into physical memory.
Increase the per process memlock (soft) limit to at least 3770 MB to lock
100% System Global Area's large pages into physical memory
LICENSE_MAX_SESSION = 0
LICENSE_SESSIONS_WARNING = 0
Initial number of CPU is 32
Number of processor cores in the system is 16
Number of processor sockets in the system is 2
CELL communication is configured to use 0 interface(s):
CELL IP affinity details:
    NUMA status: NUMA system w/ 2 process groups
    cellaffinity.ora status: cannot find affinity map at '/etc/oracle/cell/network-config/cellaffinity.ora' (see trace file for details)
CELL communication will use 1 IP group(s):
    Grp 0:
Picked latch-free SCN scheme 3
Autotune of undo retention is turned on.
IMODE=BR
ILAT =88
LICENSE_MAX_USERS = 0
SYS auditing is disabled
NUMA system with 2 nodes detected
Starting up:
Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production
With the Partitioning, OLAP, Data Mining and Real Application Testing options.
ORACLE_HOME = /u01/oracle/product/11.2.0.4
System name:    Linux
Node name:      node2.localdomain
Release:        2.6.32-131.0.15.el6.x86_64
Version:        #1 SMP Tue May 10 15:42:40 EDT 2011
Machine:        x86_64
Using parameter settings in server-side spfile /u01/oracle/product/11.2.0.4/dbs/spfileprimary_db.ora
System parameters with non-default values:
processes                = 500
sga_target               = 3760M
control_files            = "/u02/oracle/orafiles/primary_db/control01.ctl"
control_files            = "/u01/oracle/fast_recovery_area/primary_db/control02.ctl"
db_file_name_convert     = "standby_db"
db_file_name_convert     = "primary_db"
log_file_name_convert    = "standby_db"
log_file_name_convert    = "primary_db"
control_file_record_keep_time= 40
db_block_size            = 8192
compatible               = "11.2.0.4.0"
log_archive_dest_1       = "location=/u02/oracle/archivelogs/primary_db"
log_archive_dest_2       = "SERVICE=primary_db ASYNC VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=primary_db"
log_archive_dest_state_2 = "ENABLE"
log_archive_min_succeed_dest= 1
fal_server               = "primary_db"
log_archive_trace        = 0
log_archive_config       = "DG_CONFIG=(primary_db,standby_db)"
log_archive_format       = "%t_%s_%r.dbf"
log_archive_max_processes= 3
db_recovery_file_dest    = "/u02/oracle/fast_recovery_area"
db_recovery_file_dest_size= 30G
standby_file_management = "AUTO"
db_flashback_retention_target= 1440
undo_tablespace          = "UNDOTBS1"
remote_login_passwordfile= "EXCLUSIVE"
db_domain                = ""
dispatchers              = "(PROTOCOL=TCP) (SERVICE=primary_dbXDB)"
job_queue_processes      = 0
audit_file_dest          = "/u01/oracle/admin/primary_db/adump"
audit_trail              = "DB"
db_name                  = "primary_db"
db_unique_name           = "standby_db"
open_cursors             = 300
pga_aggregate_target     = 1250M
dg_broker_start          = FALSE
diagnostic_dest          = "/u01/oracle"
Mon Dec 23 17:13:45 2013
PMON started with pid=2, OS id=29108
Mon Dec 23 17:13:45 2013
PSP0 started with pid=3, OS id=29110
Mon Dec 23 17:13:46 2013
VKTM started with pid=4, OS id=29125 at elevated priority
VKTM running at (1)millisec precision with DBRM quantum (100)ms
Mon Dec 23 17:13:46 2013
GEN0 started with pid=5, OS id=29129
Mon Dec 23 17:13:46 2013
DIAG started with pid=6, OS id=29131
Mon Dec 23 17:13:46 2013
DBRM started with pid=7, OS id=29133
Mon Dec 23 17:13:46 2013
DIA0 started with pid=8, OS id=29135
Mon Dec 23 17:13:46 2013
MMAN started with pid=9, OS id=29137
Mon Dec 23 17:13:46 2013
DBW0 started with pid=10, OS id=29139
Mon Dec 23 17:13:46 2013
DBW1 started with pid=11, OS id=29141
Mon Dec 23 17:13:46 2013
DBW2 started with pid=12, OS id=29143
Mon Dec 23 17:13:46 2013
DBW3 started with pid=13, OS id=29145
Mon Dec 23 17:13:46 2013
LGWR started with pid=14, OS id=29147
Mon Dec 23 17:13:46 2013
CKPT started with pid=15, OS id=29149
Mon Dec 23 17:13:46 2013
SMON started with pid=16, OS id=29151
Mon Dec 23 17:13:46 2013
RECO started with pid=17, OS id=29153
Mon Dec 23 17:13:46 2013
MMON started with pid=18, OS id=29155
Mon Dec 23 17:13:46 2013
MMNL started with pid=19, OS id=29157
starting up 1 dispatcher(s) for network address '(ADDRESS=(PARTIAL=YES)(PROTOCOL=TCP))'...
starting up 1 shared server(s) ...
ORACLE_BASE from environment = /u01/oracle
Mon Dec 23 17:13:46 2013
ALTER DATABASE   MOUNT
ARCH: STARTING ARCH PROCESSES
Mon Dec 23 17:13:50 2013
ARC0 started with pid=23, OS id=29210
ARC0: Archival started
ARCH: STARTING ARCH PROCESSES COMPLETE
ARC0: STARTING ARCH PROCESSES
Successful mount of redo thread 1, with mount id 2071851082
Mon Dec 23 17:13:51 2013
ARC1 started with pid=24, OS id=29212
Allocated 15937344 bytes in shared pool for flashback generation buffer
Mon Dec 23 17:13:51 2013
ARC2 started with pid=25, OS id=29214
Starting background process RVWR
ARC1: Archival started
ARC1: Becoming the 'no FAL' ARCH
ARC1: Becoming the 'no SRL' ARCH
Mon Dec 23 17:13:51 2013
RVWR started with pid=26, OS id=29216
Physical Standby Database mounted.
Lost write protection disabled
Completed: ALTER DATABASE   MOUNT
Mon Dec 23 17:13:51 2013
ALTER DATABASE RECOVER MANAGED STANDBY DATABASE
         USING CURRENT LOGFILE DISCONNECT FROM SESSION
Attempt to start background Managed Standby Recovery process (primary_db)
Mon Dec 23 17:13:51 2013
MRP0 started with pid=27, OS id=29219
MRP0: Background Managed Standby Recovery process started (primary_db)
ARC2: Archival started
ARC0: STARTING ARCH PROCESSES COMPLETE
ARC2: Becoming the heartbeat ARCH
ARC2: Becoming the active heartbeat ARCH
ARCH: Archival stopped, error occurred. Will continue retrying
ORACLE Instance primary_db - Archival Error
ORA-16014: log 4 sequence# 7 not archived, no available destinations
ORA-00312: online log 4 thread 1: '/u02/oracle/fast_recovery_area/standby_db/onlinelog/o1_mf_4_9c3tk3dy_.log'
At this moment, I've lost service and I have to wait until the prmiary server goes up again to receive the missing log.
This is the rest of the log:
Fatal NI connect error 12543, connecting to:
(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=TCP)(HOST=node1)(PORT=1521)))(CONNECT_DATA=(SERVICE_NAME=primary_db)(CID=(PROGRAM=oracle)(HOST=node2.localdomain)(USER=oracle))))
VERSION INFORMATION:
        TNS for Linux: Version 11.2.0.4.0 - Production
        TCP/IP NT Protocol Adapter for Linux: Version 11.2.0.4.0 - Production
Time: 23-DEC-2013 17:13:52
Tracing not turned on.
Tns error struct:
    ns main err code: 12543
TNS-12543: TNS:destination host unreachable
    ns secondary err code: 12560
    nt main err code: 513
TNS-00513: Destination host unreachable
    nt secondary err code: 113
    nt OS err code: 0
Fatal NI connect error 12543, connecting to:
(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=TCP)(HOST=node1)(PORT=1521)))(CONNECT_DATA=(SERVICE_NAME=primary_db)(CID=(PROGRAM=oracle)(HOST=node2.localdomain)(USER=oracle))))
VERSION INFORMATION:
        TNS for Linux: Version 11.2.0.4.0 - Production
        TCP/IP NT Protocol Adapter for Linux: Version 11.2.0.4.0 - Production
Time: 23-DEC-2013 17:13:55
Tracing not turned on.
Tns error struct:
    ns main err code: 12543
TNS-12543: TNS:destination host unreachable
    ns secondary err code: 12560
    nt main err code: 513
TNS-00513: Destination host unreachable
    nt secondary err code: 113
    nt OS err code: 0
started logmerger process
Mon Dec 23 17:13:56 2013
Managed Standby Recovery starting Real Time Apply
MRP0: Background Media Recovery terminated with error 16157
Errors in file /u01/oracle/diag/rdbms/standby_db/primary_db/trace/primary_db_pr00_29230.trc:
ORA-16157: media recovery not allowed following successful FINISH recovery
Managed Standby Recovery not using Real Time Apply
Completed: ALTER DATABASE RECOVER MANAGED STANDBY DATABASE
         USING CURRENT LOGFILE DISCONNECT FROM SESSION
Recovery Slave PR00 previously exited with exception 16157
MRP0: Background Media Recovery process shutdown (primary_db)
Fatal NI connect error 12543, connecting to:
(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=TCP)(HOST=node1)(PORT=1521)))(CONNECT_DATA=(SERVICE_NAME=primary_db)(CID=(PROGRAM=oracle)(HOST=node2.localdomain)(USER=oracle))))
VERSION INFORMATION:
        TNS for Linux: Version 11.2.0.4.0 - Production
        TCP/IP NT Protocol Adapter for Linux: Version 11.2.0.4.0 - Production
Time: 23-DEC-2013 17:13:58
Tracing not turned on.
Tns error struct:
    ns main err code: 12543
TNS-12543: TNS:destination host unreachable
    ns secondary err code: 12560
    nt main err code: 513
TNS-00513: Destination host unreachable
    nt secondary err code: 113
    nt OS err code: 0
Mon Dec 23 17:14:01 2013
Fatal NI connect error 12543, connecting to:
(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=TCP)(HOST=node1)(PORT=1521)))(CONNECT_DATA=(SERVICE_NAME=primary_db)(CID=(PROGRAM=oracle)(HOST=node2.localdomain)(USER=oracle))))
VERSION INFORMATION:
        TNS for Linux: Version 11.2.0.4.0 - Production
        TCP/IP NT Protocol Adapter for Linux: Version 11.2.0.4.0 - Production
Time: 23-DEC-2013 17:14:01
Tracing not turned on.
Tns error struct:
    ns main err code: 12543
TNS-12543: TNS:destination host unreachable
    ns secondary err code: 12560
    nt main err code: 513
TNS-00513: Destination host unreachable
    nt secondary err code: 113
    nt OS err code: 0
Error 12543 received logging on to the standby
FAL[client, ARC0]: Error 12543 connecting to primary_db for fetching gap sequence
Archiver process freed from errors. No longer stopped
Mon Dec 23 17:15:07 2013
Using STANDBY_ARCHIVE_DEST parameter default value as /u02/oracle/archivelogs/primary_db
Mon Dec 23 17:19:51 2013
ARCH: Archival stopped, error occurred. Will continue retrying
ORACLE Instance primary_db - Archival Error
ORA-16014: log 4 sequence# 7 not archived, no available destinations
ORA-00312: online log 4 thread 1: '/u02/oracle/fast_recovery_area/standby_db/onlinelog/o1_mf_4_9c3tk3dy_.log'
Mon Dec 23 17:26:18 2013
RFS[1]: Assigned to RFS process 31456
RFS[1]: No connections allowed during/after terminal recovery.
Mon Dec 23 17:26:47 2013
flashback database to scn 15921680
ORA-16157 signalled during: flashback database to scn 15921680...
Mon Dec 23 17:27:05 2013
alter database recover managed standby database using current logfile disconnect
Attempt to start background Managed Standby Recovery process (primary_db)
Mon Dec 23 17:27:05 2013
MRP0 started with pid=28, OS id=31481
MRP0: Background Managed Standby Recovery process started (primary_db)
started logmerger process
Mon Dec 23 17:27:10 2013
Managed Standby Recovery starting Real Time Apply
MRP0: Background Media Recovery terminated with error 16157
Errors in file /u01/oracle/diag/rdbms/standby_db/primary_db/trace/primary_db_pr00_31486.trc:
ORA-16157: media recovery not allowed following successful FINISH recovery
Managed Standby Recovery not using Real Time Apply
Completed: alter database recover managed standby database using current logfile disconnect
Recovery Slave PR00 previously exited with exception 16157
MRP0: Background Media Recovery process shutdown (primary_db)
Mon Dec 23 17:27:18 2013
RFS[2]: Assigned to RFS process 31492
RFS[2]: No connections allowed during/after terminal recovery.
Mon Dec 23 17:28:18 2013
RFS[3]: Assigned to RFS process 31614
RFS[3]: No connections allowed during/after terminal recovery.
Do you have any advice?
Thanks!
Alex.

Hello;
What's not clear to me in your question at this point:
What I'm NOT being able to perform:
If I manually unplug the network cables from the primary site (all the network, not only the link between primary and standby node so, it's like a server unplug from the energy source).
Same situation happens if I manually disconnect the server from the power.
This is the alert logs I have:"
Are you trying a failover to the Standby?
Please advise.
Is it possible your "valid_for clause" is set incorrectly?
Would also review this:
ORA-16014 and ORA-00312 Messages in Alert.log of Physical Standby
Best Regards
mseberg

Similar Messages

Oracle data guard configuration for primary and standby db_name

I am working on configuring an active data guard for one primary DB and one standby DB. I have a few questions:
1. Can I use different db_name, db_unique_name and instance_name for primary and standby. For example: primary(db_name, db_unique_name and instance_name)=chicago. When I create standby DB with Rman backup and copy of pfile and control file from primary DB or use Grid control to create standby database. Oracle document or Grid control all keep standby db_name=chicago. Only make standby db_unique_name and instance_name=boston. Due to my application system condition, I want to make db_name=boston, not keep it as the same as primary=chicago. Is this valid configuration?
2. In primary datafiles, application system generate datafile name like this: hr_chicago_01.dbf, fn_chicago_01.dbf. When I move datafiles to standby server, if I plan to use db_name=boston for standby DB, can I rename datafiles as
hr_boston_01.dbf, fn_boston_01.dbf? In this way, datafile name match up with db_name. but I will create standby log group and members on primary and standby identically. If in future switching over, DB will not have problems.
3. If I don't use primary DB backup. Instead, I copy all datafiles, redo_log files (no control files) to standby. Then "alter database backup controlfile to trace" from promary and also " create pfile='/xxx/initSTANDBY.ora' from primary. Then modify init.ora and controlfile. Then run control.sql to bring standby DB up. After that, configure redo log shipping and apply with data guard or SQL. Is this a acceptable way to create physical standby DB?
Please advise your comments. Thanks in advance.

I want to make db_name=boston, not keep it as the same as primary=chicago. Is this valid configuration?NO. DB_NAME must be the same ("chicago") at both sites. The Standby will be using a different DB_UNIQUE_NAME (e.g. "boston") and can be using a different Instance name / SID (e.g. "boston").
can I rename datafiles Yes. The database file names can be changed.
If I don't use primary DB backup. Instead, I copy all datafiles, redo_log files (no control files) to standbyWhat is the difference between the first sentence (a backup of the primary) and the second sentence (a copy of the primary) ? A Copy is a backup.
Are you intending to differentiate between an RMAN Backup and a User-Managed (aka "scripted") backup ?
Normally, for DataGuard, tou can use non-RMAN methods to copy the database but there's no value add in this.
You'd still have to setup DataGuard ! (and I wonder if you'd have complications setting up Active DataGuard).
But remember that you MUST create the Standby controlfile from the Primary and copy it over to your Standby -- particularly as you are planning to use DataGuard. This is not created by 'alter database backup controlfile to trace' , but by 'ALTER DATABASE CREATE STANDBY CONTROLFILE AS 'filename''
Hemant K Chitale

Data guard synchronization after link down b/w primary and physical standby

Hi All,
I have configured data guard on oracle 11gr2 db. Normally switchover between my primary and physical standby happens smoothly and the Apply lag would be zero. Recently We had to test a scenario when the network link between Primary and Physical standby is completely down and Physical standby is isolated completely for more than half an hour.
When we brought up the link every thing worked smoothly but apply lag started increasing from 0 to around 3 hrs. And then it started reducing to 0. Currently Apply lag and transport lag shows 0.
But is this normal behaviour of oracle data guard that when the link between primary and physical standby is completely down, It requires 3-4 hrs for resynchronization ??? Even when during isolation, there were very few transactions happend on primary database ??
Are there any documents available for this scenario??
Thanks

Hi, after the link is up, if there were some transactions and produced archive logs it's normal to take some time for resync. To check if 3-4 hours is normal or not, you can repeat the scenario and this time check
- how many archivelogs does primary produce in this period.
- after the link is up, does archivelog transfer immediately starts from primary to standby? Is primary able to send these archivelogs parallel?
- Is there anything wrong with the apply process?
check primary & standby alert log files, and run this query on standby to check the transport and apply processes:
SELECT PROCESS, STATUS, THREAD#, SEQUENCE#, BLOCK#, BLOCKS FROM V$MANAGED_STANDBY;
regards

Data guard difference in primary and standby OS.

Dear All,
We are planning to setup a data guard in our landscape. But my question what are the prerequisites should be consider before getting in to that.
Like my environment -
1. The primary database is running with Red hat Linux 5 - IBM 64 bit and the standby database is Redhat Linux 2.1 IBM 32 bit system.
2. The primary database is running with Red hat Linux 5 - IBM 64 bit with oracle 8.1.7.4 and the standby database is
Linux 2.1 IBM 32 bit system with oracle 10g.
Is the above thing is possible? or we should maintain the same OS / DB combination in the data guard setup.
I am following the below link.
http://download-west.oracle.com/docs/cd/B19306_01/server.102/b14239/standby.htm#i72053
If anyone have thoughts on this please do advice me.
Regards
Vijay

Please explain to us what exactly you want to do with the standby database?
According to note 105047 - Support for Oracle functions in the SAP environment
12. Data Guard
You can use "Physical Standby", but Oracle provides support (not SAP).
You cannot use "Logical Standby"
So in short, no a data guard configuration with Red hat Linux 5 - IBM 64 bit with oracle 8.1.7.4 to Linux 2.1 IBM 32 bit system with oracle 10g is NOT possible.
Regards, Michael

Steps for Data Guard with one primary and 2 standby

Hi,
Database :10.2.0.4, 11.2.0.1
Os: Windows , Unix
A ----------------> Primary database
B ----------------> Standby Database 1
C ----------------> Standby Database 2
I want to configure *2 standby* databases for single primary database.
Lets take, A ,B and C are my machines.My data guard configuration will be like,*archive logs will be moving* from A to B and A to C.
If i do any switchover in between A and B , now B is primary and remaining A and C are standby databases.At this stage also , archive logs should move from B to A and B to C. Also, same should happen from C to A and C to B,If i do switchover in between B and C.If everything is fine , then i will do switchback to main Primary database(A).
How do i have to mention PFILE in all machines ,the parameters like
LOG_ARCHIVE_DEST_1=LOCATION=<PATH> -- LOCAL ARCHIVE PATH
LOG_ARCHIVE_DEST_2=SERVICE=
LOG_ARCHIVE_DEST_3=SERVICE=
FAL_SERVER=
FAL_CLIENT=
STANDBY_FILE_MANAGEMENT=
In my tnsnames.ora , primary,standby1 and standby2 are my service entries and these are same in all of my machines.
Please suggest me , how do i can configure my pfiles in all machines ?.
Thanks,
Sunand

Not yet, but now you have me interested.
Please consider Flashback.
I still have to test but here's my take:
PRIMARY SETTINGS
*.FAL_SERVER=STANDBY
*.FAL_CLIENT=PRIMARY
*.STANDBY_FILE_MANAGEMENT=AUTO
*.DB_UNIQUE_NAME=PRIMARY
*.LOG_FILE_NAME_CONVERT='STANDBY','PRIMARY'
*.log_archive_dest_1='LOCATION=USE_DB_RECOVERY_FILE_DEST VALID_FOR=(ALL_LOGFILES,ALL_ROLES) DB_UNIQUE_NAME=PRIMARY'
*.log_archive_dest_2='SERVICE=STANDBY LGWR ASYNC VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=STANDBY'
*.log_archive_dest_3='SERVICE=STANDBY2 LGWR ASYNC VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=STANDBY2'
*.LOG_ARCHIVE_DEST_STATE_1=ENABLE
*.LOG_ARCHIVE_DEST_STATE_2=ENABLE
*.LOG_ARCHIVE_DEST_STATE_3=ENABLE
*.LOG_ARCHIVE_MAX_PROCESSES=30
STANDBY 1 SETTINGS
*.FAL_SERVER=PRIMARY
*.FAL_CLIENT=STANDBY
*.STANDBY_FILE_MANAGEMENT=AUTO
*.DB_UNIQUE_NAME=STANDBY
*.LOG_FILE_NAME_CONVERT='PRIMARY','STANDBY'
*.log_archive_dest_1=LOCATION=USE_DB_RECOVERY_FILE_DEST VALID_FOR=(ALL_LOGFILES,ALL_ROLES) DB_UNIQUE_NAME=STANDBY'
*.log_archive_dest_2='SERVICE=PRIMARY LGWR ASYNC VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=PRIMARY'
*.log_archive_dest_3='SERVICE=STANDBY2 LGWR ASYNC VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=STANDBY2'
*.LOG_ARCHIVE_DEST_STATE_1=ENABLE
*.LOG_ARCHIVE_DEST_STATE_2=DEFER
*.LOG_ARCHIVE_DEST_STATE_3=DEFER
*.LOG_ARCHIVE_MAX_PROCESSES=30
STANDBY2 SETTINGS
*.FAL_SERVER=PRIMARY
*.FAL_CLIENT=STANDBY2
*.STANDBY_FILE_MANAGEMENT=AUTO
*.DB_UNIQUE_NAME=STANDBY2
*.LOG_FILE_NAME_CONVERT='PRIMARY','STANDBY2'
*.log_archive_dest_1=LOCATION=USE_DB_RECOVERY_FILE_DEST VALID_FOR=(ALL_LOGFILES,ALL_ROLES) DB_UNIQUE_NAME=STANDBY2'
*.log_archive_dest_2='SERVICE=STANDBY LGWR ASYNC VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=STANDBY'
*.log_archive_dest_3='SERVICE=PRIMARY LGWR ASYNC VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=PRIMARY'
*.LOG_ARCHIVE_DEST_STATE_1=ENABLE
*.LOG_ARCHIVE_DEST_STATE_2=DEFER
*.LOG_ARCHIVE_DEST_STATE_3=DEFER
*.LOG_ARCHIVE_MAX_PROCESSES=30
Edited by: mseberg on Nov 29, 2010 9:39 AM
The first test slapped me. Looking at 409013.1 Cascaded Standby Databases
Edited by: mseberg on Nov 29, 2010 12:49 PM

Data Guard setup after resetlogs on Primary

Hello,
On our primary database we had to perform an incomplete recovery and afterwards had to perform the command "ALTER DATABASE OPEN RESETLOGS". The problem I have is that after switching the Logical and Physical Standby databases back on they are still looking for archive log number 16133 when the log sequences have been reset back to 1.
The error message in the alert log is ....
ORA-16069: Archive Log standby database activation identifier mismatch
So the question is do I need to re-create my Logical and Physical standby databases or is there another may around this?
Thanks.

It depends on the database version, and the configuration in the databases, if the database version is 9i, it's mandatory to re-create the standby database. If the version is 10g, and the flashback database is used, it is possible that automatically the database continue the redo application.

Oracle11g heterogenous data guard Linux-ASM primary, Windows-Filesystem standby

Is it possible to setup a standby database on (Windows 64bit) OS filesystem with 32GB datafile limit that syncs with a primary database on (Linux 64bit) ASM that has datafiles that are larger than 32GB.
There is a Linux-ASM standby in place, this will be an additional standby, to be used as a reporting database using Oracle11g snapshot standby
I have used datapump to export-import from Linux-ASM to Windows-filesystem by precreating additional 32GB datafiles in the Windows database, the import works, it takes too long to import for a reporting database
it's not clear to me how to synchronize the new standby without using RMAN full or incremental backup-restores
Any suggestions or suggested document IDs are welcome!
Thank you,
Drew Salisbury

Hi Drew,
Standby files have to be the same size/distribution as the primary - that can be differnet os, in ASM etc but they have to be the same number and size.
There are some ways to speed up import - parallel etc but in a big database there is always a lot of time lost in index building post data load.
How feasible is it to shrink the datafiles in wondows nack to 32GB......?
Can you set up ASM on windows?
You could use streams or something to replicate and keep the copy you created up to date rather than using physical standby?
Cheers,
Rich

Data guard failover - timing

Hi,
My standby database is running 2 hours behind the primary - to avoid logical corruption problems.
If I do a switchover, how do I make my database roll forward to
1) -30 minutes: so rolling forward
2) - 3 hours: so using flashback to go further back then the delay defined
Regards
Nico

Hi,
As per my understanding with your issue, 2 hours behind its running, perhaps does not mean that your logs are not getting archived to standby. Try to apply the logs manually and flush the current redo from primary and then cancel the recovery session and open the logical standby or switch over the role transition.
Do you have configured the DG broker ?
- Pavan Kumar N

Steps for data guard RAC primary RAC standby

Hi,
I have some doubt in configuring data guard for rac DR to rac primary.
1) In RAC DR, physical standby will be in RAC before starting data guard for RAC Primary , right?.
2) While configuring RAC DR , should one node be down in standby?.
3) After creating the auxiliary database in standby, i mean restoring the rman backup from staging directory in DR,
Shall we have to register the services(database,instance,asm,etc..) to CRS(OCR)?. If yes ,Why?.
4) After DR configure over , Shall i shut/down one node?. Should all the nodes in DR be in up , while applying logs to
standby ?.
If anybody have made the setup and have the prepared document , requesting you to share with me.
Thank you very much,
Sunand

Hi Sunand,
Please follow the following My Oracle Support (MOS) Document ID:
MAA - Creating a RAC Physical Standby for a RAC Primary [ID 380449.1]
For further details, see the referred documents in previous post. Try practicing on VMWare because it will give you the flexibility for creating snapshots to save your work. Please see the answers for your question below:
Q1. In RAC DR, physical standby will be in RAC before starting data guard for RAC Primary , right?.
Answer: Yes, this is recommend but not mandatory. You can add a second node later on.
Q2. While configuring RAC DR , should one node be down in standby?
Answer: Yes, there are some database configuration commands that should be run in exclusive mode. You can start the remaining instance(s) after DB creation. Do not forget to set CLUSTER_DATABASE parameter to TRUE. Also, all instances except one should be close while performing switchover/failover.
Q3. After creating the auxiliary database in standby, i mean restoring the rman backup from staging directory in DR, Shall we have to register the services(database,instance,asm,etc..) to CRS(OCR)?. If yes ,Why?
Answer: Yes, this is recommended for RAC configuration to take advantage of high availability services.
Q4. After DR configure over , Shall i shut/down one node?. Should all the nodes in DR be in up , while applying logs to
standby ?
Answer: You can start up all the instances after DR creation. In this case if your one node goes down log apply services will continue to apply changes in DR.
Hope, this helps to clear your questions.
Regards,
Shahid

Using Data Guard and hot backups - 9.2.0.6.0

Hi all,
I have an existing 9.2.0.6.0 database that is setup in a DataGuard environment - one primary database with a physical standby in a separate datacenter. It is all setup and it works beautifully. On our primary database, we currently have 2 different types of backups we are doing - we do an export of the main schema (all of the application data is all in this one schema) 4 times a day, and we do a full database hot backup once a night.
My question is in regards to the hot backup - I don't know that it is even worth doing a hot backup of this database? I am trying to think of a situation where we would actually want to restore a hot backup of the primary database... If we ran into some kind of a data issue, it would probably be quickest and easiest to restore data from one of the exports, and when we did that restore (import), I assume that data change would be replicated through DataGuard to the standby site. But if there was some kind of situation where we wanted to restore a recent hot backup of the primary database, that would essentially break the Data Guard configuration, and I assume that after the hot backup was restored, we would have to somehow re-instantiate Data Guard on the standby site.
Does anyone have any input on this? If you are running with DataGuard, is it even worth it to be doing hot backups? What kind of situation would call for restoring a hot backup, instead of just failing over to the standby?
Thanks!
--Brad

If we ran into some kind of a data issue, it would probably be quickest and easiest to restore data from one of the exportsIt would be quickest to failover to stand by database than restroing from dump file. After all you maintaing stand by db for that reason.
How can you restore database up to the latest changes using export/import ? You may have to restore using rman and apply the logs.
You can backup from stand by database. You do not need to back up primary.

A Data Guard Question

Dear experts,
This time I have a question regarding Data Guard.
Database:      Oracle 10g Release 2 (10.2.0.3)
OS:          IBM - AIX 5.3 - ML-5
Data Guard:     Physical Standby
We have multiple data guard configuraiton in place and all of them are configured in "MAXIMUM PERFORMANCE" mode.
Currently, we have a separate mount point for archive logs (say /dbarch) on both primary and standby servers.
Once log is archived on primary, it is shipped to standby server and applied.
I think we are wasting space by allocating (/dbarch) on standby server, instead we can share "/dbarch" of primary with standby using NFS.
I remember reading such document. I tried to search in Oracle documentation, google, and metalink for the same but failed :((
Any help in this regard will be very helpful.
Thanks in advance.
Regards

From a DR perpespective, this sounds like a recipe for losing data.
If your primary site has a disaster, and there are logs that have not been applied to the standby then you will never be able to apply them as they will have been lost in the crash.
The point of having the standby is to eliminate a single point of failure - and this mechanism is reintroducing it!
jason.
http://jarneil.wordpress.com

Data Guard -- v$archive_log applied column shows wrong info

I'm playing with 10g Data Guard. Both Primary and Physical Standby are in Maximum Availability mode. When I query v$archived_log column applied for dest_id=2 (which is physical standby) for some files it shows NO value but alert log on both primary and standby shows file transferred info. Even on physical standby v$archived_log shows log is applied YES value. My question is : So why is Primary database's v$archived_log shows value NO ?
I am trying to setup crontab so that once I see value YES in primary's v$archived_log for dest_id = 2 then I can backup archived log file and delete it from primary database machine. But my perl script won't work because primary v$archived_log shows value NO for applied column for dest_id = 2.
Thanks.

Hi OrionNet,
I think I am looking at the wrong column and also on the wrong database for what I need to do. Let me explain what I am trying to achieve.
I have a shell script to check if archived logs are shipped from Primary to Standby AND if Standby successfully applied it or not. My shell script was looking at Primary databases using following query
select sequence#, archived, applied
from v$archived_log
where dest_id = 2 -- running on Primary BUT looking at standby archived log destination
order by sequence# ;
SEQUENCE# ARCHIEVED APPLIED
=====================
58 YES YES
59 YES YES
*60* YES NO
61 YES YES
After reading [v$archived_log reference entry in manual|http://download.oracle.com/docs/cd/B19306_01/server.102/b14237/dynviews_1016.htm#REFRN30011]
APPLIED Indicates whether the archivelog has been applied to its corresponding standby database (YES) or not (NO). The value is always NO for local destinations.
This column is meaningful at the physical standby site for the ARCHIVED_LOG entries with REGISTRAR='RFS' (which means this log is shipped from the primary to the standby database). If REGISTRAR='RFS' and APPLIED is NO, then the log has arrived at the standby but has not yet been applied. If REGISTRAR='RFS' and APPLIED is YES, the log has arrived and been applied at the standby database.
You can use this field to identify archivelogs that can be backed up and removed from disk.
I think I should use following query on Standby database and not on primary database
select sequence#, registrar, applied
from v$archived_log
where dest_id = 1 -- query running on standby so dest_id = 1 which is standby archive log destination
and registrar = 'RFS'
order by sequence# ;
SEQUENCE# REGISTRAR APPLIED
=====================
58 RFS YES
59 RFS YES
*60* RFS YES
61 RFS YES
So, my shell script should connect to standby database from primary database and evaluate which archive logs to delete after backup from primary machine.
Now I'll generate some gaps on Standby and check query again to make sure what I understand and expect is correct.
Hope I am clear now. Thanks for your help. My bad, I didn't read the manual correctly the first time.

DB_FILE_NAME_CONVERT in DATA GUARD

Hi All,
when can i use the below scenario in Data gurad (standby parameter file) ?
DB_FILE_NAME_CONVERT='<DB_UNIQUE_NAME of PRIMARY>','<DB_UNIQUE_NAME of STANDBY>'
LOG_FILE_NAME_CONVERT='<DB_UNIQUE_NAME of PRIMARY>','<DB_UNIQUE_NAME of STANDBY>'
Thanks

Hi,
No worries. In the same doc it states:
Specify the path name and filename location of the primary database datafiles followed by the standby location. This parameter converts the path names of the primary database datafiles to the standby datafile path names. If the standby database is on the same system as the primary database or if the directory structure where the datafiles are located on the standby site is different from the primary site, then this parameter is required. Note that this parameter is used only to convert path names for physical standby databases. Multiple pairs of paths may be specified by this parameter.I've got data guard running on two sites which are physically separate to the primary site. At one site we have the same paths for the data files so we have nothing for the db_file_name_convert parameter. At the other site we have different paths so we specify the full path of the primary, followed by the full path of the standby. Just the way that I have in the example above.
Have you configured your standby yet? If I were you I would keep the file paths the same if possible. It leads to fewer issues in the long run. If that's not possible, for whatever reason, put the full path of the primary followed by the full path of the standby.
Hope that makes sense?
Rob

Upgrade 11.1 to 11.2 with Data Guard In Place

I am planing on upgrading my primary and standby servers to 11gR2 and was wondering what are the best steps. I do not plan on doing a switchover during the process because down time is only about an hour and we can afford it.
Here is what I plan on doing:
1. On Primary - ALTER SYSTEM SET LOG_ARCHIVE_DEST_STATE_2=DEFER SCOPE=BOTH;
2. On Standby - ALTER DATABASE STOP LOGICAL STANDBY APPLY;
3. Upgrade the standby database then the primary database
4. On Standby - ALTER DATABASE START LOGICAL STANDBY APPLY IMMEDIATE;
5. On Primary - ALTER SYSTEM SET LOG_ARCHIVE_DEST_STATE_2=ENABLE;
Let me know if these steps don't sound right. But my question is should I disable archive during the upgrade processes? if I do could this mess up sending archive logs over to primary because there will be missing sequences numbers?
And my second question is do I need to do anything to the Data Guard broker after the upgrade. I currently use Enterprise Manager to manager the setup.
Thanks in advance for any responses. Oh and I am running on Server 2003 if it matters.
Jeff

Is this a physical or logical standby ? If logical, the steps in MOS Doc 437276.1 (Upgrading Oracle Database with a Logical Standby Database In Place) may be of ehlp
HTH
Srini

Database Smart Flash Cache and Data Guard

We are on Oracle 11.2.0.2 on OEL 5 and our considering buying fusion io (PCIe flash drives) for our Production (Primary) database.
My question is; Can we use Physical Data Guard with the Primary side using Database Smart Flash Cache and keep-ing tables and/or partitions in the Flash Cache and have a Standby side that does not have Database Smart Flash Cache?

Yes you can.
Data Guard will not check what type drive you are using. As long as you meet the database and network requirements it will work.
http://www.oracle.com/technetwork/database/features/availability/maa-090890.html
Check out these white papers if you have concerns
"Migration to Automatic Storage Management (ASM)"
"Best Practices for Creating a Low-Cost Storage Grid for Oracle Databases"
In addition here's an example of using two different file systems with Data Guard
http://pythianpang.wordpress.com/2009/07/07/data-guard-asm-primary-to-filesystem-physical-standby-using-rman-duplicate/
Best Regards
mseberg

Data Guard Failover after primary site network failure or disconnect.

Similar Messages

Maybe you are looking for