sekikn commented on PR #1199: URL: https://github.com/apache/bigtop/pull/1199#issuecomment-1770544958
@JiaLiangC My bad, [my comment](https://github.com/apache/bigtop/pull/1197#issuecomment-1769704046) was incomplete. After applying the current PR, `pg_ctl initdb` is executed before installing ranger-admin. But postgresql is installed as Ranger's dependency, so there is no executable when the command above is executed. As its result, applying the Ranger manifest fails as follows. ``` $ ./docker-hadoop.sh -d -dcp -C config_ubuntu-22.04.yaml -F docker-compose-cgroupv2.yml -G -L -k hdfs,hbase,ranger -s ranger -c 1 ... Error: Could not find command '/usr/bin/pg_ctl' Error: /Stage[main]/Ranger::Admin/Exec[initdb]/returns: change from 'notrun' to ['0'] failed: Could not find command '/usr/bin/pg_ctl' ``` So we have to reorganize the Puppet manifest, not only removing package declarations. The steps are supposed to be: 1. install python and ranger-admin (and postgres as its dependency) 2. run initdb after ensuring 1 is finished 3. startup postgres ensuring 2 is finished 4. put install.properties ensuring 1 is finished 5. run setup.sh ensuring 3 and 4 are finished In addition, the directory layout of postgres seems to differ between rpm- and deb-based systems. For example, pg_ctl doesn't seem to be located in /usr/bin, but /usr/lib/postgresql/$VERSION/bin on Debian/Ubuntu. Also, the database directory currently specified in the Puppet manifest (/var/lib/pgsql/data) probably doesn't work on deb-based systems due to the directory permission. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
