RMAN :: Output File Permissions With ASM / Grid Infrastructure
Jul 10, 2013
I've noticed that the backup files produced by RMAN (Database is 11.2.0.2.0 Enterprise Edition) )on a RHEL5.5. server which is running ASM have group permissions assigned to ASMADMIN (so the file permissions are assigned to ORACLE:ASMADMIN) as opposed to OINSTALL or any other group.
While this is not an issue (it's not currently impacting us in any way) I'm curious as to how and why this happens. I suspect it is related to the fact that the oracle executable has these permissions also but I can't find any material to confirm my suspicion.
I'm trying to install Grid Infrastructure for Oracle RAC 11g. I followed each step of pre installation tasks, but when I run run Installer it blocks without any error message.
What I see on my shell is this: ./runInstaller Avvio di Oracle Universal Installer in corso...
Verifica dello spazio Temp: deve essere maggiore di 120 MB. Effettivi 17276 MB Superato Verifica dello spazio di swap: deve essere maggiore di 150 MB. Effettivi 23999 MB Superato Verifica del monitor: deve essere configurato per visualizzare almeno 256 colori. Effettivo 16777216 Superato Preparazione per l'avvio di Oracle Universal Installer da /tmp/OraInstall2012-01-11_03-35-57AM. Attendere...[oracle@ssts2cora1 grid]$
Everything seems ok. But the window of oracle installation does not appear. The java process is active:
but it does nothing!! Nothing appears to me. I attach the log file generated by the installation process.
Using paramFile: /home/software_ssts/grid/install/oraparam.ini
Checking Temp space: must be greater than 120 MB. Actual 17276 MB Passed Checking swap space: must be greater than 150 MB. Actual 23999 MB Passed Checking monitor: must be configured to display at least 256 colors. Actual 16777216 Passed
The commandline for unzip:
/home/software_ssts/grid/install/unzip -qqqo ../stage/Components/oracle.jdk/1.5.0.17.0/1/DataFiles/\*.jar -d /tmp/OraInstall2012-01-11_03-35-57AM Verr� utilizzato il valore di umask '022' disponibile in oraparam.ini
Execvp of the child jre : the cmdline is /tmp/OraInstall2012-01-11_03-35-57AM/jdk/jre/bin/java, and the argv is /tmp/OraInstall2012-01-11_03-35-57AM/jdk/jre/bin/java -Doracle.installer.library_loc=/tmp/OraInstall2012-01-11_03-35-57AM/oui/lib/linux -Doracle.installer.oui_loc=/tmp/OraInstall2012-01-11_03-35-57AM/oui -Doracle.installer.bootstrap=TRUE -Doracle.installer.startup_location=/home/software_ssts/grid/install -Doracle.installer.jre_loc=/tmp/OraInstall2012-01-11_03-35-57AM/jdk/jre [code].......
I got huge problem trying to install grid infrastructure and oracle db. Separately oracle db works fine and installation is going without any problems as for grid not so much.
i got vmware host 4.1 with several vm's on top. One of them i openfiler which gives 2 luns for installation. Two other vm's are host of windows 2008r2(comes only 64bits i know that many of you know but just to be clear). Those two windows should be connected in oracle cluster and should have one oracle db afterwards. I have already AD domain, free disks and all required staff arranged and prepared.
I wont bodder you ppl about previous problems but current one is that i have no clue what's going wrong. I got two logs file which i paste below but there's no error message just some warning which isn't described even properly as for me.
Attached File(s)
installtion_logs.zip ( 66.62K ) Number of downloads: 0
We have a two node 11.2.0.3 Grid Infra cluster running on SLES11 Linux. Everything is ok with the SCAN and standard listeners.
However we have a listener running on each node which is used by Oracle Gateway. These listeners were created manually using info provided by Oracle (i.e. definition copied to $ORACLE_HOME/network/admin/listener.ora and started using lsnrctl). This all works.
These listeners are currently not managed by Grid Infra. We would like them to be.
I thought this was easy enough using srvctl to register them with Grid Infra using...
These commands worked but did give a warning stating that the -n flag is deprecated and will be ignored.
This has left us with the following listener config....
grid@MYNODE1:~> srvctl status listener -n mynode1 | grep mynode Listener LISTENER_MYNODE1_GW is enabled on node(s): mynode1 Listener LISTENER_MYNODE1_GW is not running on node(s): mynode1 Listener LISTENER_MYNODE2_GW is enabled on node(s): mynode1 Listener LISTENER_MYNODE2_GW is not running on node(s): mynode1
grid@MYNODE1:~> srvctl status listener -n mynode 2 | grep mynode Listener LISTENER_MYNODE1_GW is enabled on node(s): mynode2 Listener LISTENER_MYNODE1_GW is not running on node(s): mynode2 Listener LISTENER_MYNODE2_GW is enabled on node(s): mynode2 Listener LISTENER_MYNODE2_GW is not running on node(s): mynode2
You can see that Grid Infra knows about the listeners but they are available on both nodes. What we would like is a result like the following example....
grid@MYNODE1:~> srvctl status listener -n mynode1 | grep mynode Listener LISTENER_MYNODE1_GW is enabled on node(s): mynode1 Listener LISTENER_MYNODE1_GW is not running on node(s): mynode1
grid@MYNODE1:~> srvctl status listener -n mynode 2 | grep mynode Listener LISTENER_MYNODE2_GW is enabled on node(s): mynode2 Listener LISTENER_MYNODE2_GW is not running on node(s): mynode2
i.e one listener registered with just the node it is running on. This used to be possible with using the -n flag but no longer works.
I am trying for 11gR2 3 node RAC ASM on windows 2003-64-bit. I am using one SCAN ip ,placed in /etc/hosts file.I am not using DNS/GNS. Same thing done(/etc/hosts) on other nodes also.
I have run the cluvfy.bat prior to grid software install and passed very well. When i tried to install Grid Software,at prerequisites i got error as "node connectivity".
Even i ignored this error and run the finish.After moving grid software to remote nodes ,it throws error as " [INS-20802] Grid Infrastructure Configuration failed ".
It specified one log,which i am attaching here. In network side ,the ping is ok to all the nodes.
Here i am attaching the error screen shots and error log file.
I have raised SR with Oracle ,but not able to identify the root cause of this problem.I will post the solution once i got from oracle.
I am preparing to install grid infrastructure 11.2.0.3 on AIX 6.2 USING AIX LPARS. I am getting the following errors in the output when runningn the following command...
I'm installing Oracle Grid 11gr2 on windows 2003 server. I installed successfully without domain controller. But when i try to install with Active Domain controller i faced following error
i am trying to install 2 node RAC on Oracle VMs. Before the installation during the -preinst check there were few issues which were resolved (ex user equivalence). After that during the installation process of the Grid it failed at step "Configure Oracle Grid Infrastructure for a cluster". After it failed at this step, subsequent steps too failed which I asked OUI to ignore and then I ran both the post installation scripts. And then ran post crsinst which failed. Pasting below the output of the root.sh script, post crsinst and other checks.
************************************* [root@bsfrac01 grid]# sh root.sh Running Oracle 11g root.sh script...
Enter the full pathname of the local bin directory: [/usr/local/bin]:
Copying dbhome to /usr/local/bin ... Copying oraenv to /usr/local/bin ... Copying coraenv to /usr/local/bin ...
Creating /etc/oratab file... Entries will be added to the /etc/oratab file as needed by Database Configuration Assistant when a database is created.Finished running generic part of root.sh script.Now product-specific root actions will be performed.
2011-02-13 00:11:55: Parsing the host name 2011-02-13 00:11:55: Checking for super user privileges 2011-02-13 00:11:55: User has super user privileges
I have some confusions regarding to 11gR2 RAC installation:
- Is Grid Infrastructure only the solution to install Clusterware Application 11gR2? My requirement is to install only the Cluster ware Software for 11gR2 (For 11gR1 we have "linux.x64_11gR1_clusterware.zip") using ocfs2 file system, Archive Mode not enabled.
I am in process of installing Oracle 11g R2 Grid Infrastructure on 2-node virtual RAC. All went very well until the Grid Installation stuck on 65%. It is saying copying file on remote node ('Copying Oracle Home '/u01/app/11.2.0/grid' to remote nodes 'cxm-vl-rman02'). From the logs it is difficult to figure out where the bottlenecks lies.
i have installed grid(crs+asm) on my two linux node(Enterprise linux 5) setup, using openfiler(SAN) , i have configure grid Home like ORACLE_ HOME=/ u01/ home/app/11.2.0/grid.
1) i don't get any instance specific asm parameter file like init+ASM1.ora instead of that i got ab_+ASM1.dat, hc_+ASM1.dat, init.ora, orapw+ASM four files, do i need to create ASM instance specific parameter manually to create diskgroup in the same location,
2) +DATA diskgroup has been created while grid installation, but when i try to create database, its giving me error diskgroup +DATA doesn't exit or not mount, but when i check it shows me mount like below $cd $ORACLE_HOME/dbs (here oracle home is grid home)$export ORACLE_SID=+ASM1$sqlplus / as sysdba>select name,state from v$asm_diskgroup; name state DATA Mounted even i cant able to dismount diskgroup DATA, it throws an error insufficient privilege.
3) After grid installation how can one create diskgroup i mean what needs to exported as ORACLE_SID like in 10g we have $export ORACLE_SID=+ASM1 then creat diskgroup.
We install grid using Grid Installer GUI. When doing a fresh installation of Grid Infrastructure 11.2.0.2 or 11.2.0.3 in a new server , I want to apply its latest PSU as well. When should I do this PSU patching ?
A. Should I apply the GI PSU after the entire Grid installation is successfully completed ?
or
B. Should I apply the GI PSU just before running the root.sh ?
i.e. when the installer prompts you to run root.sh (see the screenshot below)
[URL].........
At this stage(before running root.sh) the GI binaries are in place . So , I can apply the PSU on these binaries? Right?Should I follow option A or B above ?
I would like to use Oracle Grid Infrastructure to protect a single-instance database.
I found these docs:
[URL]
I was unable to find a similar document for GI 11.2 . Do you know if such a document exists?
At the moment I only know that the GI is free if used for an Oracle Product, but I am not shure whether it can be used to protext external applications or single-instance databases
I tried to install 11gR2 grid infrastructure for 2 nodes RAC using VMWare as per the instructions provided in the below link.
[URL]
Until 65% installation went fine without any issues but after reaching this step "updating '/u01/app/11.2.0/grid/install/cluster.ini",it starts to hung. In the installation log file, I found these details.
i try to install oracle grid infrastructure 11r2 on cent OS 4.8 on vmware workstation 8.0.2..ASM disk is not showing ,i had also change disk discovery path to 'ORCL:*'i had already configure ASM disk using /etc/init.d/oracleasm createdisk...successfully..
I need to take RMAN full backup by every Sunday night and Wednesday night . I have the netbackup script which will take the backup to media.
Question is:1. We have 15 - backup scripts for 15 DB in that server , so if i configure crontab for backup do i need to give all 15 scripts one by one or (*) will work . ie) 00 20 * * 3,7 /app/oracle/rman/scripts/hot_db_<sid>.sh
instead of this if i give : 00 20 * * 3,7 /app/oracle/rman/scripts/hot_db_*.sh then it will works fine or not ? 2.
Already my backup will create .output file in the same location , do i need to give output file location in crontab using ( > ) ? 3.
Is the above crontab timing is correct ? (3,7) for Wednesday and Sunday 8pm? so Sunday is 0(zero) or 7 ?
I am in a state of confusion and requires a solution for this urgently:I used to use 'Execute Immediate' statement or 'Sys_RefCursor' for forming parts of query dynamically based on user inputs and then finally execute them at runtime for updating database / saving results into a variable etc.
But as per the latest requirement, I wanted to have the resultset of such a query displayed in the grid area / results area of a database tool; say, SQL Navigator, Toad or PL/SQL developer.; just like you see the results of a normal 'SELECT * FROM TABLE1'; Is this really possible ??
see the below examples:
Eg: for Execute Immediate:
declare vtest VARCHAR2 := NULL; vdisplay VARCHAR2; begin vtest := 'SELECT FIELD_1 INTO vdisplay FROM TABLE_A'; vtest := trim(vtest) || 'WHERE FIELD_2 = ''XYZ'''; execute immediate TRIM(vtest); dbms_output.put_line(vdisplay); end;
---> This would display output into the 'Output' tab of the database tool / editor;
Eg: for SysRefCursor:
declare vCur SYS_REFCURSOR; vdisplay VARCHAR2; begin OPEN vCur for 'SELECT NULL' || var1 || 'WHERE FIELD_1 = ' || Value1; FETCH vCur INTO vdisplay; dbms_output.put_line(vdisplay); end;
---> Even this would display output only into the 'Output' tab of the database tool / editor;
===> But my requirement is: Different from these, is there any way to get this output displayed into the normal output grid / result area of a database tool ? just like you see the results of a normal 'SELECT * FROM TABLE1' query ?
after precompiling the cmpre.pc file, i got cmpre.c file. when i try compiling this .c file, am facing the following error...
servername:/.../home/usr/compile-> cc -g cmpre.c ld: 0711-317 ERROR: Undefined symbol: .sqlorat ld: 0711-317 ERROR: Undefined symbol: .sqlcxt ld: 0711-345 Use the -bloadmap or -bnoquiet option to obtain more information.
(Oracle Grid & Real Application Clusters Oracle Grid Computing with RAC) from rampant tech press. I have already ordered it 3 months back and still have not received it as its not printed yet or as i am told is in press. I would be glad to have a copy soon if its published.
CREATE OR REPLACE PROCEDURE A516907.proc_data_T_DTIA_TS_ADDR is cursor CSR1 IS SELECT * FROM T_DTIA_TS_ADDR ; v_file UTL_FILE.FILE_TYPE ; BEGIN v_file := UTL_FILE.FOPEN( 'DATA_PUMP_DIR' , 'T_DTIA_TS_ADDR.txt','w',32767); FOR CS1 IN CSR1
[Code]....
The procedure above is dynamically generated. It's simple enough in that I am calling "the parent procedure" by passing the table name in. After this the procedure for creating the output data file for that table is generated and that's what we have above. I am doing this to load data into DB2.
What I need is to be able to generate the output such that if the column is of type varchar, char (anything non numeric) then the data should have " around it. If the data is of type number it should not have ".
I suppose I could use something as follows in lieu of the procedure above
Declare cursor c1 is select column_name, data_type from user_tab_Cols where table_name = 'Table Name here';
-- I would like to know how to use the utl_file.put_line -- where if the column data type is varchar 2 -- I could get ' ||"|| CS1.TS_ADDR_TY_CD||" || ' -- and if it's a number -- || CS1.CB_TS_DTIS_ID ||
I am trying to get query output into textfile.The following procedure is working fine by creating directory as follows in sys user.The output is getting onto text file and it seen on server machine even.My question is I want to see the text file on local machine also instead of everytime connecting to server machine drive.How can I perform that?
create or replace directory INFO_DIR as 'D:Swapna'; grant read, write on directory INFO_DIR to fairpoint; CREATE OR REPLACE PROCEDURE FAIRPOINT.utl_file_test_write_xls(filename in VARCHAR2 ) IS output_fileutl_file.file_type; v_pathVARCHAR2(500); v_stringVARCHAR2(4000); v_sqlstrVARCHAR2(2000); BEGIN BEGIN [code]....
I use sqplus in oracle to output the command output to text file .
I use below set environment variable.
SQL> set echo off; SQL> set linesize 3999; SQL> set feedback off; SQL> set feedback off; SQL> set termout off; SQL> set pagesize 0; SQL> spool mapping.txt select C_SIM_MSISDN,C_SIM_IMSI from RCA_SMART_CARD order by C_SIM_MSISDN;
In ouput file , it look like
SQL> select C_SIM_MSISDN,C_SIM_IMSI from RCA_SMART_CARD order by C_SIM_MSISDN; 060010007 10007 : : : SQL> spool off;
any setting or command that allow me to remove the first sql command line" SQL>select XXXX" and the last command "SQL>spool off" after start up the "spool mapping.txt"
'HH2_SGH21_SS' is a schema I dont have access to. I have a whole bunch of schemas like this and would like to get versions of all those schemas putting it in a loop in pl/sql.
1: whats the best way get this output to a variable.
2: whats the best way to get the output to a file.
I'm running a sql file in Unix (using SQLPLUS command) and i want the file output to be sent to local windows directory. I specified the windows directory name in my Spool command (Spool C:/<Directory Name> ) , But it is not working
Our Unix server is a FTP server but i don't want to FTP the spool file from unix to Windows as the spool file is Huge and it takes hours for the transfer to complete (and we have to run the script multiple times).
Is there a way i can have the spool file created in Local windows directory when i run my sql script in Unix?
We are running a store procedure by calling it from shell script.The shell script is running in Server1 and the stored procedure is running in server2(which is the data base server.).
Functionality of the stored procedure:Generating a tab delimited file and place in a directory.
Requirement: The requirement is that the store should be modified in such a way that it should drop the output file to Server1 itself(The server from where it is getting called).
Attaching the store procedure for reference. mIn the store procedure,we have a variable "lv_file_dir" which currently refers to the directory location in Databse server(Server 2) We have to change this SP such that it drops the generated file in Server1 itself(From where the shell script is running).
Attached File(s) Procedure.txt ( 3.28K ) Number of downloads: 4
