We are seeing this exact problem with our app (OAuth). Different sets of requests return different counts for the rate limit. It goes up and down and sometimes reaches 0 which is incorrect.
Any idea? On Sep 2, 11:47 pm, srikanth reddy <srikanth.yara...@gmail.com> wrote: > I have seen some inconsistency with my desktop app(Oauth) which is not > whitelisted > Adding a tweet to favorite does not update the X-RateLimit value. > Also If i remove a tweet from my favorites ( favorites\destroy), i get 404 > error (But this is removed from favorites) and the ratelimit is reset even > though i have not consumed all 150 calls. > Basically All deletes (i have seen this for status\destory, dm\destroy, > favorites\destroy for recent tweets) are giving 404. But they get deleted > successfully. X-RateLimit is behaving strange w.r.t favorites > > On Thu, Sep 3, 2009 at 12:07 PM, Waldron Faulkner <waldronfaulk...@gmail.com > > > wrote: > > > Strange events w/ Rate Limit requests. > > > I'm calling the API from my whitelisted IP and getting results that > > are all over the map. It's almost as if Twitter is load-balancing my > > requests to two different environments, each of which is keeping its > > own count of my rate limits. So my app chugs along happily thinking it > > has plenty of limits and shouldn't need to check for a while, and then > > wham, I'm getting 404's and Rate Limit exceptions. > > > Check this output from one of my apps: > > > Rate lims for acct: 7727 > > Rate lims for acct: 2002 > > 2009-09-03 02:12:04 AM: Processed 1000 tasks (∞ / min) > > Rate lims for acct: 1136 > > 2009-09-03 02:12:25 AM: Processed 2000 tasks (∞ / min) > > Rate lims for acct: 7052 > > 2009-09-03 02:12:46 AM: Processed 3000 tasks (3000 / min) > > > Notice how the rate lim requests bounce from the 7K to the 1K range > > > Then, a few seconds later, I get a ton of 404, and finally an over-the- > > rate-lims response. > > > This only happens from my whitelisted IP. I'm running the same app > > from home (account whitelisted but not the ip) and it runs without > > this problem. > > > What's up??
