I've generated changelog with
java -jar liquibase.jar --changeLogFile="./data/<insert file name>" generateChangeLog
I've got a "create table" changelog.
After it I altered table by adding a new column and runned
java -jar liquibase.jar --changeLogFile="./data/<insert file name>"
generateChangeLog
again. I expected to see incremental log with "create table" and "alter table", but it showing "ChangeLog ...... already exists!".
How to incrementally update ChangeLog?
Thank you!
The generateChangelog command is used to generate an initial changelog. If you want to append to a changelog you use the diffChangelog command.
Related
With my Spring Boot app, I have a database that is already set up but the flyway history table is empty. I would like to run flyway to generate the history table from the migration files.
When I run flyway:migrate it does not understand that my database credentials are in application.yml so I had to use a config file as below:
Approach1:
mvn -D flyway.configFiles=flyway.config -D flyway.baselineVersion=14 -D flyway.locations=classpath:db.development flyway:baseline
No error but it does not generate migration history.
Approach2:
mvn -D flyway.configFiles=flyway.config -D flywaytarget=14 -D flyway.skipExecutingMigrations=true -D flyway.locations=classpath:db.development flyway:migrate
Gives me an error:
org.flywaydb.core.api.FlywayException: Unknown configuration property: flyway.skipExecutingMigrations
Any idea how to achieve my goal?
We need to migrate our OWB repository to a new one running on 11.2. This will eventually be converted to ODI but as a 1st step we need to do this migration.
I've been following the notes in
https://docs.oracle.com/cd/E11882_01/owb.112/e17130/migrate.htm#WBINS07100
But we won't have access to a GUI for the repository assistant and need to run this from the command line.
Anybody have suggestions for how to do this?
You can run the OWB repository assistant in command line from the following location.
Go to the following location and run the command
In AIX or LINUX:
$OWB_HOME/owb/bin/unix/
./reposinst.sh
In windows:
$OWB_HOME/owb/bin/unix/reposinst.cmd
I'm currently updating a liquibase conf to springboot 2
long story short: I don't have any way to know what the previous version or configuration was.
I only have the script and the previous databaselock , which I want to continue with
My issue is the previous database lock column MD5SUM is empty. BUT my new liquibase fill the MD5SUM column.
So when I launch liquibase, it try to run all the oldest script; because he think they never run, because of the empty MD5SUM.
How can I configure liquibase to not fill MD5SUM?
(I use spring boot and CorisqSpringLiquibase)
This is a followup question to How can I create and load a second database in ddev?. It is about doing that task automatically.
One use case for this is developing a migration to Drupal from another MySQL database, and collaborating with others on the migration. If the database name can be set by ddev, additional developers can get the database created automatically, and additional databases can be added to their settings.local.php, using known values.
Try this in your project's config.yaml:
hooks:
post-start:
- exec: mysql -uroot -proot -hdb -e "CREATE DATABASE IF NOT EXISTS another_db; GRANT ALL ON another_db.* TO 'db'#'%';"
I am planning to use Jenkins with Hive. I have created some Hive scripts: test1.hql, test2.hql, test3.hql which lie on the sample server "ssh rt#test1.test2.net" location "/home/rt"
test1.hql:
select id, name from tblA;
test2.hql:
select id, name from tblB;
test3.hql:
select id, name from tblC;
How can I execute it from Jenkins?
On Jenkins up-to below is what I did:
New Job > Job name as test > Build a free-style software project
How can I point to my .hql scripts from Jenkins and what settings do I need to have? I am new to Jenkins.
Any help?
Thanks.
Found something:
Create a Jenkins job
In the configuration of that job there is a "Build" section with "Add build step"
Add a "Execute Shell" step, and put your code.
that code would be basically:
hive -f /pathtohql/test1.hql
The only issue i see with this is that the m/c with jenkins will have to have hive installed and pointing to the correct metastore & cluster.