"the sequence of write to table integrations is misconfigured" <-- any idea what might be causing this error? on my write to column at the end of this table, the hidden write to columns aren't affecting it, and there's only an issue writing to the Funded Accounts table
Checked funded accounts, can see any write to or look up loops back to funding imports that might be causing the error, there are some hidden write to columns inside funding imports, if they need to be deleted let me know, I just kept them there incase I need to revert back to a different flow (as the config took some time on those)
Ahh, okay. Yep, that may be the issue then, that would cause an "infinite loop" between tables so it's best to keep it in a single stream instead.
If you disconnect the loop, let me know if the issue continues.
can you double check this without me having to delete the columns
as I don't wanna do it and lose the column configs if it may not fix the issue
any luck?
Not so far, it's confusing haha. I couldn't find any loop back to the original table unless I'm missing one I have not checked? I have Funding Import -> Newly Funded -> Job Imports Any other tables that would be looped in...? I didn't see any writes in Emails/phones or Checker that would push data back into those other ones
Not that I'm aware of 🤷 I checked myself but couldn't find anything, working on other things right now, but this is blocking me
The data in "funding imports" is imported via csv, nothing else comes in to that table, the "enrich your data from" is legacy and not used, but even when it was, the rows were writing fine for all the write to columns in the "funding imports" table
Also, many of these columns have conditional formulas that are referencing columns that no longer exist in the tables, in case you'd like to update those while we check that other misconfiguration issue.
hey Nick S. - you're seeing this error not because of an infinite loop back to the original table, but because you have so many write to tables all connected together! tl;dr write to table is meant to be a one-off or two-off thing to pass data between a few tables. because it'll create a massive directed acyclic graph of updates that'll move from one table to another, we limit the number of times you can write from one table, to four other tables, each to two more tables, etc. somewhere around a dozen, i lost count of the number of other write to table columns that this one write to table column led to! this column is hitting that internal limit of write to table connections. to resolve this, i'd first eliminate redundant write to table actions (for example, you have 4 different writes to the same email and phone verifications table from the newly funded accounts table, and each of these write to table actions looks like they're doing the same thing, so those should just be 1 column). if that still isn't sufficient, you can output data to a tool like zapier, then send it back in via webhook to a different table!
Hey Thanks Colin, I'll take a look into this now, do you have much more info on other limitations of clay? I'd love to read about it in more detail if so, as they could affect the architectural decisions I'm taking right now
also if there's a way to see a top down view of all the relations between the tables myself that'd be great
hey nick - no other documentation on those limitations, and they're all subject to change. for example, for write to table - we're right now thinking through a huge batch of feature updates that will hopefully change / superpower the mappings between tables, and would likely affect that graph size limitation. but that's something that'll take many, many months to release. that feature update would be centered around better seeing the relations between tables too! sorry we don't have more up-to-date info here - i'll push the team to prioritize this batch of features, but it will be a huge engineering effort and will take a while
Thanks for the feedback Colin, just to be clear, do I need to clean up and delete the write to columns that are writing to the "newly funded accounts" table, or delete the write to columns the export data from "newly funded accounts"?
probably gonna need some assistance with this, as I don't want to unnecessarily delete write to columns (some are legacy) but there just in case I need to revert back to a different workflow, but if it means I need to delete them to enable this write to column from the "funding imports" table, so be it
I've deleted a fair few tables and columns now, down to trying to pinpoint the columns which might be causing the problems and remove them or figure out a patch to help me solve this limitation for now
ahh it's solved for now, thanks for your help Colin, In the end I had to just delete and add a new column once I cleaned up (refreshing the page didn't work to remove the error from the column)
glad it worked! keep me posted if you run into this again and i can help troubleshoot 🙂
.png){kind=logo}
