I am having a problem with an application that requires very low latency (less than 100 ms per txn.) Every once in a while we will see spikes where transactions take 20-30 SECONDS. We looked at the ASH report and we are seeing "log file sync" waits, but we are unclear on the exact source of the problem. I want to compare this problem period with a normally operating period. However, the only way I know of to do this is using AWR compare period reports. The problem with this is that the periods are too long (30 mins). I need some way to compare very short periods of time after the problem has occurred since I have now way of knowing when it will happen. Any advice would be appreciated.
Way to compare very short periods of activity in Oracle
oracleoracle-11g-r2tuning
Related Solutions
Rather than doing this, I'd advise on getting the character set info from the nls_database_parameters
view & then using DBCA
to create a new database, along with dbms_metadata.get_ddl
to handle tablespace creation. - Much easier and less prone to error. The character set and national character set are really the only things that are a pain to change when a database has already been created & DBCA
lets you chose them in the UI. I imagine you're needing to package some scripts up to automate creation or something?
Anyway, if you really must proceed, you can't do exactly what you're asking out of the box, but you can dump the Control File which will give you most of the information you need to construct a CREATE DATABASE
statement, including most pfile parameters (a pfile is required to start a new instance before issuing a CREATE DATABASE
statement), redo logs & character set etc.
For example:
SQL> alter database backup controlfile to trace as '/tmp/db.sql';
Database altered.
SQL> !cat /tmp/db.sql
-- The following are current System-scope REDO Log Archival related
-- parameters and can be included in the database initialization file.
--
-- LOG_ARCHIVE_DEST=''
-- LOG_ARCHIVE_DUPLEX_DEST=''
--
-- LOG_ARCHIVE_FORMAT=%t_%s_%r.dbf
--
-- DB_UNIQUE_NAME="PHIL112"
--
-- LOG_ARCHIVE_CONFIG='SEND, RECEIVE, NODG_CONFIG'
-- LOG_ARCHIVE_MAX_PROCESSES=4
-- STANDBY_FILE_MANAGEMENT=MANUAL
-- STANDBY_ARCHIVE_DEST=?/dbs/arch
-- FAL_CLIENT=''
-- FAL_SERVER=''
--
-- LOG_ARCHIVE_DEST_1='LOCATION=USE_DB_RECOVERY_FILE_DEST'
-- LOG_ARCHIVE_DEST_1='MANDATORY NOREOPEN NODELAY'
-- LOG_ARCHIVE_DEST_1='ARCH NOAFFIRM EXPEDITE NOVERIFY SYNC'
-- LOG_ARCHIVE_DEST_1='NOREGISTER NOALTERNATE NODEPENDENCY'
-- LOG_ARCHIVE_DEST_1='NOMAX_FAILURE NOQUOTA_SIZE NOQUOTA_USED NODB_UNIQUE_NAME'
-- LOG_ARCHIVE_DEST_1='VALID_FOR=(PRIMARY_ROLE,ONLINE_LOGFILES)'
-- LOG_ARCHIVE_DEST_STATE_1=ENABLE
--
-- Below are two sets of SQL statements, each of which creates a new
-- control file and uses it to open the database. The first set opens
-- the database with the NORESETLOGS option and should be used only if
-- the current versions of all online logs are available. The second
-- set opens the database with the RESETLOGS option and should be used
-- if online logs are unavailable.
-- The appropriate set of statements can be copied from the trace into
-- a script file, edited as necessary, and executed when there is a
-- need to re-create the control file.
--
-- Set #1. NORESETLOGS case
--
-- The following commands will create a new control file and use it
-- to open the database.
-- Data used by Recovery Manager will be lost.
-- Additional logs may be required for media recovery of offline
-- Use this only if the current versions of all online logs are
-- available.
-- After mounting the created controlfile, the following SQL
-- statement will place the database in the appropriate
-- protection mode:
-- ALTER DATABASE SET STANDBY DATABASE TO MAXIMIZE PERFORMANCE
STARTUP NOMOUNT
CREATE CONTROLFILE REUSE DATABASE "PHIL112" NORESETLOGS ARCHIVELOG
MAXLOGFILES 16
MAXLOGMEMBERS 3
MAXDATAFILES 100
MAXINSTANCES 8
MAXLOGHISTORY 292
LOGFILE
GROUP 1 '/u01/app/oracle/oradata/PHIL112/redo01.log' SIZE 50M BLOCKSIZE 512,
GROUP 2 '/u01/app/oracle/oradata/PHIL112/redo02.log' SIZE 50M BLOCKSIZE 512,
GROUP 3 '/u01/app/oracle/oradata/PHIL112/redo03.log' SIZE 50M BLOCKSIZE 512
-- STANDBY LOGFILE
DATAFILE
'/u01/app/oracle/oradata/PHIL112/system01.dbf',
'/u01/app/oracle/oradata/PHIL112/sysaux01.dbf',
'/u01/app/oracle/oradata/PHIL112/undotbs01.dbf',
'/u01/app/oracle/oradata/PHIL112/users01.dbf'
CHARACTER SET WE8MSWIN1252
;
-- Commands to re-create incarnation table
-- Below log names MUST be changed to existing filenames on
-- disk. Any one log file from each branch can be used to
-- re-create incarnation records.
-- ALTER DATABASE REGISTER LOGFILE '/u01/app/oracle/fast_recovery_area/PHIL112/archivelog/2013_01_10/o1_mf_1_1_%u_.arc';
-- ALTER DATABASE REGISTER LOGFILE '/u01/app/oracle/fast_recovery_area/PHIL112/archivelog/2013_01_10/o1_mf_1_1_%u_.arc';
-- Recovery is required if any of the datafiles are restored backups,
-- or if the last shutdown was not normal or immediate.
RECOVER DATABASE
-- All logs need archiving and a log switch is needed.
ALTER SYSTEM ARCHIVE LOG ALL;
-- Database can now be opened normally.
ALTER DATABASE OPEN;
-- Commands to add tempfiles to temporary tablespaces.
-- Online tempfiles have complete space information.
-- Other tempfiles may require adjustment.
ALTER TABLESPACE TEMP ADD TEMPFILE '/u01/app/oracle/oradata/PHIL112/temp01.dbf'
SIZE 937426944 REUSE AUTOEXTEND ON NEXT 655360 MAXSIZE 32767M;
-- End of tempfile additions.
--
-- Set #2. RESETLOGS case
--
-- The following commands will create a new control file and use it
-- to open the database.
-- Data used by Recovery Manager will be lost.
-- The contents of online logs will be lost and all backups will
-- be invalidated. Use this only if online logs are damaged.
-- After mounting the created controlfile, the following SQL
-- statement will place the database in the appropriate
-- protection mode:
-- ALTER DATABASE SET STANDBY DATABASE TO MAXIMIZE PERFORMANCE
STARTUP NOMOUNT
CREATE CONTROLFILE REUSE DATABASE "PHIL112" RESETLOGS ARCHIVELOG
MAXLOGFILES 16
MAXLOGMEMBERS 3
MAXDATAFILES 100
MAXINSTANCES 8
MAXLOGHISTORY 292
LOGFILE
GROUP 1 '/u01/app/oracle/oradata/PHIL112/redo01.log' SIZE 50M BLOCKSIZE 512,
GROUP 2 '/u01/app/oracle/oradata/PHIL112/redo02.log' SIZE 50M BLOCKSIZE 512,
GROUP 3 '/u01/app/oracle/oradata/PHIL112/redo03.log' SIZE 50M BLOCKSIZE 512
-- STANDBY LOGFILE
DATAFILE
'/u01/app/oracle/oradata/PHIL112/system01.dbf',
'/u01/app/oracle/oradata/PHIL112/sysaux01.dbf',
'/u01/app/oracle/oradata/PHIL112/undotbs01.dbf',
'/u01/app/oracle/oradata/PHIL112/users01.dbf'
CHARACTER SET WE8MSWIN1252
;
-- Commands to re-create incarnation table
-- Below log names MUST be changed to existing filenames on
-- disk. Any one log file from each branch can be used to
-- re-create incarnation records.
-- ALTER DATABASE REGISTER LOGFILE '/u01/app/oracle/fast_recovery_area/PHIL112/archivelog/2013_01_10/o1_mf_1_1_%u_.arc';
-- ALTER DATABASE REGISTER LOGFILE '/u01/app/oracle/fast_recovery_area/PHIL112/archivelog/2013_01_10/o1_mf_1_1_%u_.arc';
-- Recovery is required if any of the datafiles are restored backups,
-- or if the last shutdown was not normal or immediate.
RECOVER DATABASE USING BACKUP CONTROLFILE
-- Database can now be opened zeroing the online logs.
ALTER DATABASE OPEN RESETLOGS;
-- Commands to add tempfiles to temporary tablespaces.
-- Online tempfiles have complete space information.
-- Other tempfiles may require adjustment.
ALTER TABLESPACE TEMP ADD TEMPFILE '/u01/app/oracle/oradata/PHIL112/temp01.dbf'
SIZE 937426944 REUSE AUTOEXTEND ON NEXT 655360 MAXSIZE 32767M;
-- End of tempfile additions.
--
SQL>
Then you can extract the tablespace creation DDL with:
select dbms_metadata.get_ddl('TABLESPACE', tablespace_name) as ts_ddl
from dba_tablespaces;
If you want me to expand further on all of the steps required to manually create a database, just ask (if you don't already know).
Keep in mind most of the work I do is with PostgreSQL so this may or may not be 100% on the money but I think this should be close enough to be helpful.
The basic issue is that in an environment like this you are going to have a great deal of trouble managing the locks. It seems you can either do some sort of conflict resolution or conflict prevention on the lock level. Conflict prevention seems like despite the performance difficulties, it would reduce the level of user frustration significantly.
My approach here would be indeed to do the locking on the central server in a pl/sql stored procedure, which would insert into the locking table if possible, returning a value indicating success, or if that is not possible, either returning a value indicating failure, or raising an exception (what I have done in the past, for example, is to return something identifying who has the lock if it is already locked).
What I would omit is actually checking with the local server. If you have a high read to write ratio and the chance of collisions is otherwise relatively small, you are going to have to check the central server most times anyway, so there isn't much to be gained by checking the local server also. Certainly you wouldn't want to write to both the local and remote servers for locking. Keep locking simple. It is likely to be a source of pain no matter what you do.
The second thing I would suggest here is that I highly recommend expiring locks like this, perhaps after 2 hours or something. There are two major reasons for doing so. The first is that bugs in the application-layer of the code can cause locks not to be released, and secondly if this is over a web interface, HTTP is stateless and therefore you have no real way of knowing that state has dropped. In this way you can give a lock which is valid for a certain period of time, can be renewed pre-emptively (in the background if needed), and times out if the individual closes the browser window and goes home for the day. Some sort of administrative utility to free locks is also something I would recommend.
I share your sense that 7 sec to acquire a lock is a significant performance cost, but in the end I don't see any better way to do this. Your options are limited significantly by the CAP theorem and a single central locking system is probably what's needed.
I suppose another option is that one could have the central server merely lock to branch location, and have the branch location release the lock as soon as no valid locks have been held for a certain period of time. This might have the advantage of allowing faster collaboration by a team, meaning that only the first editor on a team would have to incur that cost.
Best Answer
Directly query the DBA_HIST_ACTIVE_SESS_HISTORY view if you want a narrower view than an ASH or AWR can provide. Use something like this query: