Hi, we're experiencing an issue where we can't `ku...
# generalg
gifted-breakfast-73755
09/02/2025, 4:15 PMHi, we're experiencing an issue where we can't
kubectl exec Copy code
...
2025/09/02 16:42:13 [ERROR] error syncing 'fleet-default/vrcvip-us-west-agc-managed-system-upgrade-dc9d0e': handler mcc-bundle: no chart version found for system-upgrade-controller-106.0.0, requeuing
...
2025/09/02 16:14:37 [INFO] RDPClient: Checking if dialer is built...
2025/09/02 16:14:37 [INFO] RDPClient: Dialer is not built yet, waiting 5 secs to re-check.
2025/09/02 16:14:42 [INFO] RDPClient: Checking if dialer is built...
2025/09/02 16:14:42 [INFO] RDPClient: Dialer is not built yet, waiting 5 secs to re-check
...
2025/09/02 16:42:30 httputil: ReverseProxy read error during body copy: http2: server sent GOAWAY and closed the connection; LastStreamID=267, ErrCode=NO_ERROR, debug=""
2025/09/02 16:42:30 httputil: ReverseProxy read error during body copy: http2: server sent GOAWAY and closed the connection; LastStreamID=267, ErrCode=NO_ERROR, debug=""
2025/09/02 16:42:30 httputil: ReverseProxy read error during body copy: http2: server sent GOAWAY and closed the connection; LastStreamID=267, ErrCode=NO_ERROR, debug=""b
bland-article-62755
09/02/2025, 4:19 PMAre these windows nodes?
g
gifted-breakfast-73755
09/02/2025, 4:21 PMNo, it's Ubuntu running Rancher on a single node docker install (which I know is not recommended and we are planning to migrate to K8s/K3s but haven't had time yet).
b
bland-article-62755
09/02/2025, 4:36 PMI meant the downstream clusters.
g
gifted-breakfast-73755
09/02/2025, 4:36 PMOh, those are Debian 12 nodes
b
bland-article-62755
09/02/2025, 4:37 PMI asked because the RDP is a windows thing.
bland-article-62755
09/02/2025, 4:37 PMSeems like something triggered a client and that's what the blocker is.
bland-article-62755
09/02/2025, 4:38 PMIt's probably not related to your issue then.
bland-article-62755
09/02/2025, 4:39 PMI'd check kubectl against the nodes directly. Edit the config so you're not using the rancher instance as a proxy and just see if you can
kubectl get nodesg
gifted-breakfast-73755
09/02/2025, 4:39 PMOk, this is the error we're seeing on the client side:
Copy code
Chads-MacBook-Pro-2:~ chad$ kubectl exec -it deploy/app -- bash
Error from server (InternalError): an error on the server ("<html>\r\n<head><title>502 Bad Gateway</title></head>\r\n<body>\r\n<center><h1>502 Bad Gateway</h1></center>\r\n<hr><center>nginx/1.21.0</center>\r\n</body>\r\n</html>") has prevented the request from succeeding (get deployments.apps app) Copy code
2025/09/02 16:39:38 [error] 207972#0: *30012 connect() failed (146: Connection refused) while connecting to upstream, client: 10.10.50.203, server: <http://rancher.vrcis.com|rancher.vrcis.com>, request: "GET /v3/connect HTTP/1.1", upstream: "<http://192.168.128.23:80/v3/connect>", host: "<http://rancher.vrcis.com|rancher.vrcis.com>"b
bland-article-62755
09/02/2025, 4:41 PMSo it sounds like port 80 is closed on your rancher instance.
bland-article-62755
09/02/2025, 4:43 PMnginx is returning a 502 because it' can't reach
<http://192.168.128.23/v3/connect>g
gifted-breakfast-73755
09/02/2025, 4:43 PMOk that may have been when I restarted it. I just started it back up and it automatically crashes with this in the logs:
gifted-breakfast-73755
09/02/2025, 4:43 PMUntitled.txt
b
bland-article-62755
09/02/2025, 4:46 PM6444bland-article-62755
09/02/2025, 4:46 PMThis is probably why docker isn't supported? IDK. I'd edit some of your original message and maybe someone else has an idea.
g
gifted-breakfast-73755
09/02/2025, 4:51 PMOk, will do. Thanks anyway.
19 Views