I had some TransientError issues when using queues in iterative loops. The fix (workaround) that has eliminated the issue was to nest the iterative queue command inside a try/except. If there is a transient error, the handler does one time.sleep(0.1) command, and retries. If the error re-occurs, then the g.t. one sleep cycle error handling escalates. (And yes, I really, really hated to do burn up 100 cpu millis, but was desperate. Since TEs were infrequent, the overall resource waste proved de minimus.) To date this has fixed my transient errors. Never have had a TE outside of such loops. As with everything, YMMV. HTH -stevep
On Tuesday, December 18, 2012 10:37:31 AM UTC-8, Scott Gress wrote:
Hi all,
At the moment I'm getting TransientError exceptions whenever I try to add a task to a queue in my Python app, using code like:
q = Queue('default')
q.add(Task(url=url, countdown=countdown))
It appears to be happening with every queue, every time--retrying doesn't help. Has been going on for at least 30 minutes, and blocking important activity on my site. I haven't done any deployments in the last 4 days; it just started happening on its own.
This is happening in both my production app, "premium-prod-hr" and my staging app "premium-test-hr" on site
www.craftlaunch.com.
Anybody else seeing this? Any suggestions?
Thanks,
Scott