Re: detecting gaps in logical standby environment
- From: Roger Xu <wellmetus@xxxxxxxxx>
- To: GovindanK <gkatteri@xxxxxxxxx>
- Date: Thu, 21 Jan 2010 15:57:41 -0600
I like to know when DBA_LOGSTDBY_LOG gets populated. As soon as the log
switched at the primary site or after the entire redo log file has been
transferred to the standby site?
Let's say the logsize is huge and we have a tiny pipe. At 8:00, log switched
and nothing runs at the primary site after the switch; at 9:00 standby
receives the log file and starts applying. If I use the query below at 8:30,
will I see the gap in this log file? Thanks.
**
On Wed, Jan 20, 2010 at 5:29 PM, GovindanK <gkatteri@xxxxxxxxx> wrote:
>
> Physical standby applies blocks whereas logical standby applies the
> transactions. The commit may follow in the next log and hence logical
> standby's use SCN.
> Log files with a NEXT_CHANGE# below READ_SCN are no longer needed (since
> they have been applied, if not you need them!!). The information in those
> logs has been applied or persistently stored in the database.
>
> For gaps on the Logical Standby run the following query:
> SELECT * FROM dba_logstdby_log
> WHERE next_change# >= ( SELECT read_scn FROM dba_logstdby_progress)
> ORDER BY first_change#
> /
>
>
>
> Roger Xu wrote:
>
> Kellyn,
>
> This is a cool script to detect gaps in the physical standby setup. Does
> anyone has something like this to detect gaps in the logical standby setup?
>
> Thanks,
>
> Roger Xu
>
> On Tue, Jan 12, 2010 at 5:14 PM, Kellyn Pedersen <kjped1313@xxxxxxxxx>wrote:
>
>> When I worked with standby's, I always compared from v$log_history-
>>
>> select next_change# from v$log_history where recid =
>> (
>> select max(recid) from v$log_history);
>> And compared it to the primary with a monitoring script that did this type
>> of steps-
>> sqlplus -s << EOF | read STDBY_CURRENT_ARCH
>> connect / as sysdba
>> set feedback off
>> set pagesize 0
>> select sequence# from v\$log_history
>> where recid =
>> (select max(recid) from v\$log_history);
>> EOF
>> echo Standby archive log: ${STDBY_CURRENT_ARCH}
>> sqlplus -s << EOF | read PRIMARY_CURRENT_ARCH
>> system/`crypt oracle < ${WORKING_DIR}/.pwd`@${REMOTE_SID}
>> set feedback off
>> set pagesize 0
>> select sequence#
>> from v\$archived_log
>> where recid =
>> (select max(recid) from v\$archived_log);
>> EOF
>> echo Primary archive log: ${PRIMARY_CURRENT_ARCH}
>> let ARCH_GAP_ACTUAL=${PRIMARY_CURRENT_ARCH}-${STDBY_CURRENT_ARCH}
>> echo Archive Log Gap: $ARCH_GAP_ACTUAL
>> if (( ${ARCH_GAP_ACTUAL} > ${ARCH_GAP_TOLERANCE} ))
>> then
>> print "${HOST} standby archive log gap is > ${ARCH_GAP_TOLERANCE} logs
>> behind ${REMOTE_SID} at ${DATE}" > ${OUT_FILE}
>> print "${HOST} standby archive log gap is > ${ARCH_GAP_TOLERANCE} logs
>> behind ${REMOTE_SID} at ${DATE}"
>>
>> then mail, etc., so forth and so on....
>>
>> Kellyn Pedersen
>> Multi-Platform DBA
>> I-Behavior Inc.
>> http://www.linkedin.com/in/kellynpedersen
>> www.dbakevlar.blogspot.com
>>
>> "Go away before I replace you with a very small and efficient shell
>> script..."
>>
>>
>> --- On *Tue, 1/12/10, Jiang, Lu <Lu.Jiang@xxxxxxxxxxxx>* wrote:
>>
>>
>> From: Jiang, Lu <Lu.Jiang@xxxxxxxxxxxx>
>> Subject: Find the latest SCN # in phisical standby
>> To: "ORACLE-L" <oracle-l@xxxxxxxxxxxxx>
>> Date: Tuesday, January 12, 2010, 3:43 PM
>>
>>
>> Hi all,
>>
>>
>>
>> I have set up a real time apply physical standby and trying to find how
>> Real Time it is. However it seems that it is hard to find the latest SCN# on
>> physical standby database, v$database (current_scn) only gets updated to the
>> last scn which has been archived. Here is what I got:
>>
>>
>>
>> SQL> select recovery_mode from v$archive_dest_status where dest_id=2;
>>
>> RECOVERY_MODE
>>
>> -----------------------
>>
>> MANAGED REAL TIME APPLY
>>
>>
>>
>> SQL> select name, value, time_computed from v$dataguard_stats where
>> name='apply lag';
>>
>> NAME VALUE TIME_COMPUTED
>>
>> --------------- -------------------- ------------------------------
>>
>> apply lag +00 00:00:00 12-JAN-2010 15:26:02
>>
>>
>>
>>
>>
>> *Primary last change time: *
>>
>>
>>
>> SQL> select scn_to_timestamp(current_scn) from v$database;
>>
>> SCN_TO_TIMESTAMP(CURRENT_SCN)
>>
>>
>> ---------------------------------------------------------------------------
>>
>> 12-JAN-10 03.05.41.000000000 PM
>>
>>
>>
>>
>>
>> *Standby last change time: - *not updated after last log archived
>>
>>
>>
>> SQL> select current_scn from v$database;
>>
>> CURRENT_SCN
>>
>> -----------
>>
>> 579664324
>>
>> SQL> select scn_to_timestamp(579664324) from v$database;
>>
>> SCN_TO_TIMESTAMP(579664324)
>>
>>
>> ---------------------------------------------------------------------------
>>
>> 11-JAN-10 11.27.27.000000000 PM
>>
>>
>>
>> SQL> select max(FIRST_CHANGE#),max(NEXT_CHANGE#) from v$archived_log;
>>
>> MAX(FIRST_CHANGE#) MAX(NEXT_CHANGE#)
>>
>> ------------------ -----------------
>>
>> 579266831 579664325
>>
>>
>>
>> Could someone share some light on this?
>>
>>
>>
>> Thanks,
>>
>> Lu
>>
>>
>>
>
>
Other related posts:
