-
Bug
-
Resolution: Done
-
Major
-
4.12
-
Important
-
None
-
SDN Sprint 231, SDN Sprint 232
-
2
-
Proposed
-
False
-
This is a clone of issue OCPBUGS-6651. The following is the description of the original issue:
—
Description of problem:
When running a hypershift HostedCluster with a publicAndPrivate / private setup behind a proxy, Nodes never go ready. ovn-kubernetes pods fail to run because the init container fails. [root@ip-10-0-129-223 core]# crictl logs cf142bb9f427d + [[ -f /env/ ]] ++ date -Iseconds 2023-01-25T12:18:46+00:00 - checking sbdb + echo '2023-01-25T12:18:46+00:00 - checking sbdb' + echo 'hosts: dns files' + proxypid=15343 + ovndb_ctl_ssl_opts='-p /ovn-cert/tls.key -c /ovn-cert/tls.crt -C /ovn-ca/ca-bundle.crt' + sbdb_ip=ssl:ovnkube-sbdb.apps.agl-proxy.hypershift.local:9645 + retries=0 + ovn-sbctl --no-leader-only --timeout=5 --db=ssl:ovnkube-sbdb.apps.agl-proxy.hypershift.local:9645 -p /ovn-cert/tls.key -c /ovn-cert/tls.crt -C /ovn-ca/ca-bundle.crt get-connection + exec socat TCP-LISTEN:9645,reuseaddr,fork PROXY:10.0.140.167:ovnkube-sbdb.apps.agl-proxy.hypershift.local:443,proxyport=3128 ovn-sbctl: ssl:ovnkube-sbdb.apps.agl-proxy.hypershift.local:9645: database connection failed () + (( retries += 1 ))
Version-Release number of selected component (if applicable):
4.12
How reproducible:
Always.
Steps to Reproduce:
1. Create a publicAndPrivate hypershift HostedCluster behind a proxy. E.g" ➜ hypershift git:(main) ✗ ./bin/hypershift create cluster \ aws --pull-secret ~/www/pull-secret-ci.txt \ --ssh-key ~/.ssh/id_ed25519.pub \ --name agl-proxy \ --aws-creds ~/www/config/aws-osd-hypershift-creds \ --node-pool-replicas=3 \ --region=us-east-1 \ --base-domain=agl.hypershift.devcluster.openshift.com \ --zones=us-east-1a \ --endpoint-access=PublicAndPrivate \ --external-dns-domain=agl-services.hypershift.devcluster.openshift.com --enable-proxy=true 2. Get the kubeconfig for the guest cluster. E.g kubectl get secret -nclusters agl-proxy-admin-kubeconfig -oyaml 3. Get pods in the guest cluster. See ovnkube-node pods init container failing with [root@ip-10-0-129-223 core]# crictl logs cf142bb9f427d + [[ -f /env/ ]] ++ date -Iseconds 2023-01-25T12:18:46+00:00 - checking sbdb + echo '2023-01-25T12:18:46+00:00 - checking sbdb' + echo 'hosts: dns files' + proxypid=15343 + ovndb_ctl_ssl_opts='-p /ovn-cert/tls.key -c /ovn-cert/tls.crt -C /ovn-ca/ca-bundle.crt' + sbdb_ip=ssl:ovnkube-sbdb.apps.agl-proxy.hypershift.local:9645 + retries=0 + ovn-sbctl --no-leader-only --timeout=5 --db=ssl:ovnkube-sbdb.apps.agl-proxy.hypershift.local:9645 -p /ovn-cert/tls.key -c /ovn-cert/tls.crt -C /ovn-ca/ca-bundle.crt get-connection + exec socat TCP-LISTEN:9645,reuseaddr,fork PROXY:10.0.140.167:ovnkube-sbdb.apps.agl-proxy.hypershift.local:443,proxyport=3128 ovn-sbctl: ssl:ovnkube-sbdb.apps.agl-proxy.hypershift.local:9645: database connection failed () + (( retries += 1 ))
To create a bastion an ssh into the Nodes See https://hypershift-docs.netlify.app/how-to/debug-nodes/
Actual results:
Nodes unready
Expected results:
Nodes go ready
Additional info:
- clones
-
OCPBUGS-6651 Nodes unready in PublicAndPrivate / Private Hypershift setups behind a proxy
- Closed
- is blocked by
-
OCPBUGS-6651 Nodes unready in PublicAndPrivate / Private Hypershift setups behind a proxy
- Closed
- links to