Request per hour rate exceeded- bumped off- please help


#1

Hi, we are creating digital signage systems and pull the twitter account feed in using rss. We are not using PC to do this, rather a browser based device that pulls in the xml.

We are are getting bumped with this message “<?xml version="1.0" encoding="UTF-8"?>

Rate limit exceeded. Clients may not make more than 150 requests per hour.
/statuses/user_timeline/13941562.rss

how do I around this?

I am currently compiling a solution for a client that holds a static IP and uses numerous digital signage screens. Each department need to see their own twitter feed on screen so the total request may be more that 150 an hour.

Could you advise best practice please?

Can I increase the request oer hour total? Thanks for your help


#2

Make sure you’ve read the documentation around rate limiting on this site for the big picture.

Also verify you’re using the right URLs – RSS user timeline requests should look like this in the present day: https://api.twitter.com/1/statuses/user_timeline.rss?user_id=13941562

In your kind of scenario, the simplest-most implementation isn’t going to cut it. You’ll either need to use OAuth and a dedicated account/access token to guarantee you 350 requests per hour, or you’ll need to recommend a more robust and centralized system in which you perhaps leverage the Streaming API to monitor the feeds of multiple accounts with a single connection, building a pool of Tweets on some kind of server that the digital signs then pull their data from (rather than from the API directly).


#3

Hey
Thanks for the info (-;

Our system is unable to use OAuth or an App (there is no O/S to work with) so we are stuck here, will have to figure out a workaround like you suggest using a centralised system. Shame it’s this complicated.

will get devs to limit calls made and plan a longer term solution; can I ask, how long will we be unable to make calls after exceeding limit?

Thanks for your help


#4

The duration and time of rollover when rate limited is communicated in the X-Ratelimit-* HTTP headers in every response – there will be a value you can consume that will tell you when you can begin making requests again. If you use 150 requests in the first 15 minutes, you’d have 45 minutes to wait. If you space the 150 requests over the course of a full 60 minutes, you should be fine.


#5

Thanks (-;