-
Notifications
You must be signed in to change notification settings - Fork 717
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
kubeadm on version 1.29.1 with upload-certs times out during the configMap POST #3022
Comments
can you show the full output this command, you can make it --v=5
|
in the past we have seen users complain that a single API call fails for them. but that's a problem on the host, for example the kube-apiserver restarted exactly at that moment.
and what happens if you just call kubeadm init, does it finish? |
this is just a host alias? do other API calls succeed reaching the server trough this, or only the upload-certs API calls fail? |
Yes, I've stated this in the issue too. Without |
Yes, it's just an FQDN pointing at the server and resolves fine. Also, only the upload-certs call fails. Everything before it succeeds, so the other calls reach the server fine. Also, a 504 without a proxy in the middle just means the request timed out and the api server returned the 504. |
please share the log |
Here's the entire dump. I've removed the endpoint FQDN just incase this time edited (neolit123) to collapse the log:
|
thanks for the logs, my responses here are the same as on the kubernetes ticket. some kubeadm API calls have retries, but some do not. the CreateOrUpdateConfigMap does not have retries. we could add retries to a number of those remaining functions in kubernetes master branch, but i don't think such a patch should be backported to 1.29 or earlier versions, so you have to understand what's going on your side and workaround. |
I'll do some more research / debugging and see if I can come up with something else, or atleast a workaround. Thanks. |
let's continue the discussion here: please post your results there if you find anything. |
What keywords did you search in kubeadm issues before filing this one?
kubeadm init upload-certs failure, kubadm init configmap timeout, kubeadm init client-side throttling
I found an existing issue, but it is closed. I've also commented there: kubernetes/kubernetes#112411.
Is this a BUG REPORT or FEATURE REQUEST?
Choose one: BUG REPORT
On a new Debian 12 bare metal server, I'm trying to bootstrap a cluster with kubeadm 1.29.1. If I pass
--upload-certs
, I get the timeout error. If I don't pass it, everything works fine.Here is what I run:
When it tries to post the configMap, it times out with:
Everything before it works fine (all the api calls succeed and return and HTTP 201). What's interesting is that there's no waiting taking place for 10 seconds. All this happens in a handful of seconds (it's a 6 core machine, so everything's pretty fast).
Versions
kubectl version
Client Version: v1.29.1
Kustomize Version: v5.0.4-0.20230601165947-6ce0bf390ce3
The connection to the server localhost:8080 was refused - did you specify the right host or port?
kubeadm version
kubeadm version: &version.Info{Major:"1", Minor:"29", GitVersion:"v1.29.1", GitCommit:"bc401b91f2782410b3fb3f9acf43a995c4de90d2", GitTreeState:"clean", BuildDate:"2024-01-17T15:49:02Z", GoVersion:"go1.21.6", Compiler:"gc", Platform:"linux/amd64"}
systemctl --version
systemd 252 (252.22-1~deb12u1)
+PAM +AUDIT +SELINUX +APPARMOR +IMA +SMACK +SECCOMP +GCRYPT -GNUTLS +OPENSSL +ACL +BLKID +CURL +ELFUTILS +FIDO2 +IDN2 -IDN +IPTC +KMOD +LIBCRYPTSETUP +LIBFDISK +PCRE2 -PWQUALITY +P11KIT +QRENCODE +TPM2 +BZIP2 +LZ4 +XZ +ZLIB +ZSTD -BPF_FRAMEWORK -XKBCOMMON +UTMP +SYSVINIT default-hierarchy=unified
Environment:
kubectl version
): 1.29.1uname -a
): Linux 6.1.0-13-amd64 kubeadm join on slave node fails preflight checks #1 SMP PREEMPT_DYNAMIC Debian 6.1.55-1 (2023-09-29) x86_64 GNU/LinuxWhat happened?
See above, but I get:
when I pass
--upload-certs
to kubeadm init.What you expected to happen?
That the cluster for bootstrap like it does when I don't pass
--upload-certs
.How to reproduce it (as minimally and precisely as possible)?
Install kubeadm and all the basic stuff on https://kubernetes.io/docs/setup/production-environment/tools/kubeadm/install-kubeadm/.
Then run
kubeadm init --upload-certs
.Anything else we need to know?
The problem is exactly the same as kubernetes/kubernetes#112411, although I don't think
systemd
is the problem.The text was updated successfully, but these errors were encountered: