Closed czunker closed 6 months ago
Just a thought: Or does this error happen during the communication to the k8s API?
But the default k8s client has a default retry, IIRC. It could be, the retry doesn't not happen for these errors.
Some other mentions of EOF errors: https://github.com/golang/go/issues/53472 Some of the linked issues: https://github.com/Dynatrace/dynatrace-configuration-as-code/pull/1143 https://github.com/0chain/gosdk/pull/915
As just discussed with @imilchev , this might be a race condition. It can happen that a provider is shutting down, and a new scan tries to use the provider instead of creating a new one.
Fixed in mondoohq/cnquery#3775
Describe the bug The logs show different errors when the Scan API tries to send data upstream.
Strange is also, that the asset has no name. Perhaps that is also part of the problem.
To Reproduce
??? Need to figure this out.
Steps to reproduce the behavior:
Expected behavior cnspec should be able to talk to upstream.
The asset should have a name.
Desktop (please complete the following information):