Since taskqueue is still located in labs, I thought I'd give some
feedback. ;)

I'm a bit obsessive about log clarity. I think that taskqueue should
support the following:

- a 400-series status code indicates a client error, so the task
should _not_ be requeued
- a PermanentTaskFailure should place a 500-series status code into
the logs, but the task should _not_ be requeued

There is a super-useful panel on the Dashboard of App Engine: the
"Error" urls, specifically, which urls have the most 400/500
responses. It is an great way to hone down to the pain points in your
application and correct them.

When we use taskqueue, we often will allow something to retry a few
times, and then we stop it with a PermanentTaskFailure - we do this to
prevent clogging up the queues with bad jobs (e.g., might be a job
that needs to communicate with an external server, or simply a code
bug, or some unexpected data in datastore). The issue is that this
most critical failure (i.e., the PermanentTaskFailure) is hidden
behind a 200 status code. It should be a big, fat 500.

-- 
You received this message because you are subscribed to the Google Groups 
"Google App Engine" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/google-appengine?hl=en.

Reply via email to