Are you aware of a global API Search quota based on IP ?
I am running an app that sends queries every 30 seconds to find if there are news about traffic issues. The app is operational since 2 years.
The app avoids quota limitation problems by using the information provided in http headers.
Also, in order to limit the amount of transferred data it uses the since_id parameters as explained here :
In fact, every new query will find a small number of tweets.
Since beginning of December I experience a high number of «execution timeout» issues on application side.
This kind of error, happened before, but maybe once a week. It was not an issue because we can deal with this kind of problem. Since December the number of errors increased to several hundreds a day.
Nothing changed on the application side. I searched in different forums see if one of the layers I am using experience similar issues but found nothing (Twitter Ruby Gem).
The network configuration didn’t changed and the admin guys told me they see no errors on the traffic side.
The issue is not related to a specific search but to any of the 4 queries I am running. And it is not related to time of the day (there are very few events during the night for ex).
Seems like regularly one of the queries will take much more time than usual and push the app to a timeout.
Definitively something changed in the last month that affect the app.
Any idea or feedback is welcomed,