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
Because Cloud Run containers only have an in-memory filesystem, I thought it would make sense to just pipe the whole PDF through convert instead of saving anything to file. Turns out that was a bad idea as ImageMagick takes a lot of memory during conversion, probably because it keeps every page with full resolution in memory until it is finished.
This will easily OOM kill the container if the PDF has more than a few pages.
The text was updated successfully, but these errors were encountered:
Because Cloud Run containers only have an in-memory filesystem, I thought it would make sense to just pipe the whole PDF through
convert
instead of saving anything to file. Turns out that was a bad idea as ImageMagick takes a lot of memory during conversion, probably because it keeps every page with full resolution in memory until it is finished.This will easily OOM kill the container if the PDF has more than a few pages.
The text was updated successfully, but these errors were encountered: