(MAINT) try more debug

Author: davidmalloncares

.github/workflows/test-install-matrix.yaml

@@ -103,6 +103,9 @@ jobs:
       - name: Debug SSH connectivity
         run: |
           echo ::group::debug_ssh_connectivity
+            # Install sshpass first
+            sudo apt-get update && sudo apt-get install -y sshpass
+            
             # Check if containers are running
             docker ps
             
@@ -112,35 +115,58 @@ jobs:
               echo "SSH processes in $container:"
               docker exec $container ps aux | grep sshd || echo "No sshd processes found in $container"
               
-              # Test SSH connectivity directly
+              # Test SSH connectivity directly with password
               echo "Testing direct SSH connection to $container:"
-              container_ip=$(docker inspect $container | jq -r '.[0].NetworkSettings.IPAddress')
-              echo "Container IP: $container_ip"
-              
-              # Get the mapped SSH port
               ssh_port=$(docker port $container 22 | cut -d: -f2)
               echo "SSH port mapping: localhost:$ssh_port -> $container:22"
               
-              # Test SSH connection with timeout
-              timeout 10 ssh -o StrictHostKeyChecking=no -o ConnectTimeout=5 -p $ssh_port root@localhost 'echo "SSH connection successful"' || echo "SSH connection failed to $container"
+              # Test SSH connection with password
+              echo "Testing SSH with root/root:"
+              timeout 10 sshpass -p "root" ssh -o StrictHostKeyChecking=no -o ConnectTimeout=5 -p $ssh_port root@localhost 'echo "SSH connection successful to $container"' || echo "SSH connection failed to $container"
+            done
+          echo ::endgroup::
+
+      - name: Wait for SSH to be ready
+        run: |
+          echo ::group::wait_for_ssh
+            # Wait for SSH to be available on all containers via Bolt
+            for i in {1..12}; do
+              echo "Attempt $i: Testing Bolt SSH connectivity with root/root..."
+              
+              # Test with root/root credentials
+              if bundle exec bolt command run 'echo "Bolt SSH test successful"' \
+                 --inventoryfile ./inventory.yaml \
+                 --targets all \
+                 --connect-timeout 30 \
+                 --no-host-key-check \
+                 --user root \
+                 --password root; then
+                echo "All containers are accessible via Bolt SSH!"
+                break
+              fi
+              
+              if [ $i -eq 12 ]; then
+                echo "Containers failed to become accessible after 12 attempts"
+                echo "Final inventory check:"
+                cat ./inventory.yaml
+                echo "=== Docker container status ==="
+                docker ps
+                exit 1
+              fi
+              echo "Waiting 10 seconds before retry..."
+              sleep 10
             done
           echo ::endgroup::
+
       - name: Check container SSH configuration
         run: |
           echo ::group::container_ssh_config
             for container in $(docker ps --format "table {{.Names}}" | tail -n +2); do
               echo "=== SSH config for $container ==="
               docker exec $container cat /etc/ssh/sshd_config | grep -E "(PasswordAuthentication|PubkeyAuthentication|PermitRootLogin)" || true
-              
-              echo "=== Test password auth ==="
-              ssh_port=$(docker port $container 22 | cut -d: -f2)
-              # Try common passwords
-              for password in "root" "password" "litmus"; do
-                echo "Trying password: $password"
-                timeout 5 sshpass -p "$password" ssh -o StrictHostKeyChecking=no -o ConnectTimeout=5 -p $ssh_port root@localhost 'echo "Password auth successful with: $password"' && break
-              done || echo "Password authentication failed"
             done
-          echo ::endgroup::    
+          echo ::endgroup::
+
       - name: Install PE on test cluster
         timeout-minutes: 120
         run: |