'Overwriting mode not activated' in the operational database during model deployment
Neel
New Altair Community Member
Hi All,
I am trying to deploy a rapidminer auto-model into a remote rapidminer server (9.5). The rapidminer server uses a docker based postgres database in the backend. When I am trying to deploy an auto-model into the server it complains with "problem with logging initialization: Table with name 'rm_mdm_locations' already exists and overwriting mode is not activated. Please change table name or activate overwriting mode". I feel manually changing the table name in the operational database is not a solution. Has anyone faced this problem?
I am trying to deploy a rapidminer auto-model into a remote rapidminer server (9.5). The rapidminer server uses a docker based postgres database in the backend. When I am trying to deploy an auto-model into the server it complains with "problem with logging initialization: Table with name 'rm_mdm_locations' already exists and overwriting mode is not activated. Please change table name or activate overwriting mode". I feel manually changing the table name in the operational database is not a solution. Has anyone faced this problem?
1
Best Answer
-
Hmm, nope, never saw that one, sorry. It looks like the deployments try to initialize the DB again. This should (obviously) not happen - after it has been initialized the first time, the next time should just deploy and not recreate the tables. I wonder if this could be connected to the fact that the DB is dockerized? Did you try with a plain vanilla DB first?Another thought I had: we had some more trouble with Postgres for deployments in the early days, MySQL worked a bit better. If it is possible to switch (for a test at least), this may be another thing worth trying.If you can share the log files of your Studio, there may be more information about the problem in there.Cheers,
Ingo3
Answers
-
Hmm, nope, never saw that one, sorry. It looks like the deployments try to initialize the DB again. This should (obviously) not happen - after it has been initialized the first time, the next time should just deploy and not recreate the tables. I wonder if this could be connected to the fact that the DB is dockerized? Did you try with a plain vanilla DB first?Another thought I had: we had some more trouble with Postgres for deployments in the early days, MySQL worked a bit better. If it is possible to switch (for a test at least), this may be another thing worth trying.If you can share the log files of your Studio, there may be more information about the problem in there.Cheers,
Ingo3