Thursday, April 22, 2021

Configuring OKV automation using REST APIs

This post will go through the process of creating a few simple scripts to automate OKV installation using the REST API capability of OKV.

NOTE: This information was provided by the product manager Peter Wahl who has a great blog on the features of OKV and some great "Ask Tom" sessions on OKV (and other security pieces),you can watch and learn more.

Step #1 Configure RESTful Services and download client tool

First you need to configure the OKV server for RESTful Services. The instructions can be found here. This is done by navigating to the System tab and clicking on RESTful Services.

It is recommended that you only enable the RESTful services when you are actively onboarding new endpoints.

This bring up the window below.

There are three things you want to do from this window.

Click on the "Enable" box to enable RESTful services
Download the okvrestcliepackage.zip which are the client utilities.
Save this setting to enable RESTful services.

Now that we have this file, we need to download it our client and start creating the scripts to automate this process.

I downloaded the zip file to my DB host to configure it. I unzipped it in /home/oracle/okv/rest

NOTE: you can also download it directly from the OKV hosts by using any of the commands below.

wget --no-check-certificate https://ip_address:5695/okvrestclipackage.zip  
curl -k https://ip_address:5695/okvrestclipackage.zip -o okvrestclipackage.zip 
curl -O -k https://ip_address:5695/okvrestclipackage.zip

Step #2 Create a user for the restapi steps

I create a new user in OKV called "restapi" and you can see the permissions below. Before using the new user, you must login and change the password because the password is expired by default when creating a new user.

Step #3 unzip and configure the client tool

I unzipped the client tool into my home directory on a DB server so I can put together the automation scripts. In my case I unzipped it into /home/oracle/okv/rest. This creates 3 sub directories. I am going to format the output using this command.

find . | sed -e "s/[^-][^\/]*\// |/g" -e "s/|\([^ ]\)/|-\1/"

Below is what the output looks like

.
 |-lib
 | |-okvrestcli.jar
 |-bin
 | |-okv.bat
 | |-okv
 |-conf
 | |-okvrestcli.ini
 | |-okvrestcli_logging.properties

Step #4 - Set the environment for the CLI

In order to configure OKV, I am going to need some variables set in my environment. I can do this manually, but in my case I decided to create a "setenv.sh" script that will set the variables and add the OKV script to my path to be executed. I also included the ability to pass a parameter (ORACLE_SID) so that you can use the script in a loop across multiple instances on the same host.

The 3 main variables I will be using are

OKV_RESTCLI_HOME - Location of the scripts that I am going to be installing. If I source the setenv.sh script, it will set the home to this location.

OKV_RESTCLI_CONFIG - Name of the configuration file that contains the rest CLI configuration.

OKV_HOME - Location to install OKV for the current instance. This location is $ORACLE_BASE/admin/${DB_UNIQUE_NAME}/okv_home. This follows the standard for ExaCC.

NOTE: If this is a NEW database

If you want to use these steps to configure OKV on a new database, you need to perform the following steps prior to executing "/ ./setenv.sh".

Add the new $ORACLE_SIDE for the host in the "/etc/oratab" file.
Create the directory "mkdir $ORACLE_BASE/admin/{DB Unique Name}"
Create the directory "mkdir -p $ORACLE_BASE/diag/rdbms/${DB Unique Name}/${ORACLE_SID}
Use ". oraenv" to set the environment to this $ORACLE_SID

 ###Ensure that the setenv file is sourced from the current directory to properly set variables


if [ "${BASH_SOURCE[0]}" != "${0}" ]
  then echo " "
  else echo " "
    echo "the setenv.sh file must be sourced with \". ./setenv.sh \""
   return
fi


FILE=`pwd`/setenv.sh
if [  -f "$FILE" ]
  then echo " "
  else echo " "
    echo "source this from RESTful install directory"
    return
fi



#############################################################################################
#
#  Check to see if a parameter is passed  Or ORACLE_SID is set. Otherwise exit
#
#############################################################################################



if [ $# -eq 0 ]
  then
      echo " "
      if [ -z ${ORACLE_SID} ]
         then
           echo "No \$ORACLE_SID set and nothing passed.  Exiting "
           echo " "
           return
      fi

    echo "No \$ORACLE_SID provided. Trying current \$ORACLE_SID value of : $ORACLE_SID"
    echo " "
  else

    echo "\$ORACLE_SID provided. Setting \$ORACLE_SID to  : $1"
    export ORACLE_SID=$1
    echo " "
fi



#############################################################################################
#
#  make sure SID is in the /etc/oratab so we can determine the
#
#############################################################################################


sid_found='false'
for sids in `cat /etc/oratab |  cut -d ":" -f 1`
  do
   if [ "$sids" = "${ORACLE_SID}" ]
      then sid_found='true'
   fi
  done

if [ ${sid_found} = "false" ]
  then
    echo "\$ORACLE_SID : ${ORACLE_SID} not found in /etc/oratab file. Exiting"
    return
fi


#############################################################################################
#
#  make sure SID  has exists under $ORACLE_BASE/admin to install software
#
#############################################################################################


 db_dir=`find ${ORACLE_BASE}/diag/rdbms -maxdepth 2 -name ${ORACLE_SID}`
 delimeter_cut=`echo $db_dir | sed -e 's/\(.\)/\1\n/g' | grep \/ | wc -l`
 DB_UNIQUE_NAME=`echo $db_dir | cut -d "/" -f $delimeter_cut `

 if [ -z ${DB_UNIQUE_NAME} ]
       then
        echo $db_dir
         echo "Invalid \$ORACLE_SID. Exiting script"
         echo " "
         echo " "
         return
 fi



#############################################################################################
#
#   set the environment
#
#############################################################################################

export ORAENV_ASK=NO
. oraenv
export ORAENV_ASK=YES



#############################################################################################
#
#  set the variables
#
#############################################################################################

 echo "create environment variables OKV_HOME, OKV_RESTCLI_HOME and OKC_RESTCLI_CONFIG  "


## set variables
export OKV_HOME=$ORACLE_BASE/admin/$DB_UNIQUE_NAME/okv_home
export OKV_RESTCLI_HOME=`pwd`
export OKV_RESTCLI_BIN=$OKV_RESTCLI_HOME/bin
export OKV_RESTCLI_CONFIG=$OKV_RESTCLI_HOME/conf/okvrestcli.ini

##add the OKV script to the path

     echo " "
     echo "\$OKV_HOME            :  $OKV_HOME "
     echo "\$OKV_RESTCLI_HOME    :  $OKV_RESTCLI_HOME "
     echo "\$OKV_RESTCLI_CONFIG  :  $OKV_RESTCLI_CONFIG "
     echo " "
     echo "Adding \$OKV_RESTCLI_BIN to the \$PATH  "
     echo " "
 if [[ "$PATH" =~ (^|:)"${OKV_RESTCLI_BIN}"(:|$) ]]
 then   echo " "
 else PATH=${OKV_RESTCLI_BIN}:$PATH
 fi


string=`which dcli`
echo $string
if  [[ $string == *"no dcli in"* ]]
  then export RAC_DATABASE="N"
  else export RAC_DATABASE="Y"
fi

## Set JAVA_HOME to the location of Java if it isn't set already

if [ -z ${JAVA_HOME+x} ]
  then export JAVA_HOME=$ORACLE_HOME/jdk
       echo "Setting  \$JAVA_HOME to \$ORACLE_HOME/jdk  : $JAVA_HOME"
  else
       echo "\$JAVA_HOME already set to   : $JAVA_HOME"
fi

     echo " "

Step #5 - Set initialization parameters in okvrestcli.ini file

Next, I am going to configure the initialization parameters. These are found in the okvrestcli.ini file.

You can see that the file contains a "[Default]" profile and a few other example profiles. We will start with the default profile. In this we are going to set a few of the properties.

LOG_PROPERTY - Location of the logging properties. Default location is ./conf directory.

SERVER - IP address (or DNS) of one or more OKV hosts

OKV_CLIENT_CONFIG - location of the config file. Default location is ./conf directory

USER - OKV user that has authority to administer endpoints an wallets. In this case it will be the restapi user that I created.

PASSWORD - Password for the user, or location of wallet containing the password. I am NOT going to use this as I am going to use a wallet file.

CLIENT_WALLET - I am going to use a wallet to store the password, and this is the location of the wallet file. I will be creating the autologin wallet later.

Below is what my "[Default]" configuration file looks like after my changes which is located at $OKV_RESTCLI_HOME/conf/okvrestcli.ini . I am going to use the environmental variables I set in the setenv.sh script.

NOTE: I am choosing to store my password in wallet rather than clear text in the .ini file.

You need to change the server to either the server name or the IP address.

[Default]
log_property=$OKV_RESTCLI_HOME/conf/okvrestcli_logging.properties
server=10.0.0.150
okv_client_config=$OKV_RESTCLI_HOME/conf/okvclient.ora
user=restapi
client_wallet=$OKV_RESTCLI_HOME/conf

Step #6 Create the wallet to save the password encrypted

Since I chose to put my password in a wallet, I now need to create that wallet. Using the instructions in the document (linked to at the beginning of this blog), I execute the command from the directory I installed into (/home/oracle/okv/rest)

cd /home/oracle/okv/rest
. ./setenv.sh

#####  Output from script is below #########
# create environment variables OKV_RESTCLI_HOME and OKC_RESTCLI_CONFIG
#
#   $OKV_RESTCLI_HOME    :  /home/oracle/okv/rest
#  $OKV_RESTCLI_CONFIG  :  /home/oracle/okv/rest/conf/okvrestcli.ini
#
#   Adding $OKV_RESTCLI_BIN to the $PATH</i>#



okv admin client-wallet add --client-wallet $OKV_RESTCLI_HOME/conf --wallet-user restapi
#Password: {my password}
#{
#  "result" : "Success"
#}

Step #7 Create and execute the run-me.sh script

The last step is to create the script that will be executed on the host to create the provision script. In my script, I took the default and did some checking. This script will

Ensure the variable OKV_RESTCLI_HOME is set before it can be executed.
Determine the DB_UNIQUE_NAME from the $ORACLE_BASE/diag/rdbms/*/$ORACLE_SID directory. Solving for the * should give us the DB_UNIQUE_NAME
While executing, it tells you what it believes the DB_UNIQUE_NAME is, and gives you a chance to change it if incorrect.
It will validate if the endpoint group exists by accessing OKV. If the endpoint group already exists, it does not try to create it again. If it doesn't exist it will create the endpoint group.
It will validate if the wallet exists by accessing OKV. If the wallet already exists, it does not try to create it again. If it doesn't exist it will create the wallet .
t will add the endpoint, add this endpoint to the endpoint group, and assign the wallet to this endpoint.
It will install the client software in $ORACLE_BASE/admin/$DB_UNIQUE_NAME/okv_home/okv

Below is the script I am using.

#!/bin/bash


###########################################################################################
#
#  Make sure the setenv script was executed to set the OKV_RESTCLI_HOME variable
#
###########################################################################################



if [ -v OKV_RESTCLI_HOME ]
 then echo " "
        echo "executing script with \$OKV_RESTCLI_HOME=$OKV_RESTCLI_HOME"
        echo " "
        echo " "
 else  echo " "
        echo "\$OKV_RESTCLI_HOME not set"
        echo " "
        echo "source the setenv.sh script in the RESTful Services install directory and try again."
        echo " "
        exit 0
fi


###########################################################################################
#
#  Determine the DBNAME Unique Name for this SID and have user validate
#
###########################################################################################

pushd $ORACLE_BASE/diag/rdbms/*/$ORACLE_SID > /dev/null
cd ..
export DB_UNIQUE_NAME=`printf '%s\n' "${PWD##*/}"`
popd > /dev/null


echo "DB Unique Name is identified as $DB_UNIQUE_NAME and ORACLE_SID is set to $ORACLE_SID setting"
echo " "
echo "****************************************************************************************** "
echo " "
echo "DB Name is identified as $DB_UNIQUE_NAME ---  NOTE THIS MAY NOT be the correct DB_NAME --- "
echo "      ---  NOTE THIS MAY NOT be the correct DB_NAME if there is a primary and standby  --- "
echo " "
echo "****************************************************************************************** "
echo " "
echo " "
echo "Press enter to keep this default [$DB_UNIQUE_NAME], or enter the DB Name"
read -p 'DB Name [enter for Default]    : '  newdbname

  
if [ -z "$newdbname" ]
  then export DBNAME=`echo $DB_UNIQUE_NAME`
  else export DBNAME=`echo $newdbname`
fi

echo " "
echo "Using DB Name : $DBNAME"

echo "#!/bin/bash" >  $OKV_RESTCLI_HOME/okv-install-${ORACLE_SID}.sh

###########################################################################################
#
#  check to see if wallet already exists
#  create it if it doesn't exist
#
###########################################################################################



export wallet_exists=`okv manage-access wallet check-status --wallet ${DBNAME^^} | grep Success | wc -l`
if [ $wallet_exists -ne 0 ]
  then  echo " Wallet ${DBNAME^^} already exists skipping wallet creation. assuming this is NOT the first node for the database"
  else
cat >> $OKV_RESTCLI_HOME/okv-install-${ORACLE_SID}.sh << EOF2
 okv manage-access wallet create --wallet ${DBNAME^^} --description "wallet for database ${DBNAME^^}"
EOF2
fi


###########################################################################################
#
#  check to see if endpoint Group already exists
#  create endpoint group if it doesn't exist
#
###########################################################################################


export group_exists=`okv manage-access endpoint-group check-status --endpoint-group ${DBNAME^^} | grep Success | wc -l`
if [ $group_exists -ne 0 ]
  then  echo " endpoint group ${DBNAME^^} already exists skipping group creation. assuming this is NOT the first node for the Database"
  else
cat >> $OKV_RESTCLI_HOME/okv-install-$ORACLE_SID.sh << EOF2
 okv manage-access endpoint-group create --endpoint-group ${DBNAME^^} --description "endpoint group for database ${DBNAME^^}"
EOF2
fi


export EP_NAME=`echo ${ORACLE_SID^^}_on_${HOSTNAME/.*} | sed 's/\-/\_/g'`


###########################################################################################
#
#  add commands to create/remove directories
#
###########################################################################################

cat >> $OKV_RESTCLI_HOME/okv-install-$ORACLE_SID.sh << EOF1
mkdir -pv ${OKV_HOME}
mkdir -pv ${OKV_HOME}/okv
mkdir -pv ${OKV_HOME}/tde
mkdir -pv ${OKV_HOME}/tde_seps
EOF1

cat > $OKV_RESTCLI_HOME/okv-remove-$ORACLE_SID.sh << EOF1
#!/bin/bash
rm -r ${OKV_HOME}
rm -r ${ORACLE_BASE}/okv/$ORACLE_SID
EOF1



###########################################################################################
#
#  commands to provision instance on Node
#
###########################################################################################

cat >> $OKV_RESTCLI_HOME/okv-install-${ORACLE_SID}.sh << EOF3
okv admin endpoint create --endpoint ${EP_NAME} --description "$HOSTNAME, $(hostname -i)" --type ORACLE_DB --platform LINUX64
okv manage-access endpoint-group add-endpoint --endpoint-group  ${DBNAME^^}   --endpoint ${EP_NAME}
okv manage-access wallet set-default --wallet ${DBNAME^^} --endpoint ${EP_NAME}
okv manage-access wallet add-access --wallet ${DBNAME^^} --endpoint-group  ${DBNAME^^} --access "RM_MW"
expect << _EOF
    set timeout 120
    spawn okv admin endpoint provision --endpoint ${EP_NAME} --location $OKV_HOME/okv --auto-login FALSE
    expect "Enter Oracle Key Vault endpoint password: "
    send "change-on-install\r"
    expect eof
_EOF
EOF3
chmod +x $OKV_RESTCLI_HOME/okv-install-$ORACLE_SID.sh
more $OKV_RESTCLI_HOME/okv-install-$ORACLE_SID.sh

Step #8 Execute okv-install-${ORACLE_SID}.sh to validate that enrollment works

[oracle@pmdexa1dbadm01vm01 okv]$ ./okv-ep.sh
{
  "result" : "Success"
}
{
  "result" : "Success"
}
{
  "result" : "Success"
}
spawn okv admin endpoint provision --endpoint DEMO31_on_pmdexa1dbadm01vm01 --location /u01/app/oracle/admin/demo31/wallet/okv --auto-login FALSE
Enter Oracle Key Vault endpoint password:
{
  "result" : "Success"
}

Once I ran it I removed the files that were created for the install.

Step #9 Zip it all up and place it in a location to be downloaded

Below is the scripts that will be part of the zip file.

.
 |-wallet
 | |-ewallet.p12.lck
 | |-ewallet.p12
 | |-cwallet.sso.lck
 | |-cwallet.sso
 |-conf
 | |-okvrestcli_logging.properties
 | |-okvrestcli.ini
 |-setenv.sh
 |-runme.sh
 |-lib
 | |-okvrestcli.jar
 |-bin
 | |-okv
 | |-okv.bat

Now I am ready to download this zip file to my Database Host and enroll a database.

NOTE: To change the script to work on another OKV all host I only had to make 3 changes.

Update the okvrestcli.ini file with OKV host IP
Update the okvrestcli.ini file with the the user
recreate the wallet file that contains the password for the OKV user

Step #10 Run root.sh if this is the first on a host

When a new endpoint is added, a script is downloaded into the $OKV_HOME/okv/bin/ directory called root.sh.. This script will copy the PKCS library (liborapkcs.so) in central location on the host as root. This file is needed for the database to access OKV. The location of this file is different on different OS systems.

NOTE: If you are already using a key management software on this host for Oracle databases, running root.sh could overwrite the library used.

Wednesday, April 7, 2021

Migrate your TDE wallet to Oracle Key Vault

How to migrate your local TDE wallet to Oracle Key Vault .

Here and here are the links to the 21C document that I used for to go through this process.

Also, in order to keep up on OKV and new features, the Product manager Peter Wahl has a great blog where you can learn about on the latest additions to the product, and new release information.

I am assuming that you installed the OKV server cluster by this point.

NOTE: If you are using the RestAPI method of installing the client, you can go straight to step #6. Be sure you have changed the password using the command below.

Below are the steps.

1) Add the database/host to OKV as an endpoint.

Remember in OKV, each endpoint is unique, but a wallet is shared between endpoints.

I navigate to the endpoint tab and click on the "Add" button.

I fill in the information for my database (TDETEST from my previous post). This is the CDB, as I am using a UNITED wallet for all PDBs that are a member of my CDB. Once filled in I click on the "Register" button.

Once registered, I can see it on the Endpoint screen. Note the "Enroll Token" column. This is needed to enroll the endpoint itself. Save this token, as this will be needed by the person who actually enrolls each DB host/Database.

2) Create wallet in OKV and associate it with the endpoint(s)

Now that OKV is installed and configured on the client we can create a wallet in OKV to upload the keys into. I am going to start by logging back into OKV and navigating to the wallets tab and clicking on "Create" to create a new wallet.

The screen belows comes up, and I enter the name of the new wallet to hold the keys for my CDB. I then click on save to save the new wallet.

3) Make this wallet the default wallet for the endpoint

Now go to the endpoints tab and click on the "endpoint name"

At this point you want to make this wallet the default wallet for the endpoint. Click on the "Choose Wallet" button, and it will bring up the window in front. Select the correct wallet and click on save. This will set the default wallet for this endpoint when it is enrolled in the next step.

4) Download the OKV client install file

Now that the database/host is registered in OKV (the combination of the 2 is the endpoint), I need to download the jar file which will configure the setting on the database host.

The download is initiated by logging out of the OKV console, and clicking on the "Endpoint Enrollment and Software Download" link on the logon screen. I highlighted it below.

You might not have noticed this link before ! Now click on the link, you don't need to login for this step. It will bring up the window below and in that window you will

Click on the "Submit Token" button, and it will validate the token
Click on "Enroll" to begin the download of the install file. If SMTP was configured, you can also have the software install e-mailed to the endpoint administrator.

The download file is a jar file called okvclient.jar. It is highly recommended that you rename it because it is specific to this endpoint.

5) Transfer the .jar file to the database host and install it.

The pre-requisites are in the install guide. The oracle environment during the install must be set to the database you are configuration ($ORACLE_HOME, $ORACLE_BASE, $ORACLE_SID)

My target directory is going to be "/home/oracle/app/oracle/admin/tdetest/wallet/okv" and I copied my .jar file to /home/oracle/app/oracle/admin/tdetest (which I renamed to tdetest_okv.jar).

Execute java passing the location of the jar file, followed by -d "install location"


[oracle@oracle-server okvtest]$ java -jar /home/oracle/app/oracle/admin/tdetest/tdetest_okv.jar -d /home/oracle/app/oracle/admin/tdetest/wallet/okv
Detected JAVA_HOME: /home/oracle/db_19c/jdk
Enter new Key Vault endpoint password (<enter> for auto-login):
Confirm new Key Vault endpoint password:
The endpoint software for Oracle Key Vault installed successfully.
Deleted the file : /home/oracle/app/oracle/okvfiles/okvtest/okvtest_install.jar
[oracle@oracle-server okvtest]$

If this is the first time OKV is being installed on the server, you need to execute the root.sh script (located in the /bin directory within the install location) as root. If it has already been executed, you can skip this step.

Creating directory: /opt/oracle/extapi/64/hsm/oracle/1.0.0/
Copying PKCS library to /opt/oracle/extapi/64/hsm/oracle/1.0.0/
Setting PKCS library file permissions

Finally, verify that we can connect OKV by executing "okvutil list". If successful, you will receive "default template". This script is located in the /bin directory within the install.

oracle@oracle-server bin]$ ./okvutil list
Enter Oracle Key Vault endpoint password:
Unique ID                               Type            Identifier
E8C825BB-649C-451E-98B8-E128B675FD31    Template        Default template for TDETEST_PRIMARY_NODE1

6) Review how OKV connects to the database.

WALLET_ROOT is set in the database, and within WALLET_ROOT there is an "/okv" directory where the endpoint software must be installed.
On the OS itself, a library is installed (as root if it's not already there) to take care of the encryption. The location on Linux is "/opt/oracle/extapi/64/hsm/oracle"
A link is created to a config file for this endpoint. This link is located in $ORACLE_BASE/okv/$DB_UNIQUE_NAME and links to 2 files that were part of the install. okvclient.lck, and okvclient.ora.

NOTE: okvclient.ora is the configuration file where parameters are set for the endpoint.

7) Upload the keys from the local wallet into OKV

Now we upload the keys from the local wallet into OKV.

The command is

"okvutil upload -t WALLET -l {wallet location on host} -g {key vault wallet name} -v 2

NOTE: the Key Vault wallet name is case sensitive

[oracle@oracle-server bin]$ ./okvutil upload -t WALLET -l  /home/oracle/app/oracle/admin/tdetest/wallet/tde -g tdetest -v 2
okvutil version 21.1.0.0.0
Endpoint type: Oracle Database
Configuration file: /home/oracle/app/oracle/admin/tdetest/wallet/okv/conf/okvclient.ora
Server: 10.0.0.150:5696
Standby Servers:
Uploading from /home/oracle/app/oracle/admin/tdetest/wallet/tde

Enter source wallet password:
Enter Oracle Key Vault endpoint password:
ORACLE.SECURITY.ID.ENCRYPTION.
ORACLE.SECURITY.KB.ENCRYPTION.
ORACLE.SECURITY.KT.ENCRYPTION.AQDBKozP1k8Mvwq4sH7ptKYAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
ORACLE.SECURITY.KM.ENCRYPTION.AQDBKozP1k8Mvwq4sH7ptKYAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
ORACLE.SECURITY.DB.ENCRYPTION.AQDBKozP1k8Mvwq4sH7ptKYAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
ORACLE.SECURITY.KT.ENCRYPTION.AYURdnq5XU8Rv7IipWqWgHoAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
ORACLE.SECURITY.KM.ENCRYPTION.AYURdnq5XU8Rv7IipWqWgHoAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
ORACLE.SECURITY.DB.ENCRYPTION.AYURdnq5XU8Rv7IipWqWgHoAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
ORACLE.SECURITY.DB.ENCRYPTION.MASTERKEY.BF507489CE7703B4E0536800000A8180
ORACLE.SECURITY.KM.ENCRYPTION.AXLqsppXAU9kv9JLJCcfGYAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
ORACLE.SECURITY.DB.ENCRYPTION.AXLqsppXAU9kv9JLJCcfGYAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
ORACLE.SECURITY.KT.ENCRYPTION.AXLqsppXAU9kv9JLJCcfGYAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
ORACLE.SECURITY.DB.ENCRYPTION.MASTERKEY.BF5072A8540A032BE0536800000AB0DD
ORACLE.SECURITY.KM.ENCRYPTION.AXDVlynThU8bvwblg7vruGAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
ORACLE.SECURITY.DB.ENCRYPTION.AXDVlynThU8bvwblg7vruGAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
ORACLE.SECURITY.KT.ENCRYPTION.AXDVlynThU8bvwblg7vruGAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
ORACLE.SECURITY.DB.ENCRYPTION.MASTERKEY.BF50708B8BEB0266E0536800000A7B90
ORACLE.SECURITY.DB.ENCRYPTION.MASTERKEY

Uploaded 4 TDE keys
Uploaded 0 SEPS entries
Uploaded 0 other secrets
Uploaded 6 opaque objects

Uploading private persona
Uploading certificate request
Uploading trust points

Uploaded 1 private keys
Uploaded 1 certificate requests
Uploaded 0 user certificates
Uploaded 0 trust points


Upload succeeded

Within the upload, I can see where the TDE master keys are being uploaded for my PDBs by looking at the PDB guids.

PDB PDB_ID

SQL> column name format a40
SQL> select name,guid from v$pdbs;

NAME                                     GUID
---------------------------------------- --------------------------------
PDB$SEED                                 BF5039AF39966A70E0536800000A09E1
TDEPDB1                                  BF50708B8BEB0266E0536800000A7B90
TDEPDB2                                  BF5072A8540A032BE0536800000AB0DD
TDEPDB3                                  BF507489CE7703B4E0536800000A8180

And I can look in the wallet in OKV (filtering by Symmetric key) and see the contents that was uploaded from the local wallet. In this screen I can identify the PDB key because I used tags when I created the keys.

8) Add secret to allow use of "External Store".

1) I am going to add the OKV password to the keystore as a secret to allow me to use the "EXTERNAL STORE" instead of the password.

ADMINISTER KEY MANAGEMENT ADD SECRET '0KV2021!' FOR CLIENT 'OKV_PASSWORD' TO LOCAL AUTO_LOGIN KEYSTORE '/home/oracle/app/oracle/admin/tdetest/wallet/tde_seps';

NOTE: As I pointed out in the last post

The keystore must be in a subdirectory of the WALLET_ROOT location called "tde_seps" in order to be found.
The "FOR CLIENT" entry must be 'OKV_PASSWORD' to be found.
It must be AUTO_LOGIN so that it can be opened and used.

2) I am going to add the OKV password to the keystore as a secret to allow me to auto logon to the OKV Keystore.

ADMINISTER KEY MANAGEMENT ADD SECRET '0KV2021!' FOR CLIENT 'HSM_PASSWORD' TO AUTO_LOGIN KEYSTORE '/home/oracle/app/oracle/admin/tdetest/wallet/tde';

3) I need to change the TDE_CONFIGURATION (which is dynamic).

'ALTER SYSTEM SET TDE_CONFIGURATION = "KEYSTORE_CONFIGURATION=OKV|FILE" SCOPE = BOTH;

4) I am going to bounce the database, and ensure that both my file and OKV wallets open up properly.


PDB Name   Type       WRL_PARAMETER                                      Status                         WALLET_TYPE          KEYSTORE Backed Up
---------- ---------- -------------------------------------------------- ------------------------------ -------------------- -------- ----------
CDB$ROOT   FILE       /home/oracle/app/oracle/admin/tdetest/wallet//tde/ OPEN                           AUTOLOGIN            NONE     YES
CDB$ROOT   OKV                                                           OPEN_NO_MASTER_KEY             OKV                  NONE     UNDEFINED
PDB$SEED   FILE                                                          OPEN                           AUTOLOGIN            UNITED   YES
PDB$SEED   OKV                                                           OPEN_NO_MASTER_KEY             OKV                  UNITED   UNDEFINED
TDEPDB1    FILE                                                          OPEN                           AUTOLOGIN            UNITED   YES
TDEPDB1    OKV                                                           OPEN_NO_MASTER_KEY             OKV                  UNITED   UNDEFINED
TDEPDB2    FILE                                                          OPEN                           AUTOLOGIN            UNITED   YES
TDEPDB2    OKV                                                           OPEN_NO_MASTER_KEY             OKV                  UNITED   UNDEFINED
TDEPDB3    FILE                                                          OPEN                           AUTOLOGIN            UNITED   YES
TDEPDB3    OKV                                                           OPEN_NO_MASTER_KEY             OKV                  UNITED   UNDEFINED


10 rows selected.

9) Combine the local wallet File and OKV.

Next I need to migrate the keys using the local wallet. Note this will rekey the database.

ADMINISTER KEY MANAGEMENT SET ENCRYPTION KEY IDENTIFIED BY "0KV2021!" MIGRATE USING "F1LE2021!" WITH BACKUP;

I am going to bounce the database and ensure it comes up with both Keystores opened.

PDB Name   Type       WRL_PARAMETER                                      Status                         WALLET_TYPE          KEYSTORE Backed Up
---------- ---------- -------------------------------------------------- ------------------------------ -------------------- -------- ----------
CDB$ROOT   FILE       /home/oracle/app/oracle/admin/tdetest/wallet//tde/ OPEN                           AUTOLOGIN            NONE     YES
CDB$ROOT   OKV                                                           OPEN                           OKV                  NONE     UNDEFINED
PDB$SEED   FILE                                                          CLOSED                         UNKNOWN              UNITED   UNDEFINED
PDB$SEED   OKV                                                           CLOSED                         UNKNOWN              UNITED   UNDEFINED
TDEPDB1    FILE                                                          OPEN                           AUTOLOGIN            UNITED   YES
TDEPDB1    OKV                                                           OPEN                           OKV                  UNITED   UNDEFINED
TDEPDB2    FILE                                                          OPEN                           AUTOLOGIN            UNITED   YES
TDEPDB2    OKV                                                           OPEN                           OKV                  UNITED   UNDEFINED
TDEPDB3    FILE                                                          OPEN                           AUTOLOGIN            UNITED   YES
TDEPDB3    OKV                                                           OPEN                           OKV                  UNITED   UNDEFINED

9) Combine the local wallet

That's all there is to it !

The most important notes I found during this process

WALLET_ROOT and TDE_CONFIGURATION should be used in 19c.
The password for OKV

add secret to the wallet in WALLET_ROOT/tde_seps using client 'OKV_PASSWORD'
add secret to the wallet in WALLET_ROOT/tde using client 'HSM_PASSWORD'

OKV must be installed in WALLET_ROOT/okv
The local wallet for TDE must be in WALLET_ROOT/tde

Monday, April 5, 2021

Oracle TDE encryption - Encrypting my pluggable database

This is post #1 in a series of posts explaining how to implement TDE (Transparent Data Encryption). In this first post I will take my Multitenant 19c database (remember Multitenant is mandatory with 21c) and configure TDE in my 3 (no extra license needed) pluggable databases.

The database I created for this example contains 3 PDBs as this will give me the flexibility to unplug and move PDBs around.

The names I used are

TDETEST - This is my CDB, and I will only be encrypting users tablespace and the catalog.
TDEPDB1,TDEPDB2, TDEPDB3 - My 3 PDBs. I will be encrypted all user tablespaces that make up these 3 PDBS .

Since I have only a single instance, the location I chose to put the local wallet file is under $ORACLE_BASE/admin/$ORACLE_SID/wallet (/home/oracle/app/oracle/admin/tdetest/admin).

In a RAC environment you have a few choices of where to put the wallet file. The most important thing though, is to ensure each node has an identical copy in a RAC environment.

In a RAC environment you can put the wallet file:

On the local file system on each node, but be sure they are all identical. It is best to put them within the $ORACLE_BASE to make it easier in an out-of-place upgrade.
On a shared filesystem. This can be ACFS, DBFS, or just NFS.
With ASM. It is possible to set the location to be an ASM file for example "+DATA/TDETEST/wallet"

NOTE: When deciding where to store your TDE wallet there are 2 items to keep in mind.

1) You need to backup your wallet. Without the wallet file you can't access your database. Once encryption is implemented, the wallet needs to be available.

2) The wallet needs to be backed up separate from the database. If the wallet and the Database backup are stored together, anyone can read the backup.

In my further blogs I will be converting from using a local wallet file to store my encryption keys, to using OKV along with a local wallet that caches the autologin information.

To migrate to TDE, I chose to perform the quickest method "Restore tablespace as encrypted". With my test database, I did not have a standby database.

NOTE: With a standby database the fastest way to convert to TDE would be with a "standby first" strategy. This is explained in this MAA document which includes an automated procedure. With this strategy you would convert your standby database to utilize TDE with a restore as encrypted, while the primary database stays untouched. Once the standby is converted, you would perform a switchover (to the standby) and encrypt the "current standby", which was the primary. Once both are encrypted, you would switch back and the process is completed.

Step 1 - Perform a full backup of the database immediately prior to starting the process. Since I am using "restore tablespace as encrypted" this will allow me to open the database with minimal recovery. Once backed up, you also should create a restore point to quickly identify the point after the full backup prior to the encryption.

create restore point pretde;

Step 2 - Set the location of the wallet_root, and the tde configuration. I chose to use the WALLET_ROOT parameter (new with 19 I believe) because it gives the most flexibility. Keep in mind in order to go through step 2 completely the database will need to be bounced.

alter system set WALLET_ROOT='/home/oracle/app/oracle/admin/tdetest/wallet/' scope=spfile;

startup force;

alter system set tde_configuration='KEYSTORE_CONFIGURATION=FILE' scope=both;

Step 3 - We are going to take a look at the database and the parameters that are set for the encryption wallet. Below is the formatted query I am going to be using throughout this post.

set linesize 170;
column wrl_parameter format a50
column wrl_type heading 'Type' format a10
column status heading 'Status' format a30
column fully_backed_up heading 'Backed Up' format a10
column pdb_name heading 'PDB Name' format a10

 select b.name pdb_name,wrl_type,
wrl_parameter,
status,wallet_type,
keystore_mode,
fully_backed_up
from v$encryption_wallet a,v$containers b
where a.con_id = b.con_id(+);

Below is the output of the query and the current settings as of this point. You can see that there are rows for all my PDBs, and that the status is "NOT_AVAILABLE" since I have not created a wallet or any master keys yet. You can also see that the keystore is UNITED, meaning that all the keys (both for the CDB and all the PDBs) are assumed to be contained in the same Wallet file.

Also note that the WRL_PARAMETER is set based on the WALLET_ROOT setting. The TDE wallet file is defaulted to be the WALLET_ROOT/tde directory for the CDB.

PDB Name   Type       WRL_PARAMETER                                      Status                         WALLET_TYPE          KEYSTORE Backed Up
---------- ---------- -------------------------------------------------- ------------------------------ -------------------- -------- ----------
CDB$ROOT   FILE       /home/oracle/app/oracle/admin/tdetest/wallet//tde/ NOT_AVAILABLE                  UNKNOWN              NONE     UNDEFINED
PDB$SEED   FILE                                                          NOT_AVAILABLE                  UNKNOWN              UNITED   UNDEFINED
TDEPDB1    FILE                                                          NOT_AVAILABLE                  UNKNOWN              UNITED   UNDEFINED
TDEPDB2    FILE                                                          NOT_AVAILABLE                  UNKNOWN              UNITED   UNDEFINED
TDEPDB3    FILE                                                          NOT_AVAILABLE                  UNKNOWN              UNITED   UNDEFINED

Step 4. Now I need to create the keystore and open it for the CDB, and all my individual PDBs. Note that each PDB shares the keystore with the CDB. In isolated mode, I would create an individual keystore for each PDB and they would be in subdirectories under the WALLET_ROOT location.

But first I need to create the directory to hold the keystore wallet.

mkdir /home/oracle/app/oracle/admin/tdetest/wallet/tde

ADMINISTER KEY MANAGEMENT CREATE KEYSTORE '/home/oracle/app/oracle/admin/tdetest/wallet/tde' IDENTIFIED BY "F1LE2021!";

ADMINISTER KEY MANAGEMENT SET KEYSTORE OPEN IDENTIFIED BY "F1LE2021!";
alter session set container=tdepdb1;
ADMINISTER KEY MANAGEMENT SET KEYSTORE OPEN IDENTIFIED BY "F1LE2021!" CONTAINER = CURRENT;
alter session set container=tdepdb2;
ADMINISTER KEY MANAGEMENT SET KEYSTORE OPEN IDENTIFIED BY "F1LE2021!"  CONTAINER = CURRENT;
alter session set container=tdepdb3;
ADMINISTER KEY MANAGEMENT SET KEYSTORE OPEN IDENTIFIED BY "F1LE2021!"  CONTAINER = CURRENT;

Now let's look at the encryption settings in v$encryption_wallet. Below you can see that there is a single wallet setting (UNITED keystore), and the status is "OPEN_NO_MASTER_KEY". The master key has not been set for CDB, or the PDBs.

PDB Name   Type       WRL_PARAMETER                                      Status                         WALLET_TYPE          KEYSTORE Backed Up
---------- ---------- -------------------------------------------------- ------------------------------ -------------------- -------- ----------
CDB$ROOT   FILE       /home/oracle/app/oracle/admin/tdetest/wallet//tde/ OPEN_NO_MASTER_KEY             PASSWORD             NONE     UNDEFINED
PDB$SEED   FILE                                                          CLOSED                         UNKNOWN              UNITED   UNDEFINED
TDEPDB1    FILE                                                          OPEN_NO_MASTER_KEY             PASSWORD             UNITED   UNDEFINED
TDEPDB2    FILE                                                          OPEN_NO_MASTER_KEY             PASSWORD             UNITED   UNDEFINED
TDEPDB3    FILE                                                          OPEN_NO_MASTER_KEY             PASSWORD             UNITED   UNDEFINED

Step 5. Now we create the master keys for the CDB and each PDB. A master key is needed to encrypt the tablespace encryption keys stored in the datafiles.

NOTE: I added a tag that identifies the key with the CDB or PDB it is created for. I highly recommend using tags to identify the keys within the wallet. Identifying the master encryption key for an individual PDB will be important when moving PDBs between CDBs.

ADMINISTER KEY MANAGEMENT SET encryption KEY using tag 'TDETEST MASTERKEY_APRIL19' IDENTIFIED BY "F1LE2021!" WITH BACKUP USING 'TDETEST_TDEKEY_APR1_backup';
alter session set container=tdepdb1;
ADMINISTER KEY MANAGEMENT SET encryption KEY using tag 'TDEPDB1 MASTERKEY_APRIL19' IDENTIFIED BY "F1LE2021!" WITH BACKUP USING 'TDEPDB1_TDEKEY_APR1_backup'  container=current;
alter session set container=tdepdb2;
ADMINISTER KEY MANAGEMENT SET encryption KEY using tag 'TDEPDB2 MASTERKEY_APRIL19' IDENTIFIED BY "F1LE2021!" WITH BACKUP USING 'TDEPDB2_TDEKEY_APR1_backup'  container=current;
alter session set container=tdepdb3;
ADMINISTER KEY MANAGEMENT SET encryption KEY using tag 'TDEPDB3 MASTERKEY_APRIL19' IDENTIFIED BY "F1LE2021!" WITH BACKUP USING 'TDEPDB3_TDEKEY_APR1_backup'  container=current;

And once again let's look at the settings in v$encryption_wallet. This time you will see that the wallet is open for all CDBs/PDBs except for the PDB$SEED. The wallet type is "PASSWORD" which means that the wallet needs to be manually opened with a password.

PDB Name   Type       WRL_PARAMETER                                      Status                         WALLET_TYPE          KEYSTORE Backed Up
---------- ---------- -------------------------------------------------- ------------------------------ -------------------- -------- ----------
CDB$ROOT   FILE       /home/oracle/app/oracle/admin/tdetest/wallet//tde/ OPEN                           PASSWORD             NONE     NO
PDB$SEED   FILE                                                          CLOSED                         UNKNOWN              UNITED   UNDEFINED
TDEPDB1    FILE                                                          OPEN                           PASSWORD             UNITED   NO
TDEPDB2    FILE                                                          OPEN                           PASSWORD             UNITED   NO
TDEPDB3    FILE                                                          OPEN                           PASSWORD             UNITED   NO

Step 6 - We have the master keys set and the wallets are open. We now need to implement TDE by encrypted the tablespaces in my PDBs. As I said before, in my example, I used "restore tablespace as encrypted".

Another option is to encrypt online (as of 12c). In this process the database will encrypt each datafile sequentially while the database is online and active.

NOTE : If using online encryption be aware that

It takes much longer than performing a restore, as datafiles are encrypted sequentially. Using "restore tablespace as encrypted" You can parallelize the restore across multiple channels.
The process needs enough space for 2 copies of the largest datafile. If using bigfiles, this can be quite a bit of storage.
You need to monitor the process to ensure it completes successfully.

Next step is I am going to startup mount and open the wallets, restore my 3 PDBs users tablespaces, along with the users tablespace in my CDB, and then recover and open the database.


sqlplus> shutdown immediate;
sqlplus> startup mount;
sqlplus> ADMINISTER KEY MANAGEMENT SET KEYSTORE OPEN IDENTIFIED BY "F1LE2021!";
sqlplus> alter session set container=tdepdb1;
sqlplus> ADMINISTER KEY MANAGEMENT SET KEYSTORE OPEN IDENTIFIED BY "F1LE2021!" CONTAINER = CURRENT;
sqlplus> alter session set container=tdepdb2;
sqlplus> ADMINISTER KEY MANAGEMENT SET KEYSTORE OPEN IDENTIFIED BY "F1LE2021!"  CONTAINER = CURRENT;
sqlplus> alter session set container=tdepdb3;
sqlplus> SDMINISTER KEY MANAGEMENT SET KEYSTORE OPEN IDENTIFIED BY "F1LE2021!"  CONTAINER = CURRENT;

rman target / catalog rmancat/oracle@rmancat

rman> restore tablespace users as encrypted;

rman> restore tablespace tdepdb1:users  as encrypted;
rman> restore tablespace tdepdb2:users  as encrypted;
rman> restore tablespace tdepdb3:users  as encrypted;

rman> recover database;
rman> alter database open;

Step 7 - Make sure all new tablespaces are encrypted by default

In order to ensure all new tablespaces are encrypted I am going to set the database parameter.

sql> alter system set encrypt_new_tablespaces = ALWAYS scope = both sid = '*';

Step 8 - Encrypt all credentials that contained in the root container

In order to encrypt all credentials (like scheduler credentials, and DB Link credentials) that are stored in the system catalogs, you need to login as a user granted "SYSKM" role and execute

sql> alter database dictionary encrypt credentials container = current;

Step 9 - I am going to verify that the pluggable databases are encrypted, along with the catalog.

First let's look at the existing keys using the query below

 set linesize 170;
SQL> column wrl_parameter format a50
column wrl_type heading 'Type' format a10
column status heading 'Status' format a30
column fully_backed_up heading 'Backed Up' format a10
column pdb_name heading 'PDB Name' format a10

 select b.name pdb_name,wrl_type,
wrl_parameter,
status,wallet_type,
keystore_mode,
fully_backed_up
from v$encryption_wallet a,v$containers b
where a.con_id = b.con_id(+);

I can see the keys that are created in each container, including the ROOT

PDB Name        Activation Time            Key ID                                                  Tag
--------------- -------------------------- ------------------------------------------------------- ----------------------------------------
CDB$ROOT        19-Apr-2021 05:12:41pm     AbwcWGicr0+rvyhrSB+rKQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA    TDETEST MASTERKEY_APRIL19

TDEPDB1         19-Apr-2021 05:19:11pm     AX9pkB+zQE/Wv6Qek13TeLkAAAAAAAAAAAAAAAAAAAAAAAAAAAAA    TDEPDB1 MASTERKEY_APRIL19

TDEPDB2         19-Apr-2021 05:19:11pm     AUKkROD1TE8wv0jfJhN63JYAAAAAAAAAAAAAAAAAAAAAAAAAAAAA    TDEPDB2 MASTERKEY_APRIL19

TDEPDB3         19-Apr-2021 05:19:13pm     AZUWZDWpxk9sv0GrljDFr7sAAAAAAAAAAAAAAAAAAAAAAAAAAAAA    TDEPDB3 MASTERKEY_APRIL19

I am going to use the query below to look at the encryption setting on each datafile.

set linesize 150
set pagesize 50
column pdb_name heading "PDB Name" format a15
column tablespace_name heading "Tablespace Name" format a30
column masterkeyid_base64 heading "Master Key ID" format a60
column encrypted heading "Encrypted" format a10

break on pdb_name skip 1

select pdb_name, 
       tablespace_name,encrypted,
       utl_raw.cast_to_varchar2( utl_encode.base64_encode('01'||substr(mkeyid,1,4))) || utl_raw.cast_to_varchar2( utl_encode.base64_encode(substr(mkeyid,5,length(mkeyid)))) masterkeyid_base64
 FROM (select t.name tablespace_name,
              z.name pdb_name,
              t.con_id, encrypted,
              RAWTOHEX(x.mkid) mkeyid 
        from v$tablespace t, 
             x$kcbtek x,
             v$containers z,
             cdb_tablespaces q 
        where t.ts#=x.ts# 
          and t.con_id=x.con_id 
          and t.con_id=z.con_id
          and t.con_id = q.con_id
          and t.name = q.tablespace_name)
order by pdb_name,tablespace_name;

Below is the output. I see that all the datafiles were properly encrypted and are available.

PDB Name        Tablespace Name                Encrypted  Master Key ID
--------------- ------------------------------ ---------- ------------------------------------------------------------
CDB$ROOT        SYSAUX                         NO         AbwcWGicr0+rvyhrSB+rKQA=
                SYSTEM                         NO         AbwcWGicr0+rvyhrSB+rKQA=
                TEMP                           NO         AbwcWGicr0+rvyhrSB+rKQA=
                UNDOTBS1                       NO         AbwcWGicr0+rvyhrSB+rKQA=
                USERS                          YES        AbwcWGicr0+rvyhrSB+rKQA=

TDEPDB1         SYSAUX                         NO         AX9pkB+zQE/Wv6Qek13TeLk=
                SYSTEM                         NO         AX9pkB+zQE/Wv6Qek13TeLk=
                TEMP                           NO         AX9pkB+zQE/Wv6Qek13TeLk=
                UNDOTBS1                       YES        AX9pkB+zQE/Wv6Qek13TeLk=
                USERS                          YES        AX9pkB+zQE/Wv6Qek13TeLk=

TDEPDB2         SYSAUX                         NO         AUKkROD1TE8wv0jfJhN63JY=
                SYSTEM                         NO         AUKkROD1TE8wv0jfJhN63JY=
                TEMP                           NO         AUKkROD1TE8wv0jfJhN63JY=
                UNDOTBS1                       YES        AUKkROD1TE8wv0jfJhN63JY=
                USERS                          YES        AUKkROD1TE8wv0jfJhN63JY=

TDEPDB3         SYSAUX                         NO         AZUWZDWpxk9sv0GrljDFr7s=
                SYSTEM                         NO         AZUWZDWpxk9sv0GrljDFr7s=
                TEMP                           NO         AZUWZDWpxk9sv0GrljDFr7s=
                UNDOTBS1                       YES        AZUWZDWpxk9sv0GrljDFr7s=
                USERS                          YES        AZUWZDWpxk9sv0GrljDFr7s=

And I am going to look at the catalog to be sure

select enforcement from DICTIONARY_CREDENTIALS_ENCRYPT;

ENFORCEM
--------
ENABLED

Step 10 - I am adding the ability to access the keystore without having to enter the password. This can make it much easier to open the keystore if you don't chose to make the Keystore Wallet AUTO_LOGON.

I am putting my keystore password in an AUTO_LOGIN wallet as a secret. This wallet needs to be created in a directory called TDE_SEPS. In my case since, I am using WALLET_ROOT, the directory containing the secret must be WALLET_ROOT/tde_seps.

The password portion of "ADMINISTER KEY" becomes IDENTIFIED by EXTERNAL STORE;

NOTE: I am using the phrase "FOR CLIENT 'TDE_WALLET'"

'TDE_WALLET' is necessary to

ADMINISTER KEY MANAGEMENT
     ADD SECRET 'F1LE2021!' FOR CLIENT 'TDE_WALLET'
     USING TAG 'TDE file keystore password' 
     TO LOCAL AUTO_LOGIN KEYSTORE '/home/oracle/app/oracle/admin/tdetest/wallet/tde_seps';

Now to verify it I am going to restart my database and open the wallet using the external store (rather than hardcoding in the password).

ADMINISTER KEY MANAGEMENT SET KEYSTORE OPEN IDENTIFIED BY EXTERNAL STORE;

alter session set container=tdepdb1;
ADMINISTER KEY MANAGEMENT SET KEYSTORE OPEN IDENTIFIED BY EXTERNAL STORE CONTAINER = CURRENT;
alter pluggable database  open;

alter session set container=tdepdb2;
ADMINISTER KEY MANAGEMENT SET KEYSTORE OPEN IDENTIFIED BY EXTERNAL STORE  CONTAINER = CURRENT;
alter pluggable database  open;

alter session set container=tdepdb3;
ADMINISTER KEY MANAGEMENT SET KEYSTORE OPEN IDENTIFIED BY EXTERNAL STORE CONTAINER = CURRENT;
alter pluggable database  open;

And finally verify that that the Keystores are open (using the external store) for my CDB and PDBs.

PDB Name   Type       WRL_PARAMETER                                      Status                         WALLET_TYPE          KEYSTORE Backed Up
---------- ---------- -------------------------------------------------- ------------------------------ -------------------- -------- ----------
CDB$ROOT   FILE       /home/oracle/app/oracle/admin/tdetest/wallet//tde/ OPEN                           PASSWORD             NONE     NO
PDB$SEED   FILE                                                          CLOSED                         UNKNOWN              UNITED   UNDEFINED
TDEPDB1    FILE                                                          OPEN                           PASSWORD             UNITED   NO
TDEPDB2    FILE                                                          OPEN                           PASSWORD             UNITED   NO
TDEPDB3    FILE                                                          OPEN                           PASSWORD             UNITED   NO

Step 11 - I am going to change the wallets to be AUTO_LOGIN, bounce the database and verify that the encryption settings are all correct.

sqlplus / as sysdba

sql> ADMINISTER KEY MANAGEMENT SET KEYSTORE CLOSE IDENTIFIED BY EXTERNAL STORE;
sql> ADMINISTER KEY MANAGEMENT CREATE AUTO_LOGIN KEYSTORE FROM KEYSTORE '/home/oracle/app/oracle/admin/tdetest/wallet/tde/' IDENTIFIED BY "F1LE2021!";


sql> shutdown immediate
sql> startup

And v$encryption_wallet shows me that my wallets are all open, and that they are AUTOLOGIN.

PDB Name   Type       WRL_PARAMETER                                      Status                         WALLET_TYPE          KEYSTORE Backed Up
---------- ---------- -------------------------------------------------- ------------------------------ -------------------- -------- ----------
CDB$ROOT   FILE       /home/oracle/app/oracle/admin/tdetest/wallet//tde/ OPEN                           AUTOLOGIN            NONE     NO
PDB$SEED   FILE                                                          OPEN                           AUTOLOGIN            UNITED   NO
TDEPDB1    FILE                                                          OPEN                           AUTOLOGIN            UNITED   NO
TDEPDB2    FILE                                                          OPEN                           AUTOLOGIN            UNITED   NO
TDEPDB3    FILE                                                          OPEN                           AUTOLOGIN            UNITED   NO

Now I am ready to perform a new FULL backup of the pluggable databases, and they are ready for use.

That's all there is to implementing TDE with a wallet file. Next post, I am going to convert my wallet to OKV managed wallets.

Monday, March 29, 2021

ZDRLA adds smart incremental to be even smarter.

Recently version 19.1.1.2 of ZDLRA software was released, and one the features is something called "Smart Incremental". I will walk through how this feature works, and help you understand why features like this are "ZDLRA Only".

I am going to start by walking through how incremental backups become "virtual full backups", and that will give you a better picture of how "smart incremental" is possible.

The most important thing to understand about these features is that the RMAN catalog itself is within the ZDLRA AND the ZDLRA has the ability to update the RMAN catalog.

How does a normal backup strategy work ?

That is probably the best place to start. What DBAs typically do is perform a WFDI (Weekly Full Daily Incremental) backup. To keep my example simple, I will use the following assumptions.

My database contains 3 datafile database. SYSTEM, SYSAUX, USERS, but I will only use the example of backing up datafile users.
Each of these 3 datafiles are 50 GB in size
I am only performing a differential backup which creates a backup containing the changes since the last backup (full OR incremental).
My database is in archivelog *

* NOTE: With ZDLRA you can back up a nologging database, and still take advantage of virtual fulls. The database needs to be in a MOUNTED state when performing the incremental backup.

If placed in a table the backups for datafile USERS would look this. Checkpoint SCN is the current SCN number of the database at the start of the backup.

If I were to look at what is contained in the RMAN catalog (RC_BACKUP_DATAFILE), I would see the same backup information but I would see the SCN information 2 columns.

Incremental change # is the oldest SCN contained in the backupset. This is the starting SCN number of the previous backup, this backup is based on.
Checkpoint Change # is starting SCN number of the backup. Everything newer than this SCN (including this SCN) needs to be defuzzied.

Normal backup progression (differential)

When performing an incremental RMAN backup of a datafile, the first thing that RMAN does is decide which blocks needs to be backed up. Because you are performing an incremental backup, you may be backing up all of the blocks, only some of the blocks, or even none of the blocks if the file has not changed.

This is a decision RMAN makes by querying the RMAN catalog entries (or the controlfile entries if you not using an RMAN catalog).

Now let's walk through this decision process. Each RMAN incremental differential's starting SCN is based on the beginning SCN of the previous backup (except for the full).

By looking at the RMAN catalog (or controlfile), RMAN determines which blocks need to be contained in each incremental backup.

Normal backup progression (cumulative differential)

Up to release 19.1.1.2, the recommendation was to perform a Cumulative Differential backup. The cumulative differential backup compares the starting SCN number of the last full backup to determine the starting point of the incremental backup (rather than the last incremental backup) .

The advantage of the cumulative over differential, is that a cumulative backups can be applied to the last full and take the place of applying multiple differential backups. However, cumulative backups are bigger every day that passes between full backups because they contain all blocks since the last full.

Below is what a cumulative schedule would look like and you can compare this to the differential above.

You can see that each cumulative backups starts with the Checkpoint SCN of the last full to ensure that all blocks changed since the full backup started are captured.

The RMAN catalog entries would look like this.

If you were astute, you would notice a few things happened with the cumulative differential vs the differential.

The backup size got bigger every day
The time it took to perform the incremental backup got longer
The range of SCNs contained in the incremental is larger for a cumulative backup.

ZDLRA backup progression (cumulative differential)

As you most likely know, one the most important features of the ZDLRA is the ability to create a "virtual full" from an incremental backup.,

If we look at what happens with a cumulative differential (from above), I will fill in the virtual full RMAN catalog entries by shading them light green.

The process of performing backups on the ZDLRA is exactly the same as it is for the above cumulative, but the RMAN catalog looks like this.

What you will noticed by looking at this compared to the normal cumulative process that

For every cumulative incremental backup there is a matching virtual full backup The Virtual full backup appears (from the newly inserted catalog entry) to have beeen taken at the same time, and the same starting SCN number as the cumulative incremental. Virtual full backups, and incremental backups match time, and SCN as catalog entries.
The size of the virtual full is 0 since it is virtual and does not take up any space.
The completion time for the cumulative incremental backup is the same as the differential backups. Because the RMAN logic can see the virtual full entry in the catalog, it executes the cumulative incremental EXACTLY as if it is the first differential incremental following a full backup.

Smart Incremental backups -

Now all of this led us to smart incremental backups. Sometimes the cumulative backup process doesn't work quite right. A few of the reasons this can happen are.

You perform a full backup to a backup location other than the ZDLRA. This could be because you are backing up to the ZDLRA for the first time replacing a current backup strategy, or maybe you created a special backup to disk to seed a test environment (Using a keep backup for this will alleviate this issue). The cumulative incremental backup will compare against the last full regardless of where it was taken (there is exceptions if you always use tags to compare).

You implement TDE or rekey the database. Implementing TDE (Transparent Data Encryption) changes the blocks, but does not change the SCN numbers of the blocks. A new full backup is required.

Previously, you would have to perform a special full backup to correct these issues. In the example below you can see what happens (without smart incremental) to the RMAN catalog if you perform a backup on Thursday at 12:00 to disk to refresh a development environment.

Since the cumulative backups are based on the last full backup, the Thursday - Saturday backups contain all the blocks that have changed since the disk backup started on Thursday at 12:00.

And, since it is cumulative, each days backup is larger, and takes longer.

This is when you would typically have to force a new level 0 backup of the datafile.

What the smart incremental does

Since the RMAN catalog is controlled by the ZDLRA it can correct the problem for you. You no longer need to perform cumulative backups as the ZDLRA can fill in any issues that occur.

In the case of the Full backup to disk, it can "hide" that entry, and continue to correctly perform differential backups. It would "hide" the disk backup that occured, and inform the RMAN client that the last full backup as of Thursday night is NOT the disk backup, but it is the previous virtual full backup.

In the case of the TDE, it can "hide" all of the Level 0 virtual full backups, and the L1 differential backups (which will force a new level 0).

All of this is done without updating the DB client version. All the magic is done within the RMAN catalog on the ZDLRA.

Now isn't that smart ?

Friday, March 26, 2021

ZDLRA leverages the ZFS object store in the newest release

Yes, Using ZFSSA as an on-prem object store with ZDLRA is here, and How to configure Zero Data Loss Recovery Appliance to use ZFS OCI Object Storage as a cloud repository (Doc ID 2761114.1) shows you how.

Above is the diagram from Tim Chien's "Ask Tom" session on the new feature with ZDRA release 19.1.1.2.

For those how have been reading my blog posts, and wondering why the sudden interest in ZFS as an object store, here is another reason.

The idea behind this is pretty simple, many customers are looking for an additional tier of storage behind the ZDLRA for 2 reasons

They want to extend the the recovery window onto a lower tier of storage. This may include going from a full "any point in time" recovery to a set of "recovery points"
They want an archival backup for a long period of time that is a set backup point. Keep backups are the perfect example of this. With Keep backups you get a self-contained restore point of your choosing.

Now for the magic of how all this works.

1. The first step is to configure your ZFSSA as an OCI object store. As long you are on the latest patched release of OS 8.8, this functionality is available to you. If you are unfamiliar with how to do this, in previous posts, I have walked through the steps of configuring this. Below are some places to start.

How to configure ZFS as an object store

Managing authentication with the ZFS object store

Also, here is the documentation from ZFS.

2. The second step is to configure Key Vault (OKV), which is a licensed product. Key vault is a centralized Encryption Key management system that is used to store the master encryption key for the backups. OKV is released as a virtual image, that can be installed on physical hardware, or in a virtual environment. the installation is self-contained and walks through a series of questions to finish the configuration. Easy.

WHY do I need TDE ? I'm sure you are asking this question. The "Copy-to-cloud" functionality of the ZDLRA is being utilized to present ZFS as an "OCI cloud store". It acts just like an object store in the Oracle Public Cloud. The only difference is that there is no "ARCHIVE" tier on ZFS. Since ZFS is considered a "Cloud destination", it follows the Larry rule that "All data in the Cloud is encrypted.". Because of that, the backups going to ZFS will be RMAN encrypted (no license needed for this part). The ZDLRA uses OKV to store the master keys used to encrypt the RMAN backupsets.

3. The third step is to configure the ZDLRA to utilize OKV as a client, and to point the ZDLRA to your ZFS.

One of the great things of using this solution is that the process is exactly the same as configuring the ZDLRA to send backups to the Oracle public cloud. This link points to the documentation that makes it clear how to configure this process.

That's all there is to it. The most complicated task is configuring the authentication for the OCI object store on ZFS, as it requires setting up a public and private key.

Now to walk through the workflow.

Backups -- Below is the backup workflow from the presentation. The ZDLRA creates an RMAN backupset from the backup pieces on the ZDLRA. This backupset is an RMAN encrypted backupset.

One item is NOT mentioned on this slide is compression. If your Database is using TDE, then the backup cannot be compressed when sent to the ZFS because the ZDLRA does not have the encryption master key for the database. BUT, if your database is NOT TDE enabled, then you should be using compression when sending the backups to the ZFS. As I've said earlier, the backset is an RMAN encrypted backupset. Because it is already encrypted when sent to ZFS, the ZFS will be unable to compress the backups. You can find instructions to add compression in the documentation for creating a job template. There is a setting for the template called

Compression_algorithm=>

By implementing compression on the ZDLRA you are:

Decreasing the size of the backups on the ZFS..
Decreasing the networkwork traffic between the ZDLRA and ZFS as the data is compressed before it is sent to ZFS. This can double the throughput for backups and restores.

Keep in mind, that if you restore directly to your database host from the ZFS Object store, the database host will be performing the uncompression.

Restores - Below is the restore workflow. Typically you would utilize the catalog on the ZDLRA and let the ZDLRA be the conduit for uncompressing (if it was compressed when sent to ZFS), and unencrypting it, as the ZDLRA encrypted it. The ZDLRA already has the credentials for the object store, and it has the Encryption master key available to it from OKV.

Alternately you can restore the backups directly from the ZFS object store.

This would be a 3 step process..

1) You would download the Oracle Database Cloud Backup Module . Once downloaded you would configure the database to utilize the OCI object store. The link above also contains links to documentation for the module, and to a MOS note containing the FAQ. Keep in mind that in this case you are configuring the Module for the on-premise ZFS (rather than the Oracle public cloud), and the instructions may have to be modified. The table below gives you an idea of the differences.

2) You would catalog the backup pieces. If the RMAN catalog is not available (for some reason) the MOS note mentioned below contains detail on how to list what is in the object store, and how to clean it out.

How to report or delete backup pieces stored in Cloud Object Storage by Database Backup Cloud Service without using RMAN (Doc ID 2360800.1)

The script contained in the MOS note ( odbsrmt.py) should work with a few minor changes to the instructions (since we are talking about an on-prem ZFS). I will continue to work through the changes and post the results in a future blog post.

3) You would register the restore location as an OKV endpoint (if it isn't already registered), OR you can alternately export the encryption key and create a wallet file.

Conclusion - This is a very exciting addition to the many features that the ZDLRA already provides.