During some tests that I was executing, I was often facing this issue, ORA-01466: unable to read data – table definition has changed, on data export with expdp from Oracle 19.16 in place. I am still investigating of the root cause, but as a workaround I have added the parameter CONTENT=ALL on expdp command line and it solved the issue, at least for now. The command I was executing: expdp mufalani/xxxxxx@myorclpdb directory=mufalani dumpfile=pdb1.dmp logfile=pdb1.log full=y flashback_time=systimestamp exclude=statistics When I checked the log of export the error showed for a particular table ..read more

A couple of days ago, I was playing a bit with encryption and I got a weird message when I was trying to encrypt a tablespace a second time, with a different algorithm. I will describe the situation and provide a workaround, while I’m still discussing with Oracle with an opened SR, about the issue. But I will provide a workaround as well. To develop a POC for a customer application, we have encrypted the database using TDE (Transparent Data Encryption) with the AES256 algorithm, if you are more interested to check all details regarding TDE, you can check the TDE official documentation here. Th ..read more

Some weeks ago, I was trying to upgrade a database from 12.2 to 19.16 with the AutoUpgrade tool, and I got an annoying error at the end of the upgrade. As per my standard behavior, I read the documentation and test the things before I ran it, and it was exactly what I did, even after using AutoUpgrade several times already. So, I would like to explain in this post, why you first need to “Stop blaming the software… Sometimes it’s you!!!” and RTFM (Read The Fucking Manual). First, as a best practice, I downloaded the latest version of AutoUpgrade tool, available at that time, I have placed it un ..read more

One of the things you need to care about is the TZ to be chosen while you are deploying an Oracle DB System, otherwise, you gonna suffer every time on patch the environment or restart, you end up having users complaining about the wrong time displayed by your queries. In this small post, I will show how to change the TZ configuration from GI and solve the issue. Even if the OS is correct, GI and the DB have their own configuration and on my tests, the GI was the reason for the wrong date and time display. To solve the issue, I followed the Doc ID 1627439.1 As root user, make a copy of the file ..read more

Last week, I was scripting something and I would like to replace an entire line on a init file, using sed, which I use a lot, but when you have “/” you need to use a bunch of escape characters and this is not cool. I will show in this small post, how to do it quickly and avoid mistakes by using sed in a proper way. Let’s imagine I have this Oracle init file below: *.audit_file_dest='/u01/app/oracle/db/base/admin/RMSRC/adump' *.audit_trail='db' *.compatible='19.0.0' *.control_files='/u01/app/oracle/oradata/RMSRC/controlfile/o1_mf_kfzycb9l_.ctl' *.db_block_size=8192 *.db_create_file_dest='/u01 ..read more

This week I was involved in an automation project that required a database to be named after a flash copy (storage level snapshot). So I need to write a code to use NID (DBNEWID Utility), but for the ones already familiar with the tool, it requests a confirmation to proceed or not with the renaming: Change database ID and database name OLDDBNAME to NEWDBNAME? (Y/[N]) => For other tools, Oracle has introduced environment variables or parameters, but for this particular tool (NID), I haven’t found a thing, like this, to skip the question. But to skip the question you can use the same approach ..read more

It is not uncommon to create new services on the database, in fact, it is a best practice for the level of HA of your application, for performance monitoring purposes and etc. But we are humans and we tend to commit mistakes. This was the case that I will present to you in this blog post, and it would be very nice, from Oracle if they implement the delete service by service_id: exec DBMS_SERVICE.DELETE_SERVICE( SERVICE_ID => :X); While this implementation doesn’t exist, if you run in the same situation as me, you can use the trick to solve the issue and remove the service created by mis ..read more

One of the main advantages of the Cloud Computing, is the ability to scale up and down some resources of your running environment without the long process of hardware purchasing, downtime to install it, and so on. During a test I was executing last week, I need to enlarge the boot volume of one machine running the Ubuntu Operating System. The process of resizing a disk on Oracle Cloud Infrastructure (OCI) is pretty much easy and straightforward. The documentation on Oracle is always a very good place to start, so I will leave it here Via OCI Console, to edit a block volume (Boot volume), you w ..read more

This error happens when your sysadmin makes the LUN on the OS side larger than 2TB, by increasing it, and then you need to “resync” to be aligned with the OS, but if you are using Oracle 12cR1 and newer versions, that is more likely to happen these days, you probably are locked by the wrong parameters inside the ASM instance. Let’s have a look: SQL> alter diskgroup data resize all; alter diskgroup data resize all; * ERROR at line 1: ORA-15032: not all alterations performed ORA-15099: disk '/dev/oracleasm/disks/ASM1' is larger than maximum size of 2097152 MBs   As per My Oracle Suppo ..read more

This error is caused by corrupted histograms on table stats and in this small blog post I will show how to remove and, workaround this issue. When I was checking the alert log from one instance I saw this error: ERROR at line 1: ORA-00600: internal error code, arguments: [16515], [U], [37], [10677], [1], [10], [], [], [], [], [], [] ORA-06512: on "SYS.DBMS_STATS", line 1704 ORA-06512: on "SYS.DBMS_STATS", line 14485 ORA-06512: on "SYS.DBMS_STATS", line 14615 ORA-06512: on "SYS.DBMS_STATS", line 35378 ORA-06512: on "SYS.DBMS_STATS", line 36230 ORA-06512: on "SYS.DBMS_STATS", line 36716 ORA-06 ..read more