Hi! I started receiving emails about workload spikes on my app yesterday evening. I am on the Starter Plan and it quickly used up all WUs. I allowed overages for a time, but quickly saw the charges accumulating. I disabled overages overnight and still was faced with more charges in the morning. I am not sure how that is possible. I’ve looked at the metrics and server logs, and am still stumped on the issues. I’ve emailed support and my email was transferred to a workload spike team and it could be awhile for a response. I was wondering if anyone here could help me identify the issues happening.
Are you having nested Repeating Groups ? Also what is the size of the data being displayed in your Repeating Group ? sounds like it’s something like “inventory” or something, What list were you trying to display in your Repeating Group ?
As far as I know, the list size will be multiplied by the number of rows to be displayed in your Repeating Group hence the WU spike.
Here is what the repeating group looks like in the editor. There was one user who was very active around the time of these spikes. This user has 480 items in her list. (It is an inventory/sales app). From the server logs it appears she was making updates to individual items. The list repopulates after every update so I can see how having it reload a lot could use up some WUs. I’m not seeing any actions in the server log that are more than 10 WUs. Most are 0.6 WUs. I’m not understanding where almost 800,000 WUs were used.
Sure! Here is the workflow when a user edits/updates an item’s info. There are probably more efficient ways of doing some of the things I have, but here is what I have currently. I appreciate the help!
I wonder if those ‘make changes total’ could be one step instead of running them after each ‘thing’ update.
I would like to test this because I’m not sure, but maybe DB Triggers would be an option for something that complex? Example: update other/blank only when “other/blank SKU.name-before change IS NOT other/blankSKU.name” or something.
Basically it’s checking to see if the value was actually changed before doing that workflow. If there are a bunch of ‘do searches’ in those ‘make change’ steps, that’s probably the reason.
Although, to spike to almost 800K runs, something seems off.
That is a good idea I’ll have to look into. Something definitely seems off however. This is my normal monthly workload usage and I’ve never been close to using all my Starter Plan’s WUs. And now January is right where I’d expect it to be with workflows, but suddenly ten times as many WUs being consumed. I hope I hear back from Bubble support soon.
Thanks for all the tips! I’m sure my app could be more efficient. I do have pages on that repeating group but I will look into some of the other ideas you mentioned! Thanks again
