http://www.forum.askdba.org/ Forum for Oracle DBA's en Sat, 11 Feb 2012 11:44:12 +0000 http://www.forum.askdba.org/topic/ora-29760-instance_number-parameter-not-specified#post-75 Fri, 31 Jul 2009 10:26:39 +0000 Saurabh Sood 75@http://www.forum.askdba.org/ <p>Hi,</p> <p>Try to put the following parameters in this pfile:<br /> *.cluster_database=true<br /> *.cluster_database_instances=2<br /> *.instance_type=asm</p> <p>Regards,<br /> Saurabh Sood </p> http://www.forum.askdba.org/topic/clone-database-using-hot-backup#post-46 Wed, 26 Nov 2008 07:15:52 +0000 daniesh 46@http://www.forum.askdba.org/ <p>Hi Amit,<br /> Thanks for your time and input. </p> http://www.forum.askdba.org/topic/multiple-oracle-home-rac#post-26 Tue, 30 Sep 2008 11:28:21 +0000 Amit Bansal 26@http://www.forum.askdba.org/ <p>1) Yes, you need to use only one ASM and CRS version.</p> <p>2) Yes, in this case it will support 10.2 and 10.1 databases</p> <p>3) Yes it can support. But again if there is any feature which is available in newer release , you will not be able to use it with lower release database. Starting from 11g, you have COMPATIBLE.ASM and COMPATIBLE.RDBMS disk group attributes which specify the compatibility settings for Oracle ASM and database instances. Check following for more detail</p> <p><a href="http://download.oracle.com/docs/cd/B28359_01/server.111/b31107/asmdiskgrps.htm#CHDDIGBJ" rel="nofollow">http://download.oracle.com/docs/cd/B28359_01/server.111/b31107/asmdiskgrps.htm#CHDDIGBJ</a> </p> http://www.forum.askdba.org/topic/ora-01013-user-requested-cancel-of-current-operation#post-34 Fri, 14 Nov 2008 04:22:01 +0000 Amit Bansal 34@http://www.forum.askdba.org/ <p>Hi,</p> <p>Could you actually let us know what exact command you used to shutdown the database. Also give us the full alert log content after you issued shutdown.</p> <p>ORA-1013 would normally arise when you press Ctrl + C after issuing a command at sqlplus prompt..</p> <p>-Amit </p> http://www.forum.askdba.org/topic/error-while-creating-of-snapshot#post-16 Fri, 29 Aug 2008 17:13:39 +0000 Sasi 16@http://www.forum.askdba.org/ <p>Monash_aus,</p> <p>&lt;&lt; trying to create a snapshot in a different schema to that of base &gt;&gt; </p> <p>thats an interesting one.. creating a MV/snapshot in a different schema to that of base schema.</p> <p>Snapshots were used in 8i. From 9i we should use materialized views (MV) . To create a fast refresh MV we need to create a MV log first in the source schema and give access to this MV log to the target schema.</p> <p>In case if you are not familiar with Materialized View (MV) , refer the below steps to create a MV that will get refreshed once an hour.</p> <p>----------------------------------- Start -------------------------------------------------------------------<br /> Lets say you have two schemas , SAM and JOHN .</p> <p>Now ,SAM wants to create a materialized view (TAB1_MV) on JOHN.TAB1 table,</p> <p>1st SAM should have the following privileges granted to him</p> <p>1.As sysdba grant the following privileges to SAM.</p> <p>grant create materialized view to SAM;<br /> grant global query rewrite to SAM;<br /> grant on commit refresh to SAM;</p> <p>2.Login as user JOHN and grant the required table access to SAM,</p> <p>grant select on JOHN.TAB1 to SAM;</p> <p>create materialized view log on JOHN.TAB1 with primary key;</p> <p>The above sql will create a MV log named MLOG$_TAB1 in JOHN schema.</p> <p>grant select on JOHN.MLOG$_TAB1 to SAM;</p> <p>3.Now connect to SAM schema and run the follwoing SQLs,</p> <p>create materialized view TAB1_MV<br /> build immediate<br /> refresh fast<br /> start with sysdate<br /> next sysdate+1/24<br /> with primary key<br /> enable query rewrite<br /> as<br /> SELECT * from TAB1;</p> <p>This creates the required MV and populates the data and refresh takes ple every hour.</p> <p>Hope this helps.</p> <p>-----------------------------------------------------------------------<br /> Note 1:</p> <p>To check if SAM has appropriate privileges to create a materialized view run the following,</p> <p>select grantee,privilege from user_sys_privs ; </p> <p>create materialized<br /> global query rewrite<br /> on commit refresh</p> <p>should see an output something like above</p> <p>Note 2:</p> <p>If the materialized view log name is more than 30 characters then the name would be trunacted automatically while creating MV log. This will give an error as table not found while creating MV.<br /> ----------------------------------------------------- End ----------------------------------------------</p> <p>For more info on MV refer, (9i)</p> <p><a href="http://download.oracle.com/docs/cd/B10501_01/server.920/a96567/repmview.htm" rel="nofollow">http://download.oracle.com/docs/cd/B10501_01/server.920/a96567/repmview.htm</a> </p> http://www.forum.askdba.org/topic/cpu-patch-and-catcpusql#post-62 Tue, 24 Mar 2009 13:48:08 +0000 Amit Bansal 62@http://www.forum.askdba.org/ <p>Daniesh,</p> <p>I would say it's a question which never came to my mind. This is my understanding from the captcpu.sql notes. </p> <p>Catcpu.sql inturn calls a script called catbundle.sql which is located under $ORACLE_HOME/rdbms/admin</p> <p>This script will load sql files in database by looking in bundledata_CPU.xml for file information for particular patch. It will be creating a dynamic apply sql file and run it.</p> <p>Hope this helps.</p> <p>Cheers<br /> Amit </p> http://www.forum.askdba.org/topic/pkrs-1009-failed-to-start-asm-instance#post-156 Wed, 02 Jun 2010 08:12:10 +0000 Anonymous 156@http://www.forum.askdba.org/ <p>Permissions on init+ASM2.ora is right.I am only facing this problem when i was starting ASM through srvctl command otherwise it's working fine.</p> <p>Please note this point. </p> http://www.forum.askdba.org/topic/asm-tool#post-36 Thu, 20 Nov 2008 08:21:01 +0000 Amit Bansal 36@http://www.forum.askdba.org/ <p>asmtool or asmtoolg(Gui version) helps to stamp the new disks on windows for using as ASM Disks. You can use asmtoolg utility present in %ORACLE_HOME%\bin directory which will provide GUI for stamping the disks.</p> <p>As per Oracle docs </p> <pre>asmtoolg shows the devices available on the system. Unrecognized disks are labeled as a "Candidate device." Unformatted partitions are labeled as "Oracle raw device file." Stamped ASM disks are labeled as "Stamped ASM disk," and unstamped ASM disks are labeled as "Unstamped ASM disks." The tool also shows disks that are recognized by Windows as a file system (such as NTFS). These disks are not available for use as ASM disks, and cannot be selected. </pre> <p>You can find the above information and also steps to use asmtoolg at below link</p> <p><a href="http://download.oracle.com/docs/cd/B28359_01/install.111/b28250/racstorage.htm#BABCAFIB" rel="nofollow">http://download.oracle.com/docs/cd/B28359_01/install.111/b28250/racstorage.htm#BABCAFIB</a> </p> http://www.forum.askdba.org/topic/11g-useful-sql#post-246 Fri, 22 Jul 2011 13:12:27 +0000 Saurabh Sood 246@http://www.forum.askdba.org/ <p>Amit,</p> <p>In 10.2.0.4, if we do not want to apply patch, we can do this by setting event 5614566 at session level and then using dbms_shared_pool.purge (may have to run dbmspool.sql if purge is not present)</p> <p>Cheers!! </p> http://www.forum.askdba.org/topic/dr-rebuild-data-restore-issue#post-140 Sun, 14 Mar 2010 06:15:39 +0000 Saurabh Sood 140@http://www.forum.askdba.org/ <p>Please update the complete backup script that you are using to take backup </p> http://www.forum.askdba.org/topic/error-while-addign-a-disk-to-asm-diskgroup#post-19 Fri, 12 Sep 2008 16:18:31 +0000 Saurabh Sood 19@http://www.forum.askdba.org/ <p>Anu,</p> <p>1. Does the disk appear under v$asm_disk?<br /> You can check it as:</p> <p>select group_number, name, path from v$asm_disk;</p> <p>2. Is this a new LUN created?</p> <p>3. Have you used asmtoolg to delete any existing lun ?<br /> If asmtoolg was used to delete any existing lun then it is mandatory to bounce the ASM instance and then try to add a lun from scratch.</p> <p>-Saurabh Sood </p> http://www.forum.askdba.org/topic/checking-listener-log-in-11g#post-173 Thu, 12 Aug 2010 12:51:29 +0000 Amit Bansal 173@http://www.forum.askdba.org/ <p>Starting Oracle 11g, architecture for trace file and log files have changed. Alert.log and listener.log files are now XML based. </p> <p>To check listener log in 11g, following steps need to be done</p> <p>$adrci<br /> adrci&gt; show homes<br /> ADR Homes:<br /> diag/tnslsnr/orcl11/listener<br /> adrci&gt; set home diag/tnslsnr/orcl11/listener<br /> adrci&gt; show tracefile<br /> diag/tnslsnr/orcl11/listener/trace/listener.log<br /> adrci&gt; show trace listener.log </p> http://www.forum.askdba.org/topic/error-in-invoking-target-client_sharedlib#post-69 Wed, 08 Jul 2009 04:51:41 +0000 Saurabh Sood 69@http://www.forum.askdba.org/ <p>Daniesh,</p> <p>paste the content of installation log file, it will help me to know what exactly is happening at your end.</p> <p>-- Saurabh Sood </p> http://www.forum.askdba.org/topic/monitoring-top-10-memory-consuming-process-on-aix#post-79 Tue, 25 Aug 2009 16:30:35 +0000 Amit Bansal 79@http://www.forum.askdba.org/ <p>At times, you wish to find the processes with maximum memory consumption. This can be easily done using following command. Copy contents in file and run it like as vmstat command</p> <pre> ### Usage mem_mon.sh (interval) (occurences) e.g sh mem_mon.sh 2 10 ### ### This will give top 10 memory consuming processes ##### wt_time=$1 lmt=$2 let a=0 while [[ $a -lt $lmt ]] ; do ps axwv |sort -nrk 7,7|grep -v "COMMAND" \ |awk 'BEGIN {printf ("MemoryinKb \t ProcessId \t PerCPU \t PerMem \t Command \n")}{t=t+$7 - $10 ;printf("%13d \t %10d \t %6.2f \t %6.2f \t %s \n",$7 - $10,$1,$11,$12,$13)}END{print t}'|head -10 let a+=1 sleep $wt_time done </pre> http://www.forum.askdba.org/topic/rman-duplicate-with-cold-backup#post-27 Thu, 02 Oct 2008 07:20:10 +0000 Saurabh Sood 27@http://www.forum.askdba.org/ <p>Hi Ani,</p> <p>It requires target database in either MOUNT or OPEN stage to duplicate database using RMAN. Could you give the complete RMAN command that you are using to duplicate the database? </p> http://www.forum.askdba.org/topic/register-standby-database-in-recovery-catalog#post-188 Sun, 17 Oct 2010 04:15:52 +0000 Amit Bansal 188@http://www.forum.askdba.org/ <p>Daniesh,</p> <p>Apologies, didnt read the question properly. I have not seen any configuration using different catalog for primary and standby (there could be but I havent seen it). Moreover purpose of using rman at standby site is to reduce workload on primary so that in case of failure you can restore the backup. In case of failure, you would need to connect to this very catalog. Anyways I think you can register first primary database at catalog and then take standy backups.<br /> Checked docs (<a href="http://download.oracle.com/docs/cd/B28359_01/backup.111/b28270/rcmcatdb.htm#CIHECCEF" rel="nofollow">http://download.oracle.com/docs/cd/B28359_01/backup.111/b28270/rcmcatdb.htm#CIHECCEF</a>) and they mention that registration can only be done at primary database.</p> <p>-Amit </p> http://www.forum.askdba.org/topic/ora-16055#post-154 Wed, 02 Jun 2010 08:08:05 +0000 Anonymous 154@http://www.forum.askdba.org/ <p>you have a standby database and the standby needs a log that likely you already removed where it thinks it should be /u05/arch/ </p> http://www.forum.askdba.org/topic/datapump-dumpfile-with-dates#post-152 Thu, 20 May 2010 09:34:05 +0000 Saurabh Sood 152@http://www.forum.askdba.org/ <p>Hi,</p> <p>As you mentioned earlier, and that i missed..... You are using this on windows ..</p> <p>The syntax which I have mentioned is not for windows but will work only for unix platforms only.. </p> http://www.forum.askdba.org/topic/asm-and-multipathing#post-111 Wed, 09 Dec 2009 12:29:40 +0000 deepika.dhamija@gmail.com 111@http://www.forum.askdba.org/ <p>Amit,</p> <p>I am finally able to achieve what I intended to do.</p> <p>Many Thanks to you for your valuable suggestions. </p> http://www.forum.askdba.org/topic/rman-05001-while-using-rman-duplicate#post-102 Thu, 22 Oct 2009 10:40:58 +0000 Saurabh Sood 102@http://www.forum.askdba.org/ <p>RMAN-05001 can occur while cloning if:</p> <p>1. The target server have same directory structure as of the source i.e you have to use something like:</p> <p>db_file_name_convert=("/u01/app/oracle/1020/","/u01/app/oracle/1020/")</p> <p>and while issuing:</p> <p>RMAN &gt; duplicate target database to aux;</p> <p>RMAN-05001: auxiliary filename /u01/app/1020/fact01.dbf conflicts with a file used by the target database</p> <p>SOLUTION:<br /> =========<br /> Use "NOFILENAMECHECK" clause as:</p> <p>RMAN &gt; duplicate target database to aux NOFILENAMECHECK;</p> <p>Cheers!!!<br /> Saurabh Sood </p> http://www.forum.askdba.org/topic/oracle-11g-install-performance-on-sun-t5220#post-32 Thu, 13 Nov 2008 06:44:16 +0000 Amit Bansal 32@http://www.forum.askdba.org/ <p>I have seen dbca taking time when all options like JVM,XML , OLAP are chosen. By default all these are checked in and you have to uncheck them if you do not want it. Did you see the place where DBCA was taking more time? was it during data dictionary creation phase?</p> <p>-Amit </p> http://www.forum.askdba.org/topic/asm-not-starting-up-after-system-reboot#post-15 Fri, 29 Aug 2008 07:21:04 +0000 Amit Bansal 15@http://www.forum.askdba.org/ <p>Cool!! Do update ASM alert log entry also in case issue does not resolve.. </p> http://www.forum.askdba.org/topic/asm-volumes-and-diskgroups#post-117 Wed, 30 Dec 2009 04:19:40 +0000 deepika.dhamija@gmail.com 117@http://www.forum.askdba.org/ <p>Thanks Amit for your reply.</p> <p>I am working for a company that makes protection software and we are planning to extend it to support RAC.<br /> Yes, we do realize that RAC Databases are complex and we therefore are also getting in touch with a DBA. These were some of the preliminary doubts that I had.</p> <p>Thanks anyways for your response. </p> http://www.forum.askdba.org/topic/easy-conversion-of-datablock-addressdba-to-file-and-block#post-84 Sat, 29 Aug 2009 13:39:17 +0000 amansharma 84@http://www.forum.askdba.org/ <p>Good one! DBMs-Util is a good package.I was going to add this as well in the blog post but than just felt lazy and left it :-). </p> <p>Are the comments working on my blog now?</p> <p>Aman.... </p> http://www.forum.askdba.org/topic/disabling-tablespace-threshold-alert-for-particular-tablespace#post-85 Thu, 10 Sep 2009 10:51:59 +0000 Amit Bansal 85@http://www.forum.askdba.org/ <p>To disable Threshold for tablespace (Percent type), use following statement</p> <pre> exec DBMS_SERVER_ALERT.SET_THRESHOLD( - metrics_id =&gt; DBMS_SERVER_ALERT.TABLESPACE_PCT_FULL, - warning_operator =&gt; DBMS_SERVER_ALERT.OPERATOR_DO_NOT_CHECK,- warning_value =&gt; '0', critical_operator =&gt; DBMS_SERVER_ALERT.OPERATOR_DO_NOT_CHECK,- critical_value =&gt; '0', observation_period =&gt; 1,- consecutive_occurrences =&gt; 1, instance_name =&gt; NULL, - object_type =&gt; DBMS_SERVER_ALERT.OBJECT_TYPE_TABLESPACE,- object_name =&gt; 'USERS'); </pre> <p>Replace tablespace USER with the tablespace you want to exclude from monitoring</p> <p>If you have also enabled tablespace alerts based on Bytes left, then use following</p> <pre> exec DBMS_SERVER_ALERT.SET_THRESHOLD( - metrics_id =&gt; DBMS_SERVER_ALERT.TABLESPACE_BYT_FREE, - warning_operator =&gt; DBMS_SERVER_ALERT.OPERATOR_DO_NOT_CHECK, - warning_value =&gt; '0', - critical_operator =&gt; DBMS_SERVER_ALERT.OPERATOR_DO_NOT_CHECK, - critical_value =&gt; '0', observation_period =&gt; 1, - consecutive_occurrences =&gt; 1, - instance_name =&gt; NULL, - object_type =&gt; DBMS_SERVER_ALERT.OBJECT_TYPE_TABLESPACE, - object_name =&gt; 'USERS'); </pre> http://www.forum.askdba.org/topic/how-to-connect-11gr2-rac-db-using-sql#post-159 Sat, 12 Jun 2010 11:06:24 +0000 Anonymous 159@http://www.forum.askdba.org/ <p>HI, Looking like a interesting thread going on. Its good that people are actually willing to share their views on this topic. </p> http://www.forum.askdba.org/topic/sysasm-not-granted-to-sys#post-119 Tue, 12 Jan 2010 06:25:34 +0000 Amit Bansal 119@http://www.forum.askdba.org/ <p>As per upgrade guide, you had to add a user to password file before upgrading ASM instance.</p> <p><a href="http://download.oracle.com/docs/cd/B28359_01/server.111/b28300/upgrade.htm#BABEJBFD" rel="nofollow">http://download.oracle.com/docs/cd/B28359_01/server.111/b28300/upgrade.htm#BABEJBFD</a></p> <p>Did you try recreating password file for ASM?Use orapwd to do same.<br /> If it still doesn't work, then would recommend you to raise SR. </p> http://www.forum.askdba.org/topic/exclude-parameter-in-import-script#post-89 Wed, 16 Sep 2009 16:04:25 +0000 Saurabh Sood 89@http://www.forum.askdba.org/ <p>Hi Alok,</p> <p>Here is a very quick reply (The first very basic thing):</p> <p>In 9i there is no EXCLUDE clause. To use EXCLUDE you have to use DATAPUMP.</p> <p>In 9i :<br /> 1. Get the list of all tables by using :<br /> select table_name from dba_tables where table_name not in (those 20 tables) ......and some more clauses here....</p> <p>2. Now you have a list of all the tables that you want to import. Use TABLES= clause with imp : use $ imp help=y also</p> <p>3. prepare a par file and do the import.</p> <p>Cheers!!!<br /> Saurabh Sood </p> http://www.forum.askdba.org/topic/database-shutdown-hangs#post-123 Wed, 20 Jan 2010 14:42:21 +0000 Saurabh Sood 123@http://www.forum.askdba.org/ <p>Hi Pratheep,</p> <p>Processes with status "D" cannot be killed at all. They are in "D" status because they are sleeping and waiting for a resource like "Kernel lock" etc. which they will never get.</p> <p>If you are seeing these ospid in v$process they will not allow "shutdown immediate" or "shutdown abort".</p> <p>The only way to remove such processes is to "reboot" server.</p> <p>These processes cannot be traced down.</p> <p>Check If you can apply Dead Connection detection (DCD) to avoid such situations in future.</p> <p>Cheers!!! </p> http://www.forum.askdba.org/topic/oracle-10g-installation-with-4gb-ram#post-38 Mon, 24 Nov 2008 05:08:09 +0000 Saurabh Sood 38@http://www.forum.askdba.org/ <p>Hi DK,</p> <p>/3GB /PAE support is only available if you are using server with more than 4GB RAM.</p> <p>If you have more than 4GB of RAM then you can change your server's boot.ini file and put a /3GB /PAE in it and can utilize memory upto 16GB.</p> <p>Regards,<br /> Saurabh Sood </p> http://www.forum.askdba.org/topic/monitoring-archive-gap-in-standby#post-80 Tue, 25 Aug 2009 16:35:50 +0000 Amit Bansal 80@http://www.forum.askdba.org/ <p>You can use following sql to monitor the gap in standby using following sql</p> <pre> sqlplus -s "/ as sysdba" &lt;&lt;EOF set head off set feedb off select (select name from V\$DATABASE),(select max(sequence#) from v\$archived_log where dest_id=1) Current_primary_seq,( select max(sequence#) from v$\archived_log where to_date(next_time,'dd-mm-yyyy') &gt; sysdate-1 and dest_id=2 ) max_stby,(select nvl((select max(sequence#) - min(sequence#) from v\$archived_log where to_date(next_time,'dd-mm-yyyy') &gt; sysdate-1 and dest_id=2 and applied='NO'),0) from dual) "To be applied",((select max(sequence#) from v\$archived_log where dest_id=1) - (select max(sequence#) from v\$archived_log where dest_id=2)) "To be Shipped" from dual / </pre> <p>Assumptions -<br /> Dest_id=1 --&gt; Primary DB<br /> Dest_id=2 --&gt;Standby site</p> <p>This query reports the number of archives which need's to be shipped along with number of archive log's which need to be applied. If you are looking for script, you can get the logic from following thread</p> <p><a href="http://forums.oracle.com/forums/thread.jspa?messageID=3708034" rel="nofollow">http://forums.oracle.com/forums/thread.jspa?messageID=3708034</a> </p> http://www.forum.askdba.org/topic/db-and-listener-availability-script#post-134 Thu, 11 Mar 2010 06:13:39 +0000 Amit Bansal 134@http://www.forum.askdba.org/ <p>Hi,</p> <p>Find below a simple script to monitor db and listener. It uses utl_mail procedure to connect to a mailserver db and send a mail in case of failure.<br /> You can also use mailx function to send mail.<br /> I have hard-coded password , but you can also use Oracle wallet to store password. Refer to my blog post <a href="http://askdba.org/weblog/2009/09/using-oracle-wallet-to-execute-shell-scriptcron-without-hard-coded-oracle-database-password/"> here </a> for details</p> <p>You need to pass ORACLE_SID as parameter. So to schedule a cron to check every 10 mins, you need to specify like below</p> <pre> 00,10,20,30,40,50 * * * * sh /home/oracle/monitor/db_check.sh orcl </pre> <p>Find below the script. Any suggestions/improvements are welcome</p> <pre> ################################################################### ## Check Database Availability ################################################################### ##########Env Variables ############# export DBA_MAIL1='amit.bansal@askdba.org' export DBA_MAIL2='saurabh.sood@askdba.org' export ORACLE_HOME=/oracle/ora_db export TNS_ADMIN=$ORACLE_HOME/network/admin export BASE_DIR=/home/oracle/monitor export ORACLE_SID=$1 export LOG_FILE=$BASE_DIR/$ORACLE_SID.log export PATH=$ORACLE_HOME/bin:$PATH export machinename=<code>hostname</code> ################################### cd $BASE_DIR echo "<code>date</code> " &gt; $LOG_FILE echo "Oracle Database(s) Status <code>hostname</code> :\n" &gt;&gt; $LOG_FILE db_cnt=<code>ps -ef|grep pmon|grep $ORACLE_SID|wc -l</code> if [ $db_cnt -gt 0 ]; then echo "$ORACLE_SID is Up" &gt;&gt; $LOG_FILE else echo "$ORACLE_SID is Down" &gt;&gt; $LOG_FILE $ORACLE_HOME/bin/sqlplus -s "system@maildb/oracle123" &lt;&lt;EOF &gt;&gt; $LOG_FILE alter session SET smtp_out_server = 'mailserv:100'; BEGIN UTL_MAIL.send( sender =&gt; 'webmaster@askdba.org' ,recipients =&gt; '${DBA_MAIL1}' ,cc =&gt; '${DBA_MAIL2}' ,subject =&gt; '${ORACLE_SID} is down on ${machinename}' ,message =&gt; '${ORACLE_SID} is Down on ${machinename}' ,mime_type =&gt; 'text/plain; charset=us-ascii' ,priority =&gt; 3); END; / exit EOF fi lsn_cnt=<code>ps -ef|grep tnslsnr|grep oracle|grep -v grep|wc -l</code> if [ $lsn_cnt -gt 0 ]; then echo "Listener is up" &gt;&gt;$LOG_FILE else echo "Listener is down" &gt;&gt;$LOG_FILE $ORACLE_HOME/bin/sqlplus -s "system@maildb/oracle123" &lt;&lt;EOF &gt;&gt; $LOG_FILE alter session SET smtp_out_server = 'mailserv:100'; BEGIN UTL_MAIL.send( sender =&gt; 'webmaster@askdba.org' ,recipients =&gt; '${DBA_MAIL1}' ,cc =&gt; '${DBA_MAIL2}' ,subject =&gt; 'Listener is down on ${machinename}' ,message =&gt; 'Listener is Down on ${machinename}' ,mime_type =&gt; 'text; charset=us-ascii' ,priority =&gt; 3); END; / exit EOF fi </pre> http://www.forum.askdba.org/topic/convert-from-2-node-rac-to-single-instance-database#post-130 Thu, 25 Feb 2010 05:35:13 +0000 Amit Bansal 130@http://www.forum.askdba.org/ <p>Hi Tsswamy,</p> <p>I have not done this kind of migration. I believe that starting a database with non-rac binary and removing the cluster* parameters should allow you to do same. I would recommend you to contact My Oracle Support for exact steps to do same. </p> http://www.forum.askdba.org/topic/what-is-the-processsteps-to-start-up-a-database-in-a-rac#post-121 Thu, 14 Jan 2010 01:50:00 +0000 Amit Bansal 121@http://www.forum.askdba.org/ <p>Hi,</p> <p>Sorry for late reply. Didn't get time to reply earlier.</p> <p>In case of RAC database (10.2) you need to perform following steps</p> <p>1)Start CRS (as root)<br /> #crsctl start crs</p> <p>Actually if you have enabled automatic startup of resources and configured asm-instance dependency correctly, then it will start all resources automatically. Otherwise go to below steps<br /> 2)Start Nodeapps</p> <p> $srvctl start nodeapps -n &lt;noden&gt;<br /> 3)Start ASM instance</p> <p>srvctl start asm -n &lt;noden&gt;</p> <p>Repeat steps on other node. You can use srvctl command from any node</p> <p>4) Now start cluster database<br /> $srvctl start database -d &lt;dbname&gt;</p> <p>You can also start a single instance using following command</p> <p>srvctl start instance -d &lt;db_name&gt; -i &lt;inst_name&gt; </p> http://www.forum.askdba.org/topic/script-to-generate-oracle-add-service-commands#post-243 Thu, 14 Jul 2011 07:51:16 +0000 Amit Bansal 243@http://www.forum.askdba.org/ <p>Suppose you have 10g database which you are trying to migrate to 11g. You wish to create a file to generate command for adding services. You can use below script to do same   <br />  <br /> <pre><br /> srvctl config service -d db_name &gt;services.lst<br /> cat services.lst|while read ln<br /> do<br /> echo $ln|sed 's/PREF//'|sed 's/AVAIL//'|awk -F : '{print "-s " $1 "-r" $2 "-a" $3}'|sed 's/ /,/g'|sed 's/-r,/-r/'|sed 's/,-a/ -a/'|sed 's/-a,/ -a / '|sed 's/-s,/ -s / '|sed 's/,-r/ -r / '|sed 's/-a$/ /'|sed 's/^/srvctl add service -d db_name/'<br /> done<br /> </pre></p> <p>Thanks to Alok Kumar </p>