Hi there, we are running an on-prem Metaflow inst...
# ask-metaflow
a
Hi there, we are running an on-prem Metaflow instance and nowadays even with Argo. Local execution and "--with kubernetes" exceution of our flows work pretty well. With the recent introduction of Argo into our workflow the jobs behave "shakey". Flow execution randomly breaks with following error message: "*Error (exit code 1): failed to put file: The Access Key Id you provided does not exist in our records.*" The interesting part is: This error might disappear if we retry the step without changing anything. If we deploy the flow with "--retry" the workflow will eventually finish but with totally random amount of retries per each task. The code itself seems to function as expected but Argo fails to continue totally random. Has anybody an idea whats going on? Any help appreciated
1
To whom it may concern: It seems the issue was related to a residual information in configmap which lead argo to search for an minio instance that does not exist. After removing this residual (a left over from a failed installation) everything worked as expected.