I would like to know the root cause when I send a request to these API and sometimes it's got the message something like " rdp_requester.py - n time retry. wait for x sec. " I wonder why this message appears and how can I avoid it?
I would like to know the root cause when I send a request to these API and sometimes it's got the message something like " rdp_requester.py - n time retry. wait for x sec. " I wonder why this message appears and how can I avoid it?
Please provide examples of which APIs you are calling?
If the above .py can be shared, please rename and attach it as a .txt file - after removing any credentials such as username, password etc.
Hi umer,
here below is the API that I used
https://api.refinitiv.com/data/quantitative-analytics-curves-and-surfaces/v1/curves/zc-curves
https://api.refinitiv.com/data/quantitative-analytics-curves-and-surfaces/v1/curves/forward-curves
Are you suggesting when you go to these API pages and try to get the data, you see these errors? Or are you using an example application that uses these APIs? If so, what application are you using? Or did you attempt to create an applications that uses these APIs?
Looking at the logs you submitted, it appears the IPA service is complaining about too many concurrent requests. Was this the error you were referring to in your original post? Using the same code, have you tried to limit the request load/frequency to see if it helps?
I just ran my example Article.RDPLibrary.Python.ZeroCouponCurves which uses both the above API endpoints and I did not experience any retry errors.
I also converted the main Jupyter Notebook to a .py file (removed the charts etc) and ran that from the command line and again I do not see the above issue.
Is this all zc-curve/forward curve request that gives you the retry error - or just some specific complex ones?
Hi @umer.nalla
I have found this error in a specific complex situation in my app on SELF PPE, I have the 15 curves/payload and I send 2 requests by using a concurrent method (asyncio) by this case sometimes I got an error as I said in the beginning here below is the log that i found an error
Thanks for providing the additional information that you are using concurrency -this now puts the problem into context.
If you are able to share a code snippet showing exactly what you are doing - you don't have to necessarily include the full curve request definition, but the concurrency related code would be useful to understand what you are doing.
Hi @umer.nalla
here is my sample of the code that I used concurrency I combine 15 curves in payload and send it to the endpoint sometimes it just appears the error message
Is there a reason why you are using ppe endpoints? Have you tried using the async version of the financial-contracts endpoints?
Yeah it's my own code the reason I using ppe endpoints is curves on the production is not ready yet so I need to use ppe. I also ask on the qps team about the limitation of API it seems like I can send 1 request/sec. Thanks for your all help