https://prefect.io
Join Slack
One more question, in the docs > A result subclass that matches the storage backend of your `pref...
# ask-community
d
One more question, in the docs
A result subclass that matches the storage backend of your 
prefect.config.flows.storage
 setting will automatically be applied to all tasks, if available; notably this is not yet supported for Docker Storage
Does this mean we can not use checkpointing if using Docker Storage?
d
Hey @Damien Ramunno-Johnson! You absolutely can, you’ll just need to configure a 
Result
class for your flow
There isn’t a “sensible default” for Docker
Whereas for S3 Storage, for example, it makes sense to also configure an S3 Result
d
That was my first thought, but when I added it wasn’t working, but looking at it again it seems I added the results to the Docker() section, and not the flow section
d
The 
Result
class is intended to be set on the 
Flow
object 👍
d
Yeah, just missed a () 🙂
So files are being uploaded but when I try 
@task(nout=4, cache_for=datetime.timedelta(hours=12))
it doesn’t seem to use the cache and still reruns. Does the flow need to succeed for the cache to work?
🤔 1
was running it locally via flow.run() and not an agent
d
Hey Damien sorry I missed this
Based on this doc, I believe that you’ve got everything set correctly
Perhaps try using a file-based cache for running with 
flow.run
?
AH! 
Note that when running Prefect Core locally, your Tasks' cached states will be stored in memory within prefect.context
I believe you’ll need to run with an agent to experience what you’re expecting here
d
Thanks, I did get the in memory cache to work If I ran the same flow more than once (in the same session).
d
Glad to hear it!
Apologies, I just saw your DM
d
No worries, I am sure you are busy 🙂
d
For what it’s worth, I tend to load my flow file into ipython with 
ipython -i
and then do everything with it in there (including updating tasks by copy/paste if I want to hold onto my session)
d
Yeah, that makes sense. I think we will go ahead and make a feature request request as well later as well. We have enjoyed being able to have breakpoints in the tasks (which I don’t think is possible with the agent) and some tasks we have are longer and has been nice to both continue them on a larger machine, or after needing to return to it etc. Thanks for the help
9 Views
Open in Slack
PreviousNext
Powered by

Prefect Community

Bring your towel and join one of the fastest growing data communities. Welcome to our second-generation open source orchestration platform, a completely rethought approach to dataflow automation.

Powered by