We are currently running Artifactory with the built-in Derby DB, and wanting to migrate to Oracle. So we follow the steps in https://www.jfrog.com/confluence/display/RTF/Configuring+the+Database to migrate data. In data export, I selected "Exclude Content".
We configured Artifactory with LDAP and after migration, the LDAP configs are all in place and login is working fine with LDAP. Everything seems OK but when I check the Permission and Groups, none got migrated over to Oracle. Am I missing something here? Do I need to recreate the Permissions and Groups? My understanding and assumption are that, everything will be migrated over.
Related
can someone help to setup correct jdbc connection string using oracle wallet in flyway conf?
If I tried sqlplus /#WALLET_NAME it works perfect, but by flyway it always wants credentials (no error).
I have flyway.url=jdbc:oracle:thin:/#WALLET_NAME also tried flyway.url=jdbc:oracle:thin:#WALLET_NAME
Thanks for any tips.
Maybe you are using the free flyway version. And it seems that's a paid feature.
See here and check the flyway teams label:
https://flywaydb.org/documentation/configuration/parameters/oracleWalletLocation
I tried to use Flyway. But, the free version doesn't support Oracle Wallet. And this is mandatory to use Oracle Autonomous Database on Oracle Cloud.
And the paid version is too expensive for me.
So, I decided to go back to Liquibase.
In JFrog's Artifactory Installation Documentation, there are 2 conflicting statements under the Configuring the Database, Oracle section:
In the Overview:
For Artifactory to run with Oracle you must create a dedicated Oracle database instance and then configure Artifactory to use it as described in the following sections.
Under Creating the Artifactory Oracle Database, Item #2
Recommendation: With both of the above options (Full DB and DB-Filesystem), it is recommended to create a dedicated table space and use AL32UTF8 encoding.
I bring up this issue, because we were looking to just create a new tablespace in a pre-exisiting repository type database.
Also on the site, under SQLServer is this tidbit, written by a different person:
Optimizing Artifactory when running with MS SQL Server
When running Artifactory with Microsoft SQL Server you may create the Artifactory schema on an existing server used for other applications, however for optimal performance, we recommend creating a dedicated Microsoft SQL Server database instance and then configure Artifactory to use it as described in the following sections.
So I guess my question is this. For Oracle, is it a recommendation or a requirement to have a dedicated database instance?
I have recently upgraded artifactory from version 4.9.0 to 6.1.0 along with DB migration from derby to postgreSQL. Although it is written to have a dedicated host for the postgreSQL server I have tested both the above activities on the same host and it worked fine. For the actual upgrade we have a dedicated DB instance.
But I guess by dedicated host the artifactory might be referring to best pratice.
I'm trying to reset the admin password of my SonarQube and I have already found that is executing the following query:
update users set crypted_password = '88c991e39bb88b94178123a849606905ebf440f5', salt='6522f3c5007ae910ad690bb1bdbf264a34884c6d' where login = 'admin'
But the thing is that I don't know what is the database manager where I need to execute it and how to connect to it. When I installed my SonarQube and my SonarQube Scanner I didn't specified any database manager.
The only "default" SonarQube database is the embedded H2 database, which is provided solely for initial testing. It is neither intended nor supported for production use. You will not be able to upgrade with this database. As a large red banner should be telling you at the bottom of every single page in the SonarQube interface.
But okay, to answer your question, you can use any generic client to make this update. I generally use SQuirreL SQL on the rare occasions I need to peer into the black box that is the SonarQube database.
On SQL Server 2008R2, we have a SQL login tagged with the sysadmin server role. Anytime a db is created, this login is automatically added as a user to the db. Is there a switch that will prevent this from happening?
EDIT:
Based on the answer below, I am adding more background info to explain why we want to do this.
When we publish to a new database from a Visual Studio SQL project, the publishing engine wants to add a user to the database. The username already exists as a login to the database (in the sysadmin server role -- and not much we can do about that), so the publish fails as the user gets auto-added by SQL server after the 'create database' statement. I was hoping for a switch that would tell SQL Server to not add the user automatically. So it seems we will have to find another solution on the VS side that tells the publishing engine to ignore the SQL DB user.
Assuming you're talking about the sql-server role, I don't think so. TBH if you're contemplating keeping a sys-admin out of a db, then you've got a different problem than the one you think you have...
I have a resolution to this. Download the latest SSDT package from MS and now there are new options under the Publishing Wizard Advanced... button to exclude the publishing of logins, roles, etc. The SSDT package I downloaded I found here...
https://msdn.microsoft.com/en-US/dn864412
I am trying to migrate sonar's H2(default database) to oracle database. For this I have edited sonar.properties file. I commented sonar URL and uncommented oracle URL(I made jdbc connection with oracle database and pasted that URL in sonar.properties). I have also changed sonar usename and password in sonar.properties. When I start sonar it gives me error and ask me to create tables in oracle database and even if I create tables in oracle it still ask me to create such more tables.
Do I need to make some more configuration/file changes? Am I missing something?
Any help would be highly appreciated.