Yesterdat I tried to install the latest version of Tracks on Textdrive using the standard lighttpd setup as recommended by Textdrive. Unfortunately, when I visit the url I get the dreaded ‘application error’ . I have used the same setup for a couple of other Rails apps that run just fine so I’m pretty sure this is something specific to Tracks.

Nothing the logs are all empty, but when I run rake test:units—trace I get some errors.

I have attached the backtrace as an image because I get a ‘precondition failed’ error when I include it as text. Something to do with mod_security I think. It’s a little hard to read because the picture can’t be larger than 800*600, if someone knows a better way to include it, I’m all ears.

Is this Tracks 1.043 or the trunk (which I wouldn’t advise running on a shared server)? It looks like rake is having problems with the schema and/or the database. Did rake db:migrate go OK? You could try going into your Tracks directory and doing:

to see if it can connect to your db OK.

Did you double check that the shebang lines are appropriate for TextDrive in the files in script/ and also the public/dispatch.* files?

I run Tracks myself on TextDrive (as do others), so it’s probably a minor config error somewhere.

According to the installation.html file it’s version 1.042. Rake db:migrate went fine and script/console also works.

The shebang in the dispatch files was #!/usr/bin/env ruby, which I changed to #!/usr/local/bin/ruby. Didn’t work though.. is it necessary to change the shebang in all the files in the script directory as well? They are not being used for running the server, right?

I was surprised I couldn’t get it to work because you are on Textdrive yourself. It is probably a minor config error, but I have no idea how to figure out what it is exactly.

I forgot to update the version number in installation.html in 1.043, so that probably won’t help you. But if you followed the download link for 1.043, you should be fine.

As for the rest, I’m baffled. If script/console is working, then it’s getting to your db OK, and the shebang is OK in the dispatch files, then there shouldn’t be a problem. No, you don’t have to change the script files as well, but I do it anyway because I always forget that I haven’t when I need to run one of those scripts and spend a few minutes wondering why it’s not working.

I’m sure you did the setup fine, but it might be worth double checking that you’ve set everything up as per the instructions on the TextDrive KB, and that there are no typos or other minor problems in any of the scripts you created, and that you’ve created an entry for ‘production’ in database.yml as well as one for ‘development’ (they can refer to the same database. I assume that you got your port set up correctly and so on for lighttpd. When you try to run rails.sh, what happens? Does it return an error? Does it keep on running even after you’ve got the application error through the browser?

If you’re really stuck and you trust me with your login details (and I can quite see why you might not!) I could log in to your shell in and take a look at what you’ve got set up to see if I can figure out where it’s going wrong.

I have the same problem. I followed the installation instructions, and my own instructions for installing on TextDrive, but I only get “Application error (Rails)“ when I try to access Tracks.

I tried the rake test:units—trace too, and got errors too:

  1) Error:
test_delete_context_deletes_todos_within_it(ContextTest):
ActiveRecord::StatementInvalid: SQLite3::SQLException: near “DISTINCT”: syntax error: SELECT count(DISTINCT todos.id) AS count_all FROM todos LEFT OUTER JOIN projects ON projects.id = todos.project_id WHERE (todos.context_id = 1)

  2) Error:
test_delete_project_deletes_todos_within_it(ProjectTest):
ActiveRecord::StatementInvalid: SQLite3::SQLException: near “DISTINCT”: syntax error: SELECT count(DISTINCT todos.id) AS count_all FROM todos LEFT OUTER JOIN contexts ON contexts.id = todos.context_id WHERE (todos.project_id = 1)

Travholt: I still don’t know what to suggest - you’re using 1.043, too right? I’m sure it must be a very minor config error somewhere, wrong permissions or something. Is there nothing illuminating in the logs?