RE: RMAN - Some basic Qs.
Qs.4 Is it easy to integrate 3rd part tools like Legato etc with RMAN for enabling taking of backups onto backup devices OR does it need an extensive setup ? I found it to be fairly easy to get RMAN working with Legato NetWorker. We had to go through a few hoops to get it working on a Veritas Cluster, but the problems were related to implementing Legato NetWorker on the cluster rather than the RMAN/Legato interface. The setup and configuration is well documented. -Original Message- Sent: Tuesday, April 01, 2003 10:19 AM To: Multiple recipients of list ORACLE-L Qs.4 Is it easy to integrate 3rd part tools like Legato etc with RMAN for enabling taking of backups onto backup devices OR does it need an extensive setup ? -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Ed Bittel INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: veritas backup via RMAN troubles
Title: veritas backup via RMAN troubles I had a remarkably similar experience a few months ago with Legato NetWorker and performed all of the steps you listed with the same results. The problem turned out to be very simple. The SA installed the 64-bit version of the Legato Networker client because it is a 64-bit server. However, we were running a 32-bit version of Oracle on it. Installing the 32-bit client solved the problem. I am mentioning this since your post did not indicate if you are using 64-bit Oracle on that server. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Testa, Joe Sent: Wednesday, March 26, 2003 1:04 PM To: Multiple recipients of list ORACLE-L Subject: veritas backup via RMAN troubles AIX 5.1, oracle 9ir2(9.2.0.1), veritas 4.5 followed instructions for installing veritas on disk, got the agent, did NOT relink oracle executable, but ran the oracle install script so the link from /usr/openv/netbackup/bin/libobk.a64 is linked to $ORACLE_HOME/lib/libobk.a Set(and unset multiple times), LIBPATH, LD_LIBRARY_PATH. Continue to get ORA-27211 Failed to load media management library. Been thru RMAN chapters multiple times, I'm totally out of ideas(and read every damn document on metalstink about this error). I've also tried to use the dummy MML library and that works fine. What the heck am I doing wrong? thanks, joe
RMAN: delete archivelog... backed up 1 times to tape
I am trying to use the DELETE ARCHIVELOG command that is available in version 9.2.0.1. The complete syntax I wish to use is this command: RMANdelete archivelog until time 'trunc(sysdate)-2' backed up 1 times to tape; I have tried this as part of a backup script and as a standalone command after allocating a maintenance channel, but it doesn't work. Furthermore, this command DOES NOT return records: RMAN list archivelog all backed up 1 times to tape; But this command DOES return records: RMAN list archivelog all backed up 0 times to tape; I know I have backups of archivelogs on tape because they are reported when I issue this command: RMAN list backup of archivelog from time 'sysdate-3' device type sbt; Does anyone know what is going on here? Thanks, Ed / RDBMS Version: 9.2.0.1 OS: Sun Solaris 8 Product: RMAN (9.2.0.1) */ -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Ed Bittel INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: Skipping a table on import (Using FGAC)
Suggestion: Investigate using FGAC.
I just spent part of my day working on a solution to a very similar problem.
I wanted to do a user level export but restrict the records from one table
based on the value of a timestamp field. I did not want to have multiple
exports. I originally considered a combination of tablespace and table
level exports, but it made the nightly import process rather difficult to
perform the way I would like.
The solution I settled on was to use Oracle's fine-grained access control
(FGAC). This feature has been available since 8i. I found FGAC a bit
difficult to get my head around, but I had no problems after I worked
through the example. I got the idea from Metalink DOC_ID: 162914.1 which
explains how to use FGAC to skip a table during an export. I also referred
to the Application Developer's Guide documentation.
The effect of the FAGC I implemented is restrict which records the EXPORT
user can see in a table. This effects only the EXPORT user. All other
users see all of the records.
RECORD COUNT ISSUED RUN AS ME (I SEE ALL RECORDS)
SQL select count(*) from bigjet.message;
COUNT(*)
--
342559
Export run as user EXPORT (HE SEES FILTERED RECORDS)
exp export/freewilly@WHALEJET file=test.dmp tables=bigjet.message
About to export specified tables via Conventional Path ...
Current user changed to BIGJET
EXP-00079: Data in table MESSAGE is protected. Conventional path may only
be exporting partial table.
. . exporting tableMESSAGE 52445 rows exported
EXP-00091: Exporting questionable statistics.
Export terminated successfully with warnings.
The basic steps I followed are listed below.
/
CREATE Function to return predicate
used to filter records for the EXPORT user
/
CREATE OR REPLACE FUNCTION export.export_message(obj_schema varchar2,
obj_name varchar2)
RETURN VARCHAR2 IS d_predicate VARCHAR2(2000);
BEGIN
IF sys_context('USERENV','SESSION_USER')='EXPORT' THEN
d_predicate := 'SYS_LAST_CHANGED_TS TRUNC(SYSDATE-2) ';
ELSE
d_predicate := '';
END IF;
RETURN d_predicate;
END export_message;
/
/
CREATE FGAC Policy Group
/
EXECUTE DBMS_RLS.CREATE_POLICY_GROUP('BIGJET','MESSAGE','MESSAGE_GROUP');
/
CREATE FGAC Policy in the Policy Group
/
EXECUTE
DBMS_RLS.ADD_GROUPED_POLICY('BIGJET','MESSAGE','MESSAGE_GROUP','EXPORT_POL',
'EXPORT','EXPORT_MESSAGE');
/
ENABLE FGAC Policy
/
EXECUTE DBMS_RLS.ENABLE_GROUPED_POLICY
('BIGJET','MESSAGE','MESSAGE_GROUP','EXPORT_POL',TRUE);
/
DISABLE FGAC Policy
/
EXECUTE DBMS_RLS.ENABLE_GROUPED_POLICY
('BIGJET','MESSAGE','MESSAGE_GROUP','EXPORT_POL',FALSE);
-Original Message-
Is there a
way to skip
certain tables during an import?
--
Please see the official ORACLE-L FAQ: http://www.orafaq.net
--
Author: Stephen Lee
INET: [EMAIL PROTECTED]
--
Please see the official ORACLE-L FAQ: http://www.orafaq.net
--
Author: Ed Bittel
INET: [EMAIL PROTECTED]
Fat City Network Services-- 858-538-5051 http://www.fatcity.com
San Diego, California-- Mailing list and web hosting services
-
To REMOVE yourself from this mailing list, send an E-Mail message
to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L
(or the name of mailing list you want to be removed from). You may
also send the HELP command for other information (like subscribing).
Size of a Long Field
How do you determine the size, in bytes, of a long field? Please. No tape measure jokes. Ed
Veritas Quick IO Oracle Performance
I was asked to evaluate the potential benefit of using Veritas Quick I/O for one of our databases. I put together some simple tests to gauge the expected impact of Veritas Quick I/O on Oracle performance. The tests were designed so that the only variable when running the test scripts was type of datafile the scripts were reading from or writing to (i.e., Quick I/O or non-Quick I/O). The results of the tests were very surprising. Only the performance of large DDL operations, such as copying tables and creating indexes, was significantly improved after implementing Veritas Quick I/O for an Oracle datafile. Performance of DML operations ranged from marginally improved to markedly worse after implementing Veritas Quick I/O for an Oracle datafile. Large selects were particularly poor performing with Quick IO. Are these results typical? Are there any issue I should be aware of? We're running Oracle 8.1.6.3 on Solaris with Veritas Database Edition for Oracle 2.1.1. TIA, Ed __ Do You Yahoo!? Send FREE video emails in Yahoo! Mail! http://promo.yahoo.com/videomail/ -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Ed Bittel INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
Logminer Question
I'm analyzing archive logs from an 8.1.6.3 database running on Solaris. There are many entries in v$logmnr_contents with a NULL username and DATA_OBJ# and DATA_OBJD# values that don't match anything I can find in sys.obj$. What would account for these entries in v$logmnr_contents? Example === SELECT username, OPERATION, count(*) FROM v$logmnr_contents group by username, operation / USERNAME OPERATION COUNT(*) -- -- WANT_NUTS COMMIT 301 WANT_NUTS DELETE4 WANT_NUTS INSERT4 WANT_NUTS INTERNAL884 WANT_NUTS START 301 WANT_NUTS UPDATE 296 ZEDWARDS COMMIT3 ZEDWARDS INSERT 446 ZEDWARDS INTERNAL 15 ZEDWARDS START 4 OKSERVER1 COMMIT 20 OKSERVER1 DELETE 30 OKSERVER1 INSERT 296 OKSERVER1 INTERNAL269 OKSERVER1 START20 OKSERVER1 UPDATE 163 SYSCOMMIT 4034 SYSDELETE 291 SYSINSERT 71 SYSINTERNAL 6997 SYSSTART 4036 SYSUPDATE 4346 COMMIT7 DELETE 1235 INSERT 2242 INTERNAL843 START 2 UNSUPPORTED 266 UPDATE59002 37 rows selected. FYI, data from this database is being replicated via snapshot to a reporting database. Could that be responsible for the NULL username entries? Solaris 7 Oracle 8.1.6.3 __ Do You Yahoo!? Find the one for you at Yahoo! Personals http://personals.yahoo.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Ed Bittel INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
Where 1 = 1
I was monitoring a load test of my company's new application when I came across something interesting. One of the more monstrous queries had the following as part of its WHERE clause: 'AND 1 = 1' Has anyone seen this used before? What is it supposed to achieve? __ Do You Yahoo!? Make a great connection at Yahoo! Personals. http://personals.yahoo.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Ed Bittel INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
wait events v$filestat.readtim
Are the 'db file scattered read' and 'db file sequential read' waits directly related to the readtim values in v$filestat? For every 1/100th sec of v$filestat.readtim, should I see corresponding wait time reported for the 'db file scattered read' or 'db file sequential read'? If this is the case, where would I find the corresponding I/O value(s) for the 'direct path read' wait event? __ Do You Yahoo!? Make a great connection at Yahoo! Personals. http://personals.yahoo.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Ed Bittel INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
v$tempstat question
I've been trying to track down the source of high 'direct path read' and 'direct path write' waits. It appears the waits are related to locally managed temporary tablespaces. However, I'm puzzled by what is reported in v$tempstat. SQL select FILE#, READTIM, WRITETIM, AVGIOTIM, LSTIOTIM, MINIOTIM, MAXIORTM, MAXIOWTM from v$tempstat; FILE# READTIM WRITETIM AVGIOTIM LSTIOTIM MINIOTIM MAXIORTM MAXIOWTM - --- -- -- -- 1 3001 0 6393 78 2 2 252 443 636 0 1387188 How could the values for MAXIORTM (maximum time spent doing a single read) be substatially greater than those reported for READTIM (time spent doing all reads)? This situtation persists even after all of the users have logged out of the database, so it doesn't appear to be a case of the read having not yet completed. FYI: We're running Oracle 8.1.6.3 on Sun Solaris. -Ed __ Do You Yahoo!? Make a great connection at Yahoo! Personals. http://personals.yahoo.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Ed Bittel INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
Re: SP Non-Parse CPU Calculation IS Incorrect
I don't know if anyone else uses Statspack, but I couldn't help but notice the incorrect '% Non Parse CPU' result in the instance efficiency section of the report. Non-Parse CPU % calculation taken from STATSPACK 8.1.7: STATSREP.SQL: round(100*1-(:prscpu/:tcpu),2). It should be: round(100*(1-(:prscpu/:tcpu),2)). Here's an example using actual statspack values for 'CPU used by this session' and 'parse time cpu'. select round(100*1-(23255/234162),2) pctval from dual; PCTVAL -- 99.9 select round(100*(1-(23255/234162)),2) pctval from dual; PCTVAL -- 90.07 So instead of .10% of CPU being used to parse as reported by Statspack, the database is actually using 10% of CPU to parse. Using the statspack calculation, my database could be using 50% of CPU to parse and the report would still report a high non-parse CPU %. I guess it pays to remember My Dear Aunt Sally. __ Do You Yahoo!? Make a great connection at Yahoo! Personals. http://personals.yahoo.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Ed Bittel INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
Experience using Legato on a Veritas Cluster?
If anyone on the list has any experience using Legato NetWorker on a Veritas cluster and wouldn't mind responding to some questions, please send me a e-mail. My questions pertain to configuration of Legato on a Veritas cluster and some concerns I've developed about recoverability after a node failover. Thanks, Ed __ Do You Yahoo!? Make international calls for as low as $.04/minute with Yahoo! Messenger http://phonecard.yahoo.com/ -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Ed Bittel INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
Legato NetWorker RMAN: A troublesome pair?
The subject line may be overstating things a bit, but I'm pretty frustrated at the moment. Backups are working fine when I run them manually (i.e., from a command line or within the RMAN utility). The problem is trying to schedule the Oracle backups through the NetWorker Administration GUI. Scheduled filesystem backups work fine. Our attempts to schedule a working backup fail without generating any useful errors in /nsr/logs/messages or /nsr/logs/daemon.log. Basically, the logs just say that the savegroup failed. Setting the NSR_SB_DEBUG_FILE in the nsrnmo script hasn't helped because nothing is being written to the log file. The Legato knowledgebase, if you can call it that, has been of little use (other than to confirm that the output we have received isn't specific enough to debug the problem). What I've seen on MetaLink leads me to wonder if people who use the Legato NetWorker/RMAN actually schedule their backups through the Administration GUI or fall back to good 'ol reliable cron jobs. So, if you use Legato NetWorker for your Oracle backups: Do you schedule them through the Admin GUI or through cron? Do you know of anything we should check that may not have been covered in the NetWorker administration manual? If you've had similar problems, I love to hear what the problem was. TIA for your help. Ed = Ed Bittel, Oracle DBA Executive Jet Technology Services ~~ If you're in Ohio and work with Oracle why not join the Ohio Oracle Users Group? Visit OOUG on-line at: http://www.ooug.org ~~~ __ Do You Yahoo!? Get personalized email addresses from Yahoo! Mail http://personal.mail.yahoo.com/ -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Ed Bittel INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
