Start a new topic
Solved

Facebook data import has many problems

Hi,


I experience a lot of problems with importing Facebook data to Google Data Studio lately. It is very slow to load, if it loads at all.. and besides that I just keep getting all these system errors in Google Data studio if I change the date. This is the case for multiple accounts and multiple dashboard I am making.

What causes the Facebook data import to be so slow? And what can we do to change this?


Kind regards,


Tim Mallant


Hi Tim,

Sorry to hear that you've been having issues with the Insights connector lately.


I checked the logs and found that you are getting errors for going over the page's request quota for Insights and Facebook is throttling your requests, which is why it's acting so slowly:

Rate limit error {"error":{"message":"(#32) Page request limit reached", "type":"OAuthException","code":32,"fbtrace_id":"xxxx"}}


This also seems to align with the usage we track as these seems abnormally high recently:

image


You can also use this API usage tools here:

https://supermetrics.com/tools/usageTracking



For more information about the #32 error, see "Page-level Rage Limiting" in this link:

https://developers.facebook.com/docs/graph-api/advanced/rate-limiting


As noted there, it can take up to an hour for the throttle to be lifted again and the page to get requests again. There's some general advice about avoiding this issue in the article. Generally you need to make the requests either smaller or trigger fewer in the same time span (may require some report re-design to do this).


-Thank you

Supermetrics Team

I can't tell if I'm being rated limited but I'm having similar issues. Pulling a few hundred thousand cells is very very slow. I understand that much data pulling is slow, but lately it's been more slow than usual. This is what I found in the use tracker:


image


Actually I answered my owned question when I looked at the api request MTD as compared to the average execution time. The more API requests to Facebook, the longer the execution time and it starts to spike in execution time at 1k+ API calls a day. 


image


Login or Signup to post a comment