Connection reset by peer error

ivan.lovrencic

Hi everyone,

I am trying to extract large amounts of market depth data from TRTH API, and I use python as my API client. After the data is extracted, I am streaming the data (in an effort to conserve memory), but during the stream, I often get the Connection reset by peer error. I assume this is because the server closes the connection. Because of this, I am unable to actually extract and parse larger amounts of data.

Is there a workaround for this? What would be the best way to extract large amounts of data from the TRTH API? Should I create multiple jobs and split the API call into more API jobs with smaller time intervals?

Thank you in advance!

zoya faberov

Hello @ivan.lovrencic ,

If the data set that you are downloading is truly large, if you are consistently interrupted on specific very large requests, I would suggest: partitioning the very large requirement, usually this involves dividing your required time interval into sub-intervals, on request. Also you may wish to look into Key Mechanisms -> Server Driven Paging, this can be done on the result, see the pagination as required (this can not be parallelized as requests could be) also the page size adjustment is described toward the bottom, as by default it is 250 records.

If you are observing something that is out of the ordinary (such as all requests are failing all of a sudden) , the first place that I check is Service Alerts on My Refinitiv- this case save you time.

If the issue is re-occurring, but is not consistent, it may be helpful to investigate the specific incident, when it happens, including the service status, with the product support, as discussed in this previous discussion thread via Refinitiv Helpdesk Online -> Product -> RTH