fix: add opt-in net.ifnames=0 for Azure images

README.md

@@ -137,6 +137,18 @@ Default: `true`
 
 Type: `bool`
 
+### hpc_azure_disable_predictable_net_names
+
+Whether to disable predictable network interface names by adding `net.ifnames=0`
+to the kernel command line (via the bootloader system role).
+
+This keeps kernel names such as `ib0`, `ib1`, ... instead of `ibP...` on IPoIB,
+but it also affects Ethernet naming (e.g. `eth0` instead of `enP...`).
+
+Default: `true`
+
+Type: `bool`
+
 ### hpc_install_system_openmpi
 
 Whether to install OpenMPI that comes from AppStream repositories and does not have Nvidia GPU support.

defaults/main.yml

@@ -23,6 +23,7 @@ hpc_install_nvidia_fabric_manager: true
 hpc_install_nvidia_imex: true
 hpc_install_rdma: true
 hpc_enable_azure_persistent_rdma_naming: true
+hpc_azure_disable_predictable_net_names: true
 hpc_install_system_openmpi: true
 hpc_build_openmpi_w_nvidia_gpu_support: true
 hpc_install_moneo: true

tasks/main.yml

@@ -578,6 +578,21 @@
         name: nvidia-imex.service
         enabled: true
 
+- name: Disable predictable network interface names on Azure (net.ifnames=0)
+  when:
+    - hpc_azure_disable_predictable_net_names
+  block:
+    - name: Configure net.ifnames via bootloader role
+      include_role:
+        name: fedora.linux_system_roles.bootloader
+      vars:
+        bootloader_settings:
+          - kernel: ALL
+            options:
+              - name: net.ifnames
+                value: "0"
+                state: present
+
 - name: Install RDMA packages
   when: hpc_install_rdma
   block:

templates/test-azure-health-checks.sh.j2

@@ -206,10 +206,10 @@ test_health_log() {
     pass "health.log file exists"
 
     echo "Checking: health.log for errors"
-    if grep -Ei "fail|fault|error" "$log_file" | grep -Eiv "success" > /dev/null 2>&1; then
+    if grep -Eiu '^(fail|fault|error)\b' "$log_file" | grep -Eiv "success" > /dev/null 2>&1; then
         echo ""
         echo "--- Error excerpts from health.log ---"
-        grep -Ei "fail|fault|error" "$log_file" | grep -Eiv "success" | head -20
+        grep -Eiu '^(fail|fault|error)\b' "$log_file" | grep -Eiv "success" | head -20
         echo "--- End of excerpts ---"
         echo ""
         fail "FAIL/FAULT/ERROR found in health.log"

tests/tests_default.yml

@@ -20,6 +20,7 @@
     hpc_install_docker: "{{ hpc_install_nvidia_container_toolkit }}"
     hpc_install_azurehpc_health_checks: "{{ hpc_install_nvidia_container_toolkit }}"
     hpc_install_diagnostics: false
+    hpc_azure_disable_predictable_net_names: false
   tasks:
     - name: Skip unsupported architectures
       include_tasks: tasks/skip_unsupported_archs.yml

tests/tests_include_vars_from_parent.yml

@@ -65,6 +65,7 @@
             hpc_install_docker: "{{ hpc_install_nvidia_container_toolkit }}"
             hpc_install_azurehpc_health_checks: "{{ hpc_install_nvidia_container_toolkit }}"
             hpc_install_diagnostics: false
+            hpc_azure_disable_predictable_net_names: false
 
         - name: Cleanup
           file:

tests/tests_skip_toolkit.yml

@@ -23,6 +23,7 @@
     hpc_install_docker: "{{ hpc_install_nvidia_container_toolkit }}"
     hpc_install_azurehpc_health_checks: "{{ hpc_install_nvidia_container_toolkit }}"
     hpc_install_diagnostics: false
+    hpc_azure_disable_predictable_net_names: false
   tags:
     - tests::reboot
   tasks: