Open
Description
What happened?
When running upgrade-cluster.yml with env vars download_run_once: true and download_localhost: false the /tmp/kubespray_cache/ directory isn't created so the playbook fails on task Download_container | Copy image to ansible host cache.
What did you expect to happen?
Create cache directory
How can we reproduce it (as minimally and precisely as possible)?
Run upgrade-cluster.yml or cluster.yml with same env vars as listed below with ansible control node not being on the kubernetes node that will be chosen as download_delegate.
OS
Debian 12
Version of Ansible
ansible [core 2.16.14]
config file = /home/brbik/github/infra/kubernetes/kubespray/ansible.cfg
configured module search path = ['/home/brbik/github/infra/kubernetes/kubespray/library']
ansible python module location = /home/brbik/github/infra/kubernetes/kubespray/venv/lib/python3.13/site-packages/ansible
ansible collection location = /home/brbik/.ansible/collections:/usr/share/ansible/collections
executable location = /home/brbik/github/infra/kubernetes/kubespray/venv/bin/ansible
python version = 3.13.3 (main, Apr 9 2025, 07:44:25) [GCC 14.2.1 20250207] (/home/brbik/github/infra/kubernetes/kubespray/venv/bin/python3)
jinja version = 3.1.6
libyaml = True
Version of Python
Python 3.13.3
Version of Kubespray (commit)
Network plugin used
calico
Full inventory with variables
[all]
kube-master-dev-1 ansible_host=192.168.0.156
kube-master-dev-2 ansible_host=192.168.0.158
kube-master-dev-3 ansible_host=192.168.0.157
kube-worker-dev-1 ansible_host=192.168.0.155
kube-worker-dev-2 ansible_host=192.168.0.154
kube-worker-dev-3 ansible_host=192.168.0.153
[kube_control_plane]
kube-master-dev-1
kube-master-dev-2
kube-master-dev-3
[etcd]
kube-master-dev-1
kube-master-dev-2
kube-master-dev-3
[kube_node]
kube-worker-dev-1
kube-worker-dev-2
kube-worker-dev-3
[k8s_cluster:children]
kube_control_plane
kube_node
Command used to invoke ansible
ansible-playbook -i inventory.ini -u root upgrade-cluster.yml -e download_run_once=True download_localhost=False
Output of ansible run
TASK [download : Download_container | Copy image to ansible host cache] ***************************************************************************************************************************************************************************************
fatal: [kube-master-dev-1]: FAILED! => {"changed": false, "cmd": "/usr/bin/rsync --delay-updates -F --compress --archive --rsh='/usr/bin/ssh -S none -o StrictHostKeyChecking=no -o UserKnownHostsFile=/dev/null -o ControlMaster=auto -o ControlPersist=30m -o ConnectionAttempts=100 -o UserKnownHostsFile=/dev/null' --out-format='<<CHANGED>>%i %n%L' [email protected]:/tmp/releases/images/quay.io_calico_node_v3.29.3.tar /tmp/kubespray_cache/images/quay.io_calico_node_v3.29.3.tar", "msg": "Warning: Permanently added '192.168.0.156' (ED25519) to the list of known hosts.\r\nrsync: [Receiver] change_dir#3 \"/tmp/kubespray_cache/images\" failed: No such file or directory (2)\nrsync error: errors selecting input/output files, dirs (code 3) at main.c(829) [Receiver=3.2.7]\nrsync: [Receiver] write error: Broken pipe (32)\n", "rc": 3}
Anything else we need to know
No response