You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'm trying to install a large R package (~65MB) using sqlmlutils onto SQL Server 2019 CU1. I get the following error when doing so: Error: Installation of packages failed with error C stack usage 138399847 is too close to the limit
I can run install.packages() and install the package against a local R installation; that works fine. I was also able to run install.packages() within sp_execute_external_script in SQL Server 2017.
It looks like this issue is related to package size. Almost all of the package size is due to a serialized neural network model. If I strip that out, the package size goes from 67MB to 156KB and installation completes successfully. I'd like to keep the neural network model together with this package, so splitting it out is not a great long-term solution for me.
The text was updated successfully, but these errors were encountered:
Hi,
We are investigating why this is failing. We do have size limitations, but it should not be the case here. I'll follow up with you offline @feaselkl . Another workaround in the meantime is using TSQL and create external library.
Thanks,
N
I'm trying to install a large R package (~65MB) using sqlmlutils onto SQL Server 2019 CU1. I get the following error when doing so:
Error: Installation of packages failed with error C stack usage 138399847 is too close to the limit
I can run install.packages() and install the package against a local R installation; that works fine. I was also able to run install.packages() within sp_execute_external_script in SQL Server 2017.
It looks like this issue is related to package size. Almost all of the package size is due to a serialized neural network model. If I strip that out, the package size goes from 67MB to 156KB and installation completes successfully. I'd like to keep the neural network model together with this package, so splitting it out is not a great long-term solution for me.
The text was updated successfully, but these errors were encountered: