Why is the Baserow's initiation routine when initiating the docker container so long and cpu intensive? Shouldn't it be a simple schema update, why does the migration work take tens of minutes on a 2vcore? Once the database is initiated, it runs without problem. But the initiation is really a killer. It often breaks on lower speced servers and is the reason why I, although I like the project, won't be using it. You should really check your initiation routine and find a better way to handle migrations.
Hey @HideBuz! The initialization of the database schema shouldn't take much time. We suspect it's the installation of all the templates that's consuming the resources before your instance starts. What happens if you set BASEROW_TRIGGER_SYNC_TEMPLATES_AFTER_MIGRATION=false? If you're running Baserow on lower-spec servers, then we'd recommend setting BASEROW_RUN_MINIMAL=true. This will allow the project to run with minimal hardware, but it does come with a concurrency and performance cost. We hope that helps!
@@baserow thanks for answering. When I set templates to false, the initiatialization still crashes. I think it has to do with Celery consuming all of memory and on 2GB it simply just crashes. The minimum requirement should be 3 or 4GB of RAM, but the thing is: it's just the installation! Why not just import a SQL file on a new instance? After initalization the RAM falls back to 1GB so it's not like it requires a 4GB model to begin with. It's just your migrations are problematic.
@@HideBuz Apologies for the delay in response. Do you mind if we switch to the Baserow community forum to help with your question/issue? community.baserow.io/
![Introduction to Baserow's Application Builder [Webinar]](http://i.ytimg.com/vi/bTvqn3AT4D8/mqdefault.jpg)
![Introduction to Baserow's Application Builder [Webinar]](/img/tr.png)
Great update!! 🤩 😁
Why is the Baserow's initiation routine when initiating the docker container so long and cpu intensive? Shouldn't it be a simple schema update, why does the migration work take tens of minutes on a 2vcore? Once the database is initiated, it runs without problem. But the initiation is really a killer. It often breaks on lower speced servers and is the reason why I, although I like the project, won't be using it. You should really check your initiation routine and find a better way to handle migrations.
Hey @HideBuz! The initialization of the database schema shouldn't take much time. We suspect it's the installation of all the templates that's consuming the resources before your instance starts. What happens if you set BASEROW_TRIGGER_SYNC_TEMPLATES_AFTER_MIGRATION=false? If you're running Baserow on lower-spec servers, then we'd recommend setting BASEROW_RUN_MINIMAL=true. This will allow the project to run with minimal hardware, but it does come with a concurrency and performance cost.
We hope that helps!
@@baserow thanks for answering. When I set templates to false, the initiatialization still crashes. I think it has to do with Celery consuming all of memory and on 2GB it simply just crashes. The minimum requirement should be 3 or 4GB of RAM, but the thing is: it's just the installation! Why not just import a SQL file on a new instance? After initalization the RAM falls back to 1GB so it's not like it requires a 4GB model to begin with. It's just your migrations are problematic.
@@HideBuz Apologies for the delay in response. Do you mind if we switch to the Baserow community forum to help with your question/issue?
community.baserow.io/