3 ms·

Show HN: TeamPostgreSQL, neat admin interface for PostgreSQL

10 points by johnyzee 17y ago

I am just getting ready to release TeamPostgreSQL 1.05. You can check it out here: http://www.teampostgresql.com It is a web-based admin interface for PostgreSQL, with a lot of work put into making it a user-friendly and productive way to work with the database. Hopefully this makes it ideal for when different people need to access the database, including developers, testers, support staff, managers etc. This will be the first release after several months of beta testing and user feedback. Some feature highlights: * Pretty neat web interface (GWT powered AJAX GUI) * Comprehensive database management, including database objects and tablespace management * Easy yet powerful data viewing, using one-click search, relationship navigation, filtering and more * Upload and download files directly to/from binary fields * Tabbed SQL editor with SQL auto-completion * Favorites function for easy storing/sharing of snippets and scripts * Import/export of database objects and entire databases I'd love to hear what you think!

lazyant 17y ago
Anecdote: I selected a trigger, clicked on "Generate DDL" and it just stayed on "Generating, please wait" forever
- johnyzee 17y ago
  Thanks, will fix. It is actually a permission getting denied for the demo account (there is an error message being obscured by the progress dialog). Note that it should work fine when you run as administrator.
kls 17y ago
Please don't take this as knocking your hard work and I like your product, but I don't think another admin interface is not truly what the market needs. PGAdmin does pretty well and the only significant feature that I see in your front end over PGAdmin (in my very limited exposure to your product), is that your product is web based. While a good feature, it is not the killer feature that knocks the competition dead. I can easily manage my off-site Postgres databases with PGAdmin. To me where many of the tools that support the open databases are lacking, is in the analytic department. This is truly the hard stuff to do in the database world. adding features like telling me my longest running query, my most CPU intensive query or my most memory intensive query (etc. etc.) and providing a clean UI to display and refine (make recommendations) this information goes along way to helping developers and pay for themselves rapidly. These features sell themselves. Provide these in a good UI and I would definitely look at your product given that Postgres is my turn to DB any time I have a choice.
- johnyzee 17y ago
  Thanks for the comments, I do appreciate the input. A couple of advantages over pgAdmin for an offsite database is that it will not drop connections, and it can easily be configured for SSL using https. Also, clients don't have to install anything. Feature-wise the focus at this point has mostly been on powerful data navigation, i.e. you can do a quick search for a customer id, click the reference to the purchase order table, click to download a pdf contract directly from a binary field (just an example). The tool provides significant time-savings for these tasks, though you may not discover this unless you delve into the product a little. More generally there are many little productivity improvements and details that support the user that all add up to provide value, which may not be noticed at a quick glance. I hope to be able to communicate these as efficiently as possible. Going forward there is a lot of room for the kind of analytical features you mentioned. This is still an early version of the product with focus on core administration tasks. In fact management of running processes is due in the next release.
  - kls 17y ago
    A couple of things, powerful data navigation and export to pdf from the UI are rely nice features and they will make life easier for the end user, but in my experience the main user of a DB admin application are developers, DBA's and admins. Most of which are not as concerned with the actual contents of the data, as they are with the structures of the data as well as the performance of the data. Developers general concern themselves with the data when they are setting up test, or diagnosing a data related problem but in the coarse of a day most of their focus is on the structure and performance of the data and database. The point being, generally your call center rep or your accountant (the end user) uses a custom interface dedicated to data navigation and data viewing (coupled with workflow) these features will most likely not be exposed to them and they are the users most advantaged by them, particularly the export to PDF feature. Again, please don't take this as knocking your hard work, I think that powerful data navigation is important for those times that you do have to directly manage the data and think other DB admin front ends are severely lacking in that departments, but I urge you to keep in mind that in most organizations almost all the end users that are interested in the actual data are buffered away from a raw DB interface, even report writers who know SQL fairly well are buffered with tools such as Cognos and Crystal Reports. But I think you have the basis to cure a pain point that plagues a lot of organizations given that you now have a good base product to build a robust development tool on. A portion of (software) developers do not like and look at relational databases as second class to actually writing software. If you can mold your tool in to providing deep analysis and diagnostics and marry that with solution engine that provides recommendations on how to optimize and refine the data structure and query structures so that they are more performant, more secure, less resource intensive and you can do this in a seamless way you are scratching an itch that a good portion of developers cant reach.

10 points by johnyzee 17y ago

lazyant 17y ago

Anecdote: I selected a trigger, clicked on "Generate DDL" and it just stayed on "Generating, please wait" forever

johnyzee 17y ago
Thanks, will fix. It is actually a permission getting denied for the demo account (there is an error message being obscured by the progress dialog). Note that it should work fine when you run as administrator.

kls 17y ago

Please don't take this as knocking your hard work and I like your product, but I don't think another admin interface is not truly what the market needs. PGAdmin does pretty well and the only significant feature that I see in your front end over PGAdmin (in my very limited exposure to your product), is that your product is web based. While a good feature, it is not the killer feature that knocks the competition dead. I can easily manage my off-site Postgres databases with PGAdmin. To me where many of the tools that support the open databases are lacking, is in the analytic department. This is truly the hard stuff to do in the database world. adding features like telling me my longest running query, my most CPU intensive query or my most memory intensive query (etc. etc.) and providing a clean UI to display and refine (make recommendations) this information goes along way to helping developers and pay for themselves rapidly. These features sell themselves. Provide these in a good UI and I would definitely look at your product given that Postgres is my turn to DB any time I have a choice.

johnyzee 17y ago
Thanks for the comments, I do appreciate the input. A couple of advantages over pgAdmin for an offsite database is that it will not drop connections, and it can easily be configured for SSL using https. Also, clients don't have to install anything. Feature-wise the focus at this point has mostly been on powerful data navigation, i.e. you can do a quick search for a customer id, click the reference to the purchase order table, click to download a pdf contract directly from a binary field (just an example). The tool provides significant time-savings for these tasks, though you may not discover this unless you delve into the product a little. More generally there are many little productivity improvements and details that support the user that all add up to provide value, which may not be noticed at a quick glance. I hope to be able to communicate these as efficiently as possible. Going forward there is a lot of room for the kind of analytical features you mentioned. This is still an early version of the product with focus on core administration tasks. In fact management of running processes is due in the next release.
- kls 17y ago
  A couple of things, powerful data navigation and export to pdf from the UI are rely nice features and they will make life easier for the end user, but in my experience the main user of a DB admin application are developers, DBA's and admins. Most of which are not as concerned with the actual contents of the data, as they are with the structures of the data as well as the performance of the data. Developers general concern themselves with the data when they are setting up test, or diagnosing a data related problem but in the coarse of a day most of their focus is on the structure and performance of the data and database. The point being, generally your call center rep or your accountant (the end user) uses a custom interface dedicated to data navigation and data viewing (coupled with workflow) these features will most likely not be exposed to them and they are the users most advantaged by them, particularly the export to PDF feature. Again, please don't take this as knocking your hard work, I think that powerful data navigation is important for those times that you do have to directly manage the data and think other DB admin front ends are severely lacking in that departments, but I urge you to keep in mind that in most organizations almost all the end users that are interested in the actual data are buffered away from a raw DB interface, even report writers who know SQL fairly well are buffered with tools such as Cognos and Crystal Reports. But I think you have the basis to cure a pain point that plagues a lot of organizations given that you now have a good base product to build a robust development tool on. A portion of (software) developers do not like and look at relational databases as second class to actually writing software. If you can mold your tool in to providing deep analysis and diagnostics and marry that with solution engine that provides recommendations on how to optimize and refine the data structure and query structures so that they are more performant, more secure, less resource intensive and you can do this in a seamless way you are scratching an itch that a good portion of developers cant reach.