Importing Percona notary data to CockroachDB

Corda Enterprise supports two highly-available notary implementations: MySQL and JPA notaries. The MySQL notary requires a Percona database, and is a deprecated implementation.

To migrate from a MySQL notary to a JPA notary you must change the backend database and database schema. After notary data has been validated in the CockroachDB database, the JPA Notary Setup and Notary Worker Configuration processes can be followed.

Before beginning you must have:

  • An ssh accessible Percona cluster in read-only mode, or in read/write mode with no active connections.
  • An ssh accessible CockroachDB cluster.
  • A CockroachDB database cluster installed to /opt/cockroachdb.
  • A corda user within the CockroachDB cluster.
  • The required binaries are present in the PATH of the Percona and Cockroach machines.

Consider exchanging database ssh keys to avoid password requests when accessing the databases.

  1. The data in the Percona database must be extracted. To dump the corda database tables from the Percona database, use the following command, replacing $PERCONA with the address of your Percona database machine:

    ssh $PERCONA "bash -s -x -v" <<EOF
    mysqldump -u root --skip-lock-tables corda notary_committed_transactions notary_committed_states notary_request_log | gzip > dump.sql.gz
    EOF
    

    The dumped data will be stored locally in the /home/mysql directory.

  2. The local data can now be copied to CockroachDB nodes. Replace the Cockroach node addresses before running the following command:

    #  REPLACE WITH YOUR COCKROACH NODES
    COCKROACH_NODES="[email protected] \
    [email protected] \
    [email protected]"
    COCKROACH_DEST_PATH=/opt/cockroachdb/cockroach-data/extern/notary
    for NODE in $COCKROACH_NODES; do
      ssh $NODE "mkdir -p $COCKROACH_DEST_PATH"
      scp dump.sql.gz $NODE:$COCKROACH_DEST_PATH
    done
    

3. After copying the data to CockroachDB, the data must be imported to the corda_mysql database. The corda_mysql database is used to contain the MySQL data until it can be imported to the final corda database with the correct database schema. The following command will import the Percona data in the mysql schema: bash # Get the first node from the previous variable. COCKROACH_FIRST_NODE=${COCKROACH_NODES%% *} ssh $COCKROACH_FIRST_NODE "bash -s -x -v" << EOF /opt/cockroachdb/bin/cockroach sql --certs-dir /opt/cockroachdb/certs/ DROP DATABASE IF EXISTS corda_mysql; CREATE DATABASE IF NOT EXISTS corda_mysql; USE corda_mysql; # Note the location is the "data folder"/notary IMPORT MYSQLDUMP 'nodelocal:///notary/dump.sql.gz'; EOF The Percona data has now been imported into a Cockroach database called corda_mysql.

The code blocks above can be combined into a single bash script. Take care to replace the Percona and CockroachDB addresses in the script before use:

#!/usr/bin/env bash
# REPLACE THESE WITH YOUR MACHINE(S)
PERCONA=[email protected]
COCKROACH_NODES="[email protected] \
[email protected] \
[email protected]"
# END OF REPLACE
COCKROACH_FIRST_NODE=${COCKROACH_NODES%% *}
COCKROACH_ROOT=/opt/cockroachdb
COCKROACH_IMPORT_DIR=notary
COCKROACH_DEST_PATH=$COCKROACH_ROOT/cockroach-data/extern/$COCKROACH_IMPORT_DIR
echo --Dumping MySQL tables
time ssh $PERCONA "bash -s -x -v" <<EOF
rm dump.sql.gz
mysqldump -u root --skip-lock-tables corda notary_committed_transactions notary_committed_states notary_request_log | gzip > dump.sql.gz
ls -lh dump.sql.gz
EOF
echo --Copying export from mysql to local machine
scp $PERCONA:/home/mysql/dump.sql.gz .
echo --Making Cockroach data folders for import and copying export
for NODE in $COCKROACH_NODES; do
  ssh $NODE "mkdir -p $COCKROACH_DEST_PATH"
  scp dump.sql.gz $NODE:$COCKROACH_DEST_PATH
done
echo --Import dump into cockroach
time ssh $COCKROACH_FIRST_NODE "bash -s -x -v" << EOF
$COCKROACH_ROOT/bin/cockroach sql --certs-dir $COCKROACH_ROOT/certs/
DROP DATABASE IF EXISTS corda_mysql;
CREATE DATABASE IF NOT EXISTS corda_mysql;
USE corda_mysql;
IMPORT MYSQLDUMP 'nodelocal:///$COCKROACH_IMPORT_DIR/dump.sql.gz';
EOF
echo --Done
  1. To migrate the database schema, run the following script, replacing the Cockroach node addresses:

    #!/usr/bin/env bash
    # REPLACE THESE
    COCKROACH_NODES="[email protected] \
    [email protected] \
    [email protected]"
    #  We need a UUID to uniquely identify the data - you should generate your own
    UUID="d12888c8-cd00-11ea-a5c9-77fe2bd43698"
    # END OF REPLACE
    COCKROACH_FIRST_NODE=${COCKROACH_NODES%% *}
    CORDA_SCHEMA=corda
    #  Begin...
    echo Converting Cockroach tables
    COCKROACH_ROOT=/opt/cockroachdb
    time ssh $COCKROACH_FIRST_NODE "bash -s -x -v" <<EOF
    $COCKROACH_ROOT/bin/cockroach sql --certs-dir $COCKROACH_ROOT/certs/
    drop database if exists $CORDA_SCHEMA cascade;
    create database if not exists $CORDA_SCHEMA;
    create table $CORDA_SCHEMA.notary_committed_states (
        state_ref varchar(73) not null,
        consuming_transaction_id varchar(64) not null,
        constraint id1 primary key (state_ref)
        );
    create table $CORDA_SCHEMA.notary_committed_transactions (
        transaction_id varchar(64) not null,
        constraint id2 primary key (transaction_id)
        );
    create table $CORDA_SCHEMA.notary_request_log (
        id varchar(76) not null,
        consuming_transaction_id varchar(64),
        requesting_party_name varchar(255),
        request_timestamp timestamp not null,
        request_signature bytes not null,
        worker_node_x500_name varchar(255),
        constraint id3 primary key (id),
        index (consuming_transaction_id)
        );
    create table $CORDA_SCHEMA.notary_double_spends (
        state_ref varchar(73) not null,
        request_timestamp timestamp not null,
        consuming_transaction_id varchar(64) not null,
        constraint id4 primary key (state_ref, consuming_transaction_id),
        index (state_ref, request_timestamp, consuming_transaction_id)
        );
    create user if not exists corda;
    grant select on database $CORDA_SCHEMA to corda;
    grant insert on database $CORDA_SCHEMA to corda;
    grant select on table $CORDA_SCHEMA.* to corda;
    grant insert on table $CORDA_SCHEMA.* to corda;
    select 'CONVERTING notary_committed_states' as status;
    insert into
        $CORDA_SCHEMA.notary_committed_states
    select
        CONCAT(UPPER(ENCODE(issue_transaction_id, 'hex')), ':', TO_HEX(issue_transaction_output_id)) as state_ref,
        UPPER(ENCODE(consuming_transaction_id, 'hex')) as consuming_transaction_id
    from
        corda_mysql.notary_committed_states ncs
    --------------------------------------------------------------------------------
    select 'CONVERTING notary_committed_transactions' as status;
    insert into
        $CORDA_SCHEMA.notary_committed_transactions
    select
        UPPER(ENCODE(transaction_id, 'hex')) as transaction_id
    from
        corda_mysql.notary_committed_transactions;
    --------------------------------------------------------------------------------
    select 'CONVERTING notary_request_log' as status;
    insert into
        $CORDA_SCHEMA.notary_request_log
    select
        CONCAT('$UUID', TO_HEX(r.request_id)) as id,
        ENCODE(r.consuming_transaction_id, 'hex') as consuming_transaction_id,
        r.requesting_party_name as requesting_party_name,
        r.request_date::timestamp as request_date,
        r.request_signature as request_signature,
        r.worker_node_x500_name as worker_node_x500_name
    from
        corda_mysql.notary_request_log r;
    EOF
    

    The script has four distinct operations:

  • Creating a database named corda.
  • Creating a corda user and giving appropriate permissions.
  • Creating tables within the corda database.
  • Importing the notary data from the corda_mysql database into the corda database.
  1. You can validate the notary data by checking the row counts in both databases. To run the row count, run the following script, replacing the Percona and Cockroach node addresses with your machine addresses:
    #!/usr/bin/env bash
    # REPLACE THESE WITH YOUR MACHINE(S)
    PERCONA=[email protected]
    COCKROACH_NODES="[email protected] \
    [email protected] \
    [email protected]"
    # END OF REPLACE
    COCKROACH_FIRST_NODE=${COCKROACH_NODES%% *}
    CORDA_SCHEMA=corda
    COCKROACH_ROOT=/opt/cockroachdb
    echo PERCONA
    ssh $PERCONA "bash -s -x -v" <<EOF
    mysql -u root
    select count(*) as PERCONA_notary_committed_states from $CORDA_SCHEMA.notary_committed_states;
    select count(*) as PERCONA_notary_committed_transactions from $CORDA_SCHEMA.notary_committed_transactions;
    select count(*) as PERCONA_notary_request_log from $CORDA_SCHEMA.notary_request_log;
    EOF
    echo COCKROACH
    ssh $COCKROACH_FIRST_NODE "bash -s -x -v" <<EOF
    $COCKROACH_ROOT/bin/cockroach sql --certs-dir $COCKROACH_ROOT/certs/
    select count(*) as COCKROACH_notary_committed_states from $CORDA_SCHEMA.notary_committed_states;
    select count(*) as COCKROACH_notary_committed_transactions from $CORDA_SCHEMA.notary_committed_transactions;
    select count(*) as COCKROACH_notary_request_log from $CORDA_SCHEMA.notary_request_log;
    EOF
    

After the notary data migration is complete, at least one JPA notary node must be provisioned. To set up a JPA notary node, see the JPA Notary Setup and Notary Worker Configuration documentation. After the JPA notary is operating, you should run the Notary Health Check tool) to ensure that the network is up and responsive.

Was this page helpful?

Thanks for your feedback!

Chat with us

Chat with us on our #docs channel on slack. You can also join a lot of other slack channels there and have access to 1-on-1 communication with members of the R3 team and the online community.

Propose documentation improvements directly

Help us to improve the docs by contributing directly. It's simple - just fork this repository and raise a PR of your own - R3's Technical Writers will review it and apply the relevant suggestions.

We're sorry this page wasn't helpful. Let us know how we can make it better!

Chat with us

Chat with us on our #docs channel on slack. You can also join a lot of other slack channels there and have access to 1-on-1 communication with members of the R3 team and the online community.

Create an issue

Create a new GitHub issue in this repository - submit technical feedback, draw attention to a potential documentation bug, or share ideas for improvement and general feedback.

Propose documentation improvements directly

Help us to improve the docs by contributing directly. It's simple - just fork this repository and raise a PR of your own - R3's Technical Writers will review it and apply the relevant suggestions.