Param ‘_datafile_write_errors_crash_instance’ , TRUE or FALSE?

Since there’s a new parameter, “_datafile_write_errors_crash_instance” to prevent the intance to crash when a write error on a datafile occurs .  But.. should I use this or not.  The official text of this parameter:

This fix introduces a notable change in behaviour in that
from onwards an I/O write error to a datafile will
now crash the instance.

Before this fix I/O errors to datafiles not in the system tablespace
offline the respective datafiles when the database is in archivelog mode.
This behavior is not always desirable. Some customers would prefer
that the instance crash due to a datafile write error.

By |August 26th, 2011|Categories: Database, RAC|Tags: , , |0 Comments

Restoring OCR on

As the documentation about restoring an OCR did not work as it should be, the following has been succesfully tested…

Log on as “root”
# . oraenv +ASM1
# ./ocrconfig -showbackup
# ./ocrconfig -showbackup manual

Stop the cluster on all nodes

# crsctl stop crs

By |May 15th, 2011|Categories: Database, RAC|Tags: , , |0 Comments

Just another RMAN backup-script for Windows

I was asked to deliver a backup-script for a windows-database – really ! However I’m not an o.s.-scripting guru at all and certainly not on Windows, I came to the following. It looks a lot like the one I use on Linux, just tried to translate it into a .cmd-file. The challenge was to get it in one-command file.
I noticed there  are a lot of RMAN backup-scripts, but not very much executed from a cmd-file:

By |May 16th, 2010|Categories: Database, RMAN|Tags: , |1 Comment

Apex : wwv_flow.accept, PLS-00306, wrong number of types of arguments

Database:, Apex 3.2.1. Application works fine, until there’s a bit of load of users logging in. They get a http-404 message. Full message derived from error-logging of Apache:

[Wed Mar 10 09:21:58 2010] [error]

  • [ecid: 1268209318:,0] mod_plsql: /pls/apex_app/wwv_flow.accept HTTP-404 ORA-06550: line 22, column 3: PLS-00306: wrong number or types of arguments in call to ‘ACCEPT’ ORA-06550: line 22, column 3: PLS-00306: wrong number or types of arguments in call to ‘ACCEPT’ ORA-06550: line 22, column 3:PL/SQL: Statement ignored

    When looking at the error-logging of Apex, there are a massive amount of the same errors.

    The package ‘www_flow.accept’ is a genuine Apex-package, so it doesn’t look like an application thing. After poking around on Metalink and forums, the following was found:

    Metalink: this could be the result of database bug 5228292 or 4752541 (fixed in introducing bug 5705795, also fixed in, also one-off patch available) and the problem went away after issuing ALTER SYSTEM FLUSH SHARED_POOL as SYS. So yes, this is a workaround, and yes I put this in a job every night. Did it help? Unfortunately not.

    In the forums it seems that not using ‘clearing caches’ in the sessions is causing some trouble, and one of the solutions (sorry, could not find the URL and the author anymore) we used, and I quote the author:

    I had the exact same problem when I tried to submit my login credintials. I got around it by adding a process to my login page.

    Process type: Clear Cache For Current Session (removes all state for current session)
    Process point: On Load – Before header.

    After I start using this process I haven’t seen the error.

    Implemented this (by our developer), and … now we wait. Don’t know of it helped yet, but it looks like an old story: not neatly programming will be punished, also by Apex.  Whether it helped or not with me, maybe it helps you to solve your problem or gives you a clue.

    By |March 13th, 2010|Categories: Database|Tags: , |0 Comments

    ORA-03114 / 03113 while running dbms_java package

    Short note this time. Got errors (ora-03114 / ora-03113)  while running any dbms_java script on a database. Took a while, but then I found out it only happens when I’m connected  through sql*net.  Directly on the server gave no problem at all. And… it only occurs in my test-environment, not in development, and not in production.

    What is different in my test-environment? Ahh…., on that node there are two different versions of the database installed, a and a And.. only one listener, from the – software tree.

    Stopped the listener of the, and started the listener of the (after configuring), and it worked again! The only thing is now that the other database has the problem of the dbms_java..   Configured two listeners, one at port 1521, the other on 1526, reconfigured tnsnames on the application server,  problem solved.

    The cause of this failure? I think it has to do with environment-variables, but I was too lazy to dig thoroughly, so any suggestions are welcome.

    For the record, the test script I used (it does nothing here, deleted a line with a call to a java program);

    set serveroutput on
    l_output dbms_output.chararr;
    l_lines number;
    dbms_output.get_lines(l_output, l_lines);
    for i in 1 .. l_lines loop
    end loop;


    ORA-03114: not connected to ORACLE

    ERROR at line 1:
    ORA-03113: end-of-file on communication channel

    By |March 5th, 2010|Categories: Database|Tags: |0 Comments