[OT] False timeout error after URL retrieval

[J. Landman Gay]

I am using "get url" to retrieve data from an AWS server. The server 
scripts are in Ruby on Rails and they return the correct data. This has 
been working fine for over a year until last week when the IT person 
upgraded Ruby to the latest version. Suddenly we're getting timeout 
errors all over the place and the app has become unusable.

Debugging shows that the data is returned correctly but LC continues to 
wait until the timeout period expires, and then reports a timeout error. 
It is acting like it doesn't know the transmission has completed, or 
there is no EOF marker (or whatever is used to designate the end of the 
data transfer.) The data is gzip formatted, normally sent as a stream, 
but we also tried saving it as a file and then sending the file without 
success.

The Ruby person has been troubleshooting for a week and can't find 
what's wrong. Reverting to an older version of Ruby fixes it, but we 
don't know why, and for various other reasons we definitely need to 
update to the new version.

I said I'd ask here in case someone familiar with Ruby and Passenger has 
a clue why this should suddenly be happening. The client is a little 
anxious at this point.

-- 
Jacqueline Landman Gay         |     jacque at hyperactivesw.com
HyperActive Software           |     http://www.hyperactivesw.com