Uh oh!
There was an error while loading. Please reload this page.
Normalize handling 403 for rate limits on 'dataset.create()'. - #2089
Conversation
| # We need to wait to stay within the rate limits. | ||
| # The alternative outcome is a 403 Forbidden response from upstream. | ||
| # See: https://cloud.google.com/bigquery/quota-policy |
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
Uh oh!
There was an error while loading. Please reload this page.
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
Uh oh!
There was an error while loading. Please reload this page.
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
Uh oh!
There was an error while loading. Please reload this page.
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
Uh oh!
There was an error while loading. Please reload this page.
A 429 ("Too many requests") would be way more appropriate than a 403
("Forbidden"), but there is nothing we can do about it here.
Addresses:
#2089 (comment)dhermes
commented
Aug 11, 2016
We could deal with backend changes if / when tests fail. You can make it a bit less brittle via: 'rate limit'inerror.message.lower() |
In 10e4f51, I figured out that we capture the JSON error structure, which means we can use the documented |
Exclude Forbidden exceptions which do not have 'rateLimitExceed' as the reason for one of their errors. Addresses: #2089 (comment).
dhermes
commented
Aug 11, 2016
Nice find! LGTM |
* fix: Transformers with non-standard column names through errors * fix
Fixes#2075.