You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on May 27, 2024. It is now read-only.
We currently have a hosted instance of the validator at http://transittools.forest.usf.edu/, but it's just a running instance of the JAR snapshot, which means that it's using Hibernate with HSQLDB. This gets unstable when running in production and inserting a lot of data. Plus, this virtualized machine only has 4GB of RAM.
There are a number of options for a permanent hosted instance of the validator. Ideally we'd get it running on a platform-as-a-service like Amazon Lamba (etc.) or Google App Engine, as there should be relatively low traffic and both of those services have a free tier. However, I believe both of the above require changes to the project to leverage Hibernate.
For now, I believe the most straight-forward path for something stable in the near term will be to leverage another internal server at CUTR for hosting the database server, which will free up transittools for just the validator.
@Suryakandukoori I believe the other CUTR virtualized machine already has MS SQL Server running on it, so let's just re-use this instead of installing a new DBMS (e.g., MySQL).
The text was updated successfully, but these errors were encountered:
As @Suryakandukoori worked on this, he discovered that several of the SQL statements behind some of the Hibernate view model objects do not work with Microsoft SQL Server. Therefore, to accomplish this task we need to abstract some of these SQL statements so they work beyond just HSQLDB.
Summary:
We currently have a hosted instance of the validator at http://transittools.forest.usf.edu/, but it's just a running instance of the JAR snapshot, which means that it's using Hibernate with HSQLDB. This gets unstable when running in production and inserting a lot of data. Plus, this virtualized machine only has 4GB of RAM.
There are a number of options for a permanent hosted instance of the validator. Ideally we'd get it running on a platform-as-a-service like Amazon Lamba (etc.) or Google App Engine, as there should be relatively low traffic and both of those services have a free tier. However, I believe both of the above require changes to the project to leverage Hibernate.
For now, I believe the most straight-forward path for something stable in the near term will be to leverage another internal server at CUTR for hosting the database server, which will free up transittools for just the validator.
@Suryakandukoori I believe the other CUTR virtualized machine already has MS SQL Server running on it, so let's just re-use this instead of installing a new DBMS (e.g., MySQL).
The text was updated successfully, but these errors were encountered: