In short, the sum total of our WU pie chart does not add up to the reported total WU for our environment. The pie chart under-counts by more than 50%. Hereās a video below showing this as well as how drilling into a pie-chart wedge results in a similar discrepancy where the sum total of the drill-down pie chart does not equal the size of the wedge we started from.
Anyone else seeing this? If so, please share your discrepancy numbers if possible - thank you!
How about if you drill in - whatās that comparison looking like for a subgraph relative to itās total reported wedge value?
Sorry - my API section drill down shows 663 WU but the high level overview shows 1,953, so yes, weāre seeing the same thing 
Iād submit a bug report as well but Iām still waiting since three weeks ago for Bubble to explain this WU issue 
Oh, is it possible that the drilled-down chart only displays the top 10 workflow uses? I think thatās whatās going on here. Bad UX (it would be nice to have an āotherā section or just let us see every usageā¦)
Edit: then again, I see that the workflow drilldown shows way more than 10 and has the āOtherā wedgeā¦
Yikes! Itās possible that theyāre just limiting the number of slices being shown. But as you say, they have an āOtherā category. Iāll file a bug report that references this threadā¦ need to have an official statement.
I never thought the new WU unit pricing, with super expensive pricing that is completely untethered from actual cost basis, would somehow NOT be the biggest issue with the new WU pricing model, but given the lack of admin controls, notifications triggers not working, and the actual calculations being incorrect and not fully visibleā¦
One more data point hereā¦ the numbers arenāt even close.
Thanks for filing the bug report, @zelus_pudding. Iām guessing there is an explanation for what we are seeing, and I am now super curious to find out what that explanation is.
Yeah, thatās wild! Bubble has seen our report and itās been triaged to their tier 2 support team. Details to come.
Hi @zelus_pudding - Thanks for sharing what youāre seeing. We have seen a few reports of this come in where the top metric above the bar graph is correct and accurately represents the total workload units, but the drill down is not correct and is missing workload units. Specifically:
In both cases, the top level metric is correct but the pie chart is not. Our team is working on getting a fix together for this, as we know not having accurate metrics when drilling down into workload unit consumption can make it challenging to optimize your app and address any new issues that might need investigating. If you have specific questions about your app and workload units, please reach out to our team and we can take a look. Iāll follow up here when a fix is in place.
It is concerning. How does one debug / optimize correctly when half or more of the data does not show. I wish Bubble would prioritize getting these things correct, everything with the new pricing rollout bar the notice period has been half baked and continues to be.
Ohā¦ This always happens when I click a specific day in the month view then try and drill down. I just thought it was expected behaviour that we canāt drill down on specific days
We are prioritizing this - our team is working on it right now. If you have a specific question you canāt get the answer to right now because of this issue, please get in touch with support.
This topic was automatically closed after 14 days. New replies are no longer allowed.
Hi all - Apologies for the tardy update here! We pushed out a fix for this on December 20 and then were monitoring things over the holiday season and I neglected to update this thread. Pie charts post December 20 are displaying data correctly and they should match what you see in the bar chart at the top of your logs page. Unfortunately our fix did not retroactively update charts from before December 20, so if things are incomplete when analyzing longer periods of time, thatās why - as a result I recommend looking at the last week or so of data when trying to analyze trends if youāre diving in to workload unit analysis! As always, please reach out to our team (our new chatbot is pretty incredible) if you see anything that you need us to further investigate.
