Though I guess since the sandbox is going away in favor of test accounts, none of what I just posted matters now unless the limits for test accounts are different than those for live accounts.
On Wednesday, November 21, 2012 8:12:34 AM UTC-5, Jeremy Aube wrote: > > There are two types (rate names) of RateExceeded errors: > OperationsPerMinute and RequestsPerMinute. If you're getting > RequestsPerMinute, waiting 30 seconds to a minute should resolve the issue > every time. > > There is a problem with the Sandbox, however, where it's possible to > trigger OperationsPerMinute (RateExceeded) even though you should really be > getting a SizeLimitError. The issue, I believe, is that the SizeLimitError > threshold is higher than the OperationsPerMinute threshold. > > The reason this is a problem is that it's possible to hit > OperationsPerMinute on every request, regardless of how long you wait. For > example, let's say the SizeLimitError threshold is 10,000 operations (more > or less), and the OperationsPerMinute threshold is 5,000 operations. If I > send a request with 7,000 operations, I'm going to keep getting an > OperationsPerMinute error, with a recommendation to wait 30 seconds. It > doesn't matter how many times I wait 30 seconds, however, because my single > request exceeds the OperationsPerMinute threshold. This is the problem with > the current limits on Sandbox. > > If the errors thresholds were reversed, there wouldn't be a problem. Now > when I send my 7,000 operation request, I get a SizeLimitError, which > immediately lets me know I should reduce the size of the request. If I send > a single request with 10,000 operations, I still get the SizeLimitError, so > I can do the right thing. I would only get the OperationsPerMinute error > if, say, I was sending 3 request with 4,000 operations each, in which case > waiting 30 seconds would be the right course of action. > > Currently in sandbox, if I get the OperationsPerMinute error, I make > another request 30 seconds later, then one more request 30 seconds after > that if I get the same error. If I get that error a third time, then I know > I'm in between thresholds and need to reduce the number of requests. Or I > just arbitrarily limit the number of requests to a size that I know has > worked well in the past. > > On Friday, November 16, 2012 9:03:01 AM UTC-5, mm wrote: >> >> Hey Kelly/Anash - >> In my experience, I got a RateExceeded error for the first call I made to >> create campaigns. I discovered that if I limited the number of operations >> in the call to some very small number (my production limit is 1000, but in >> this case I think I had to go to 5), the call succeeded. So I believe that >> the notion of 'don't use the sandbox for load testing' extends to the >> volume of data you attempt to create in each call also. It would be nice to >> get a more appropriate error, but it's not clear where/how to file a bug >> report for the sandbox. >> >> HTH - >> mm >> > -- -- =~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~ Also find us on our blog and discussion group: http://adwordsapi.blogspot.com http://groups.google.com/group/adwords-api =~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~ You received this message because you are subscribed to the Google Groups "AdWords API Forum" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/adwords-api?hl=en
