These forums have been archived and are now read-only.

The new forums are live and can be found at https://forums.eveonline.com/

EVE Technology Lab

 
  • Topic is locked indefinitely.
 

questions about rate limits

First post
Author
Previous Topic Next Topic
Antice Eton
The Scope
Gallente Federation
#1 - 2015-01-23 17:54:50 UTC
Scenario:

I am working on a desktop application that, when started, will refresh its list of alliances, and sovereignty from the EVE API. From what I gather, that's two separate requests, to be made one after the other. No further requests will be made unless the application is restarted.

Question #1:
Is rate limiting based on user agent, IP, or both?
If this application proves popular, and 1000 people start it up, at once, but from different IP addresses, should I encounter any issues?

Question #2:
What is the response when something gets rate limited? Eg: if some user, for some reason, decides to rapidly restart the application so that the requests made exceed the rate limit, what would be the HTTP response?

thank you in advance
CCP FoxFour
C C P
C C P Alliance
#2 - 2015-01-24 14:43:12 UTC  |  Edited by: CCP FoxFour
Antice Eton wrote:
Scenario:

I am working on a desktop application that, when started, will refresh its list of alliances, and sovereignty from the EVE API. From what I gather, that's two separate requests, to be made one after the other. No further requests will be made unless the application is restarted.

Question #1:
Is rate limiting based on user agent, IP, or both?
If this application proves popular, and 1000 people start it up, at once, but from different IP addresses, should I encounter any issues?

Question #2:
What is the response when something gets rate limited? Eg: if some user, for some reason, decides to rapidly restart the application so that the requests made exceed the rate limit, what would be the HTTP response?

thank you in advance


1) The rate limit is IP based not User-Agent. Please do send proper user-agents however so that we can monitor it's usage. This is mainly so that if we see something funny or bad happening we know who to contact. Our first course of action, assuming your application is not harming us, is to contact you in order to request you make any change we need. If we cannot contact you, or you don't respond within a reasonable amount of time, then we look at blocking things.

2) I actually don't know... >.< Keep in mind for the XML API the rate limit is ~30 req/s so it would be pretty hard for a user restarting an application a bunch to hit that unless the application is going to break that limit on its normal startup anyways.

@CCP_FoxFour // Technical Designer // Team Tech Co

Third-party developer? Check out the official developers site for dev blogs, resources, and more.
Antice Eton
The Scope
Gallente Federation
#3 - 2015-01-26 18:08:20 UTC
CCP FoxFour wrote:


1) The rate limit is IP based not User-Agent. Please do send proper user-agents however so that we can monitor it's usage. This is mainly so that if we see something funny or bad happening we know who to contact. Our first course of action, assuming your application is not harming us, is to contact you in order to request you make any change we need. If we cannot contact you, or you don't respond within a reasonable amount of time, then we look at blocking things.


Thanks.

For the user agent, is there a particular standard that is expected? I figure I could just put my app name and github address but if there is a structure that makes it easier for your loggers then i'll happily use that.