on 06-10-2013 7:51 AM
Hi,
In the production environment log files are generated and ships to Standby server.
I wants to know whether it is configured with data guard or it is using shell scripts to transferring logs and applying logs
How can I can determine, if dataguard is configured on a database
1. How can i find whether data guard is configured or not.?
2. In which server, should i check ? (Primary DB or Standby DB)
3. Server , DG is configuring? (In Primary or In Standby)
These are the settings in the Standby Server;
PHYSICAL STANDBY
MAXIMUM PERFORMANCE
PRD
Services Informational 0 1 0 NO
08-JUN-13
ARC0: Archival started
Log Transport Services Informational 0 2 0 NO
08-JUN-13
ARC1: Archival started
08-JUN-13
ARC0: Becoming the 'no FAL' ARCH
Log Transport Services Informational 0 4 0 NO
08-JUN-13
...........
...........
...........
...........
...........
FACILITY SEVERITY DEST_ID MESSAGE_NUM ERROR_CODE CAL
------------------------ ------------- ---------- ----------- ---------- ---
TIMESTAMP
---------------
MESSAGE
--------------------------------------------------------------------------------
Log Apply Services Informational 0 38 0 NO
10-JUN-13
Managed Standby Recovery not using Real Time Apply
Log Apply Services Informational 0 39 0 NO
10-JUN-13
FACILITY SEVERITY DEST_ID MESSAGE_NUM ERROR_CODE CAL
------------------------ ------------- ---------- ----------- ---------- ---
TIMESTAMP
---------------
MESSAGE
--------------------------------------------------------------------------------Managed Standby Recovery not using Real Time Apply
Log Apply Services Informational 0 40 0 NO
10-JUN-13
Managed Standby Recovery not using Real Time Apply
I have attached result of dataguard_status.
regards,
zerandib
Hi Zerandib,
you can make it the complex way like Orkun suggested or you just check the init parameters "LOG_ARCHIVE_DEST_n" for a service entry.
Oracle Documentation: http://docs.oracle.com/cd/E11882_01/server.112/e25513/initparams123.htm#I1010086
http://docs.oracle.com/cd/E11882_01/server.112/e25608/log_arch_dest_param.htm#SBYDB01107
shell> slqplus / as sysdba
SQL> show parameter log_archive_dest
The redo stream is not transferred by LNS (implicitly by LGWR or ARC), if you do not find a parameter value with "SERVICE". The redo transport by script (or third party applications like Libelle DBShadow) is the only possibility in such cases.
Regards
Stefan
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hi Zerandib,
1. How can i find whether data guard is configured or not.?
Check the GV$DATAGUARD_STATUS table content in order to understand the data guard has been configured. Additionally, check the tnsnames.ora to see the remote database connection string
2. In which server, should i check ? (Primary DB or Standby DB)
You can check at both sides. As I noted in the first item, go to tnsnames.ora and find the remote database connection
3. Server , DG is configuring? (In Primary or In Standby)
Could you clarify this?
>> I wants to know whether it is configured with data guard or it is using shell scripts to transferring logs and applying logs
After you identify the standby system check the RFS process running ath the standby, execute the command at the sqlplus, on primary;
SQL> alter system switch logfile;
Then, check the trace like below, in the "alert_<DBSID>.log" at the standby side;
Archived Log entry 15385 added for thread 1 sequence 17413 rlc 807919249 ID 0xd78daf39 dest 2:
RFS[129]: No standby redo logfiles created
RFS[129]: Opened log for thread 1 sequence 17414 dbid -511463543 branch 807919249
If you see the string as soon as you created the redo, it means that the redologs are shipped by Oracle not by script.
You can find the additional info;
http://oradbtps.blogspot.com/2012/04/oracle-dataguard-status.html
Best regards,
Orkun Gedik
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hi Orkun,
1. I have checked the tnsnames.ora in standby server;
There i think, remote db is configured. --> HOST = erpdv101
But also note that, there is a comment (#) in front of (HOST = erpsb101).
So only effective line is HOST = erpdv101 , which is the primary database
It having following settings;
PRD.WORLD=
(DESCRIPTION =
(ADDRESS_LIST =
(ADDRESS =
(COMMUNITY = SAP.WORLD)
(PROTOCOL = TCP)
# (HOST = erpsb101)
(HOST = erpdv101)
(PORT = 1527)
)
)
(CONNECT_DATA =
(SID = PRD)
(GLOBAL_NAME = PRD.WORLD)
)
)
I have checked the GV$DATAGUARD_STATUS in standby server as well;
Its having some records like this; Im unable to analyses how it should be when DG is there and DG not there
I have attached DATAGUARD_STATUS result as well
Managed Standby Recovery not using Real Time Apply ----> Whats mean by this line!
(Is that implies log applying is not real time? But will that implies DG is configured;)
Log Apply Services | Informational | 0 | 38 | 0 NO |
10-JUN-13
Managed Standby Recovery not using Real Time Apply
Log Apply Services | Informational | 0 | 39 | 0 NO |
10-JUN-13
FACILITY | SEVERITY | DEST_ID MESSAGE_NUM ERROR_CODE CAL |
------------------------ ------------- ---------- ----------- ---------- ---
TIMESTAMP
---------------
MESSAGE
--------------------------------------------------------------------------------Managed Standby Recovery not using Real Time Apply
Log Apply Services | Informational | 0 | 40 | 0 NO |
10-JUN-13
I will check the RFS process running in the standby.
by the way whats meant by "redo logs are shipped by oracle." - Is that implies DG is configured
regards,
zerandib
Hi Zerandib,
>> Managed Standby Recovery not using Real Time Apply ----> Whats mean by this line!
http://docs.oracle.com/cd/B19306_01/server.102/b14239/log_apply.htm#i1022881
Best regards,
Orkun Gedik
hi Orkun,
SQL> alter system switch logfile;
Then, check the trace like below, in the "alert_<DBSID>.log" at the standby side;
Archived Log entry 15385 added for thread 1 sequence 17413 rlc 807919249 ID 0xd78daf39 dest 2:
RFS[129]: No standby redo logfiles created
RFS[129]: Opened log for thread 1 sequence 17414 dbid -511463543 branch 807919249
If you see the string as soon as you created the redo, it means that the redologs are shipped by Oracle not by script.
Orkun Gedik
I have performed this action;
alter system switch logfile;
and checked the status of both alert_<SID>.log file (In primary & standby);
It took 2-3 minutes to change the status of the alert log file. (to appear the entry there....)
Is that implies, it using shell scripts to transferring log files between servers?
(screenshot is attached)
-------------------
Hope these information will be also useful;
In Primary DB log shipping is happens using program called "brsend", which consists of;
brarchive -u / -f -c -d stage -s
In Standby Server logs are applying using a program called "brapply" , which consists of;
brarchive -u system/passwd -c -f -d disk -m 30 -sd
regards,
zerandib
Hi again,
As far as I understand from the trace that the log shipment does not performed by Oracle. So, stop the scheduled jobs and try to execute "alter system switch logfile;" statement and check the alter log at the standby again. You should see that the redolog files have not been transferred from primary to standby. It means that the files are transferred by using scripts.
To configure the Dataguard, please follow the guide;
http://www.oracle.com/us/solutions/sap/wp-ora4sap-dataguard11g-303811.pdf
Best regards,
Orkun Gedik
User | Count |
---|---|
76 | |
9 | |
8 | |
7 | |
6 | |
5 | |
5 | |
5 | |
5 | |
5 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.