You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The swipnet project is servicing ~13000 requests/minute for items, but the target servers can only really handle ~500 concurrent items. As a result, most of those 13000 requests are walking away empty.
Furthermore, the "Tracker rate limiting is active" message does not seem to communicate what we thought it'd communicate. People tend to see it as an error rather than a "slow down" notice.
I think making some more global information available might help:
[10:04:29] <yipdw> or maybe just making the requested:granted ratio public
[10:05:02] <yipdw> will help illustrate that the right thing to do in a rate-limiting situation is not to drive the request rate higher
[10:05:07] <arkiver> maybe in the upper right corner
[10:06:01] <yipdw> it's possible that one motivation to start up a billion pipelines is lack of global information
[10:06:16] <yipdw> if we provide that maybe it'll help illustrate what "3300% of demand" means
We could also return this information as part of the "rate limited" response, which would make it possible to display that information in the "tracker rate limiting is active" message.
The text was updated successfully, but these errors were encountered:
The swipnet project is servicing ~13000 requests/minute for items, but the target servers can only really handle ~500 concurrent items. As a result, most of those 13000 requests are walking away empty.
Furthermore, the "Tracker rate limiting is active" message does not seem to communicate what we thought it'd communicate. People tend to see it as an error rather than a "slow down" notice.
I think making some more global information available might help:
We could also return this information as part of the "rate limited" response, which would make it possible to display that information in the "tracker rate limiting is active" message.
The text was updated successfully, but these errors were encountered: