Ejecuta cualquier Skill en Manus
con un clic

Ejecuta cualquier Skill en Manus con un clic

upgrading-network-operator

Estrellas7

Forks4

Actualizado9 de abril de 2026 a las 15:23

Upgrade NVIDIA Network Operator to a new version. Use when changing NETOP_VERSION, running upgrade-network-operator.sh, migrating between operator versions, handling CRD schema changes, or rolling back a failed upgrade.

Instalación

Instalar con Codex o Claude Copia este prompt, pégalo en Codex, Claude u otro asistente, y deja que revise la página de la skill y la instale por ti.

Ejecutar en Manus

Fuente

Mellanox

Mellanox/netop-tools

Abrir repositorio de GitHub Ver repositorios del creador

Descarga

Ejecutar en Manus

Ocupaciones relacionadasSOC

Basado en la clasificación ocupacional SOC

Administradores de redes y sistemas informáticosOcupaciones informáticas y matemáticas·SOC 15-1244

SKILL.md

readonly

name	upgrading-network-operator
description	Upgrade NVIDIA Network Operator to a new version. Use when changing NETOP_VERSION, running upgrade-network-operator.sh, migrating between operator versions, handling CRD schema changes, or rolling back a failed upgrade.

Upgrading Network Operator

Upgrade the Network Operator Helm release to a new version with zero-downtime node rolling.

Supported Versions

24.7.0, 24.10.0, 24.10.1, 25.1.0, 25.4.0, 25.7.0, 25.10.0, 26.1.0 (default)

Pre-Upgrade Checklist

Verify target version exists: helm search repo nvidia/network-operator --versions
Check K8s compatibility (26.1.0 requires K8s 1.32+)
Generate config for new version with CREATE_CONFIG_ONLY=1 first
Review YAML diffs between old and new generated configs
Ensure all nodes are healthy: kubectl get nodes

Upgrade Procedure

# 1. Set new version
export NETOP_VERSION="26.1.0"
source global_ops.cfg

# 2. Preview config changes (dry run)
export CREATE_CONFIG_ONLY=1
cd usecase/${USECASE}
${NETOP_ROOT_DIR}/ops/mk-config.sh
# Review generated YAML

# 3. Run upgrade
export CREATE_CONFIG_ONLY=0
${NETOP_ROOT_DIR}/upgrade/upgrade-network-operator.sh

What the upgrade script does

Cordons all worker nodes
Scales Network Operator deployment to 0 replicas
Regenerates config: mk-values.sh, mk-nic-cluster-policy.sh, mk-network-cr.sh
Applies updated NicClusterPolicy
Applies new CRDs via applycrds.sh
Applies network resources via apply-network-cr.sh
Updates NIC config (if NIC_CONFIG_ENABLE=true)
Runs helm upgrade
Uncordons worker nodes

Post-Upgrade Verification

kubectl get pods -n ${NETOP_NAMESPACE}
./ops/getnetwork.sh
./ops/checksriovstate.sh
./ops/syncsriov.sh
./ops/checkipam.sh

Version-Specific Notes

Version	Key Changes
25.10.*	`MAINTENANCE_OPERATOR_ENABLE` defaults to `NIC_CONFIG_ENABLE` value
26.1.*	`MAINTENANCE_OPERATOR_ENABLE` defaults to `true` independently
25.1.0	Uses `nicConfigurationOperator` parameter (older API)
25.4.0+	Uses `deployNodeFeatureRules` for NFD

Rollback

If upgrade fails, restore previous config and re-run:

export NETOP_VERSION="<previous_version>"
source global_ops.cfg
${NETOP_ROOT_DIR}/upgrade/upgrade-network-operator.sh

Common Failures

Symptom	Fix
Helm chart not found	Verify `NETOP_VERSION` and `helm repo update`
CRD validation errors	Check feature gate compatibility with new version
Nodes stuck cordoned	Run `source ops/cordon.sh && uncordon`
Finalizers blocking delete	`./ops/getfinalizers.sh`, then patch to remove

Más de este repositorio

mismo repositorio

configuring-netop-platform

Mellanox/netop-tools

Configure netop-tools for a specific hardware platform and use case. Use when setting up global_ops_user.cfg, choosing a use case (sriovnet_rdma, sriovibnet_rdma, hostdev, ipoib, macvlan), selecting IPAM type, configuring NETOP_NETLIST device list, setting feature flags, or switching between platforms (DGX, Dell, Lenovo, OCI, SuperMicro).

2026-04-097

deploying-network-operator

Mellanox/netop-tools

Deploy NVIDIA Network Operator to a Kubernetes cluster via netop-tools. Use when installing network operator, setting up RDMA networking, deploying SR-IOV, running ins-network-operator.sh, generating Helm values, applying network CRDs, or troubleshooting deployment failures.

2026-04-097

managing-netop-devices

Mellanox/netop-tools

Manage SR-IOV virtual functions, PCI devices, link speed, and RDMA testing for NVIDIA Network Operator. Use when setting VFs, querying PCI devices, forcing link speed, running RDMA bandwidth tests, checking RDMA capability, managing node labels and taints, or cordoning nodes.

2026-04-097

testing-netop-configs

Mellanox/netop-tools

Run and create tests for netop-tools configuration generation. Use when running unitest.sh, validating YAML generation, creating new test cases, checking test baselines, debugging test failures, or verifying config changes don't break existing tests.

2026-04-097

troubleshooting-network-operator

Mellanox/netop-tools

Diagnose and fix NVIDIA Network Operator issues in Kubernetes. Use when pods can't get RDMA network, SR-IOV VFs not showing, IP allocation failures, network attachment missing, must-gather diagnostics, operator pods crashing, NicClusterPolicy errors, or IPAM pool exhaustion.

2026-04-097

Upgrading Network Operator

Upgrade the Network Operator Helm release to a new version with zero-downtime node rolling.

Supported Versions

24.7.0, 24.10.0, 24.10.1, 25.1.0, 25.4.0, 25.7.0, 25.10.0, 26.1.0 (default)

Pre-Upgrade Checklist

Verify target version exists: helm search repo nvidia/network-operator --versions

Check K8s compatibility (26.1.0 requires K8s 1.32+)

Generate config for new version with CREATE_CONFIG_ONLY=1 first

Review YAML diffs between old and new generated configs

Ensure all nodes are healthy: kubectl get nodes

Upgrade Procedure

# 1. Set new version export NETOP_VERSION="26.1.0" source global_ops.cfg # 2. Preview config changes (dry run) export CREATE_CONFIG_ONLY=1 cd usecase/${USECASE} ${NETOP_ROOT_DIR}/ops/mk-config.sh # Review generated YAML # 3. Run upgrade export CREATE_CONFIG_ONLY=0 ${NETOP_ROOT_DIR}/upgrade/upgrade-network-operator.sh

What the upgrade script does

Cordons all worker nodes

Scales Network Operator deployment to 0 replicas

Regenerates config: mk-values.sh, mk-nic-cluster-policy.sh, mk-network-cr.sh

Applies updated NicClusterPolicy

Applies new CRDs via applycrds.sh

Applies network resources via apply-network-cr.sh

Updates NIC config (if NIC_CONFIG_ENABLE=true)

Runs helm upgrade

Uncordons worker nodes

Post-Upgrade Verification

kubectl get pods -n ${NETOP_NAMESPACE} ./ops/getnetwork.sh ./ops/checksriovstate.sh ./ops/syncsriov.sh ./ops/checkipam.sh

Version-Specific Notes

Version

Key Changes

25.10.*

MAINTENANCE_OPERATOR_ENABLE defaults to NIC_CONFIG_ENABLE value

26.1.*

MAINTENANCE_OPERATOR_ENABLE defaults to true independently

25.1.0

Uses nicConfigurationOperator parameter (older API)

25.4.0+

Uses deployNodeFeatureRules for NFD

Rollback

If upgrade fails, restore previous config and re-run:

export NETOP_VERSION="<previous_version>" source global_ops.cfg ${NETOP_ROOT_DIR}/upgrade/upgrade-network-operator.sh

Common Failures

Symptom

Fix

Helm chart not found

Verify NETOP_VERSION and helm repo update

CRD validation errors

Check feature gate compatibility with new version

Nodes stuck cordoned

Run source ops/cordon.sh && uncordon

Finalizers blocking delete

./ops/getfinalizers.sh, then patch to remove