Closed FabioAntunes closed 3 months ago
Looks like you're running into RequestInProgressException. This happens when the issue cert and get cert calls are too close to each other. If there's a way you can slow down your issuance on your end that'd be a good mitigation, I will prioritize handling this exception in the source code as well.
Thanks for the quick reply! Would this always be an issue? We are using this along with istio-csr if we have a surge on pod creation would we always be in this position? If so this partnership between istio-csr and aws-privateca might not be feasible for us. We maybe should use this project for our ingress certificates and then use regular cert-manager issuer for istio.
On Thu, 11 Jul 2024 at 23:19, Brady Siegel @.***> wrote:
Looks like you're running into RequestInProgressException. This happens when the issue cert and get cert calls are too close to each other. If there's a way you can slow down your issuance on your end that'd be a good mitigation, I will prioritize handling this exception in the source code as well.
— Reply to this email directly, view it on GitHub https://github.com/cert-manager/aws-privateca-issuer/issues/329#issuecomment-2224045750, or unsubscribe https://github.com/notifications/unsubscribe-auth/AATNIIM2XXCB5LO6QKOUGM3ZL4AF5AVCNFSM6AAAAABKXYRNTSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDEMRUGA2DKNZVGA . You are receiving this because you authored the thread.Message ID: @.***>
So this will be an issue until we fix it. We are always open to a contribution from the community as well!
What I meant was: if the problem was on the aws side with the rate limiting and therefore it's an upstream issue, or is this on the aws-privateca-issuer
?
It's a client exception that the aws-privateca-issuer
should just retry on when trying to drive a certificate to a completed state.
This is to say we can support your use case when we prioritize the fix here.
Per @bmsiegel -- The fix would be here: https://github.com/cert-manager/aws-privateca-issuer/blob/main/pkg/aws/pca.go#L80-L87
Probably want to just update the client config to retry if RequestInProgress Exception is encountered
The fix for this is included in the release: https://github.com/cert-manager/aws-privateca-issuer/releases/tag/v1.3.0. Please reopen if you're still seeing this issue.
Describe the expected outcome
Create certificates.
Describe the actual outcome
We are getting the following errors can someone point us what could be the possible root cause?
This was fresh installation on a cluster we were on version 1.2.5 and even after updating to 1.2.7 the issue persists.
The pod starts fine and signs a couple of certificates and after a while it just keeps on erroring:
Are we being rate limited by the AWS?
Steps to reproduce
No response
Relevant log output
No response
Version
v1.2.7
Have you tried the following?
Category
Supported Workflow Broken
Severity
Severity 1