Play app on AWS dies with - Oops, an error occured This exception has been logged with id 6i1972ib6.

[jaybytez]

I am doing my first super quick test of running an application on AWS.

My AWS instance is a micro EC2 (I first tried Amazon Linux 64bit and then I shifted to Ubuntu 13 32 bit).

I am running Play 2.2.1, with ebean, and using the default H2 (until I get all my app functionality completed to shift to MySQL or RDS).

I build the dist, scp it over to my AWS instance, and the unzip it (I have already installed the OpenJDK 7).

I run the app with evoluations on and 512 memory.  It runs fine for about 30-40 minutes without any issues.

Then around an hour after the application has been running, a request to my main page spews:

Oops, an error occured

This exception has been logged with id 6i1972ib6.

The server logs (with much of logging set to Debug), show that suddenly ebean is not aware of the Tables and throws an error:

[ESC[31merrorESC[0m] application -

! @6i1941l85 - Internal server error, for (GET) [/] ->

play.api.Application$$anon$1: Execution exception[[PersistenceException: Query t

hrew SQLException:Table "EVENT" not found; SQL statement:

The crazy thing is I just messed with my application for 30 minutes, saving all kinds of data into different tables, including this table.  Yet suddenly the application dies because it cannot find the table.  Is this an issue with running H2?  Or is this something related to a Unix that I need to do like ulimit (that this error is masking)?

Thanks for the help.

[jaybytez]

Well, I created a MySQL/RDS and converted my production sample to use the MySQL instance and the problem resolved itself.

Not sure if there was a memory issue with H2 on the micro instance that flushed the H2 memory causing the table to be removed.

But it appears to be working and running on the new DB instance.