Closed paddatrapper closed 2 years ago
You don't have any retry strategy setup right ?
For now I could suggest you to try setup AWS_MAX_ATTEMPTS and AWS_RETRY_MODE to mitigate this, but I'll have a look on how I could setup a default retry strategy for this. https://boto3.amazonaws.com/v1/documentation/api/latest/guide/retries.html
Re queuing might be a good idea too.
It's more that it seems to crash the thread and so prevent any future uploads from happening
It's more that it seems to crash the thread and so prevent any future uploads from happening
Oh right, yes, this is bad.
Sorry, probably should have worded the ticket better
No worries, I think this issue raises a bigger problem, this tool is not capable of crashing and recovering as we create a new tmp dir at every start up. This makes the life of sysadmins a bit more difficult to even re upload the files by hand. For the local archive storage this wasn't really a topic, but with s3 it is.
Here is what should be done:
This should also help for unexpected shutdowns. But this should have a dedicated issue.
I'll push some quick fixes, but the big missing part will need some work.
Sounds good. I can also ask @loydbanks to implement some of this next week if you want?
I already have a branch for this, it's alright. But thanks.
Solved with #132
It seems that this exception isn't handled properly