Facebook
Twitter- Server operating system version
- Ubuntu 24.04.3 LTS
- Plesk version and microupdate number
- Plesk Obsidian 18.0.75 Web Pro Edition
Hello,
I’m currently blocked on a PostgreSQL 16.11 administration task in Plesk and would appreciate clarification on the correct and supported workflows.
I initially attempted to create a new PostgreSQL admin user via the Plesk GUI, in order to replace the previous one also created from the interface on a fresh install.
During this process, Plesk reported blocking dependencies and I could not complete the operation from the GUI. To proceed, I created the new admin role directly in PostgreSQL (backend) as a superuser.
From PostgreSQL’s perspective:
* databases (oneDB, twoDB) are owned by postgres
* the new admin role (adminuser) exists and is a superuser
* the previous role (prevadminuser) does not own databases or objects
However, when managing Authentication section of PostgreSQL via the Plesk interface, Plesk reports the following error:
ERROR: role "prevadminuser" cannot be dropped because some objects depend on it
DETAIL: owner of database oneDB
owner of database twoDB
This blocks further user management in the GUI, even though PostgreSQL system catalogs show that these databases are not owned by prevadminuser.
Additional concern: PostgreSQL import workflow and GUI visibility
According to the official Plesk documentation (https://support.plesk.com/hc/en-us/...reSQL-database-dump-on-Plesk-for-Linux-server), PostgreSQL imports are performed using command-line tools.
In practice, when following this approach, the imported database may not appear correctly in the Plesk GUI, and it is unclear whether such databases are fully managed by Plesk (including backups). In general, PostgreSQL management workflows under Plesk often lead to errors or blocked actions even when PostgreSQL itself is in a valid state.
Due to this I'm currently creating new DBs from GUI and later importing existing one using the DB and USER pair I just created from GUI using --no-owner --no-privileges --clean
Request for clarification
Could you please clarify:
1. The supported/recommended way to introduce or replace a PostgreSQL admin user in Plesk
2. How PostgreSQL roles are expected to be managed when changes are made directly in PostgreSQL
3. Whether there is a supported procedure to refresh/resynchronize Plesk’s view of PostgreSQL roles/users/databases
4. Whether databases imported using the documented CLI methods are fully:
* visible in the Plesk GUI
* managed by Plesk
* included in backups
5. The recommended PostgreSQL import/export workflow in Plesk that avoids errors and inconsistent states
At the moment, PostgreSQL administration through Plesk is difficult to reason about, and the documentation does not clearly define safe and supported workflows for production use.
Thank you for your help and clarification.
I’m currently blocked on a PostgreSQL 16.11 administration task in Plesk and would appreciate clarification on the correct and supported workflows.
I initially attempted to create a new PostgreSQL admin user via the Plesk GUI, in order to replace the previous one also created from the interface on a fresh install.
During this process, Plesk reported blocking dependencies and I could not complete the operation from the GUI. To proceed, I created the new admin role directly in PostgreSQL (backend) as a superuser.
From PostgreSQL’s perspective:
* databases (oneDB, twoDB) are owned by postgres
* the new admin role (adminuser) exists and is a superuser
* the previous role (prevadminuser) does not own databases or objects
However, when managing Authentication section of PostgreSQL via the Plesk interface, Plesk reports the following error:
ERROR: role "prevadminuser" cannot be dropped because some objects depend on it
DETAIL: owner of database oneDB
owner of database twoDB
This blocks further user management in the GUI, even though PostgreSQL system catalogs show that these databases are not owned by prevadminuser.
Additional concern: PostgreSQL import workflow and GUI visibility
According to the official Plesk documentation (https://support.plesk.com/hc/en-us/...reSQL-database-dump-on-Plesk-for-Linux-server), PostgreSQL imports are performed using command-line tools.
In practice, when following this approach, the imported database may not appear correctly in the Plesk GUI, and it is unclear whether such databases are fully managed by Plesk (including backups). In general, PostgreSQL management workflows under Plesk often lead to errors or blocked actions even when PostgreSQL itself is in a valid state.
Due to this I'm currently creating new DBs from GUI and later importing existing one using the DB and USER pair I just created from GUI using --no-owner --no-privileges --clean
Request for clarification
Could you please clarify:
1. The supported/recommended way to introduce or replace a PostgreSQL admin user in Plesk
2. How PostgreSQL roles are expected to be managed when changes are made directly in PostgreSQL
3. Whether there is a supported procedure to refresh/resynchronize Plesk’s view of PostgreSQL roles/users/databases
4. Whether databases imported using the documented CLI methods are fully:
* visible in the Plesk GUI
* managed by Plesk
* included in backups
5. The recommended PostgreSQL import/export workflow in Plesk that avoids errors and inconsistent states
At the moment, PostgreSQL administration through Plesk is difficult to reason about, and the documentation does not clearly define safe and supported workflows for production use.
Thank you for your help and clarification.
