able to deploy issues with models

ansible-playbook.yml

@@ -1,13 +1,13 @@
 ---
 - name: Deploy Ollama + Open WebUI 
-  hosts: your_gpu_servers
+  hosts: all
   become: yes
 
   vars:
     ollama_port: 11434
     open_webui_port: "{{ lookup('env', 'OPEN_WEBUI_PORT') | default('3000') }}"
     ollama_volume_name: ollama_ollama
-    open_webui_volume_name: ollama_open_webui
+    open_webui_volume_name: ollama_open-webui
     webui_secret_key: "vk4HCy3FnlF0fqRT8MClJSGVmEdsP2tK"   # ← move to ansible-vault in production!
 
   tasks:
@@ -26,7 +26,6 @@
         state: started
         restart_policy: unless-stopped
         pull: true
-        tty: true
         published_ports:
           - "{{ ollama_port }}:11434"
         volumes:
@@ -36,11 +35,6 @@
             count: -1               # -1 = all GPUs  (or use device_ids: ['0', '1'] for specific)
             capabilities:
               - - gpu               # nested list as per docs
-        comparisons:
-          image: strict
-          env: strict
-          volumes: strict
-        # container_default_behavior: no_defaults   # optional, modern default
       register: ollama_deploy
 
     - name: Deploy Open WebUI container