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
Each invocation of cloudexec creates a new entry in blob storage even if only one line changes. This probably is inconsequential normally but I had a binary in my local directory and very quickly stored a gig in blob storage.
The text was updated successfully, but these errors were encountered:
First thought: hash the zip and don't upload it if it's the same as a previous upload.. Obviously the whole repo will be re-zipped every time, but we could at least skip the upload step.
Second thought: we could check the git status and re-zip everything if the commit changes or if there are uncommitted changes in any file besides cloudexec.toml.
Either way, we'll need to add some kind of repo info to the job's state + the ability to use a previous job's input + logic for replacing the previous job's input with a new cloudexec.toml.
Each invocation of cloudexec creates a new entry in blob storage even if only one line changes. This probably is inconsequential normally but I had a binary in my local directory and very quickly stored a gig in blob storage.
The text was updated successfully, but these errors were encountered: