Reviewing our logs we can see that in the last 10 days we have started to get 429s back for hitting a rate limit, which I can’t see that we’ve experienced before.
Nothing in our request approach has changed, we submit approx. 10,000 instruments each day in asynchronous batches of 100, which generally takes around 1 min – which would I guess put us well under the formal rate limit (200) but above the advised top end (60)?
Our re-try approach – which is configurable – is a 5 re-try but with a geometric sequence ^2 so should provide breathing space
What seems to be happening recently is we are getting a max tries and timing out.
Are you able to review our usage and advise what we are breaching on your side, or if you are having recent performance issues? We have plenty of options on our side to remediate, although our time window is very small.
This is the endpoint we call just after 22.30 UK Time each night M-F:
https://selectapi.datascope.refinitiv.com/RestApi/v1/Extractions/ExtractWithNotes
"@odata.type": "#DataScope.Select.Api.Extractions.ExtractionRequests.EndOfDayPricingExtractionRequest",
