diff --git a/content/cumulus-netq-412/Installation-Management/Install-NetQ/Install-NetQ-System.md b/content/cumulus-netq-412/Installation-Management/Install-NetQ/Install-NetQ-System.md index 28203aab2b..663fb2c421 100644 --- a/content/cumulus-netq-412/Installation-Management/Install-NetQ/Install-NetQ-System.md +++ b/content/cumulus-netq-412/Installation-Management/Install-NetQ/Install-NetQ-System.md @@ -36,7 +36,7 @@ You can install NetQ either on your premises or as a remote, cloud solution. If | Server Arrangement | Hypervisor | Requirements & Installation | | :--- | --- | :---: | | Single server | KVM or VMware | {{<link title="Set Up Your Virtual Machine for a Single Cloud Server" text="Start install">}} | -| High-availability cluster | KVM or VMware | {{<link title="Set Up Your Virtual Machine for a Cloud HA Server Cluster" text="Start install">}}| +| High-availability cluster | KVM or VMware | {{<link title="Set Up Your Virtual Machine for a Cloud HA Server Cluster" text="Start install">}}| ## Base Command Manager diff --git a/content/cumulus-netq-412/Monitor-Operations/Monitor-the-RoCE-Service.md b/content/cumulus-netq-412/Monitor-Operations/Monitor-the-RoCE-Service.md index 034baa7d0a..e8e3828618 100644 --- a/content/cumulus-netq-412/Monitor-Operations/Monitor-the-RoCE-Service.md +++ b/content/cumulus-netq-412/Monitor-Operations/Monitor-the-RoCE-Service.md @@ -27,7 +27,7 @@ netq check roce ## View RoCE Counters Networkwide in the UI -1. From the header or {{<img src="https://icons.cumulusnetworks.com/01-Interface-Essential/03-Menu/navigation-menu.svg" height="18" width="18">}} menu, select **Spectrum-X**, then **RoCE**. +1. From the header or {{<img src="https://icons.cumulusnetworks.com/01-Interface-Essential/03-Menu/navigation-menu.svg" height="18" width="18">}} Menu, select **Spectrum-X**, then **RoCE**. 2. Select either **RoCE switches** or **RoCE DPUs**. diff --git a/content/cumulus-netq-412/Troubleshoot-Issues/_index.md b/content/cumulus-netq-412/Troubleshoot-Issues/_index.md index c661400660..24cefd9bb4 100644 --- a/content/cumulus-netq-412/Troubleshoot-Issues/_index.md +++ b/content/cumulus-netq-412/Troubleshoot-Issues/_index.md @@ -72,63 +72,62 @@ Verified installer version FINISHED ... ``` {{< /expand >}} -<!-- -## Troubleshoot Installation and Upgrade + +## Troubleshoot NetQ Installation and Upgrade Issues Before you attempt a NetQ installation or upgrade, verify that your system meets the {{<link title="Install the NetQ System" text="minimum VM requirements">}} for your deployment type. {{%notice note%}} -If an upgrade or installation process stalls or fails, run the `netq bootstrap reset` command to stop the process, followed by the `netq install` command to re-attempt the installation or upgrade. +If an upgrade or installation process stalls or fails, run the {{<link title="bootstrap" text="netq bootstrap reset">}} command to stop the process, followed by the {{<link title="install" text="netq install">}} command to re-attempt the installation or upgrade. {{%/notice%}} -## Known Installation and Upgrade Issues {{<tabs "TabID113" >}} {{<tab "Upgrade Issues">}} -| Error | Setup | Solution | +| Error Message | Deployment Type | Solution | | ---- | ---- | ---- | -| Cannot upgrade a non-bootstrapped NetQ server. Please reset the cluster and re-install.| | Only a server that has been bootstrapped and has a valid `/etc/app-release` file can be upgraded.<br> 1. Run the `netq bootstrap reset` command. <br> 2. Run the `netq install` command. | -| Unable to get response from admin app. | | Re-run the `netq upgrade bundle <tarball>` command. If the retry fails with same error, reset the server and run the `install` command:<br> 1. Run the `netq bootstrap reset` command <br> 2. Run the `netq install` command. | -| Unable to get response from kubernetes api server. | | Re-run the `netq upgrade bundle <tarball>` command. If the retry fails with same error, reset the server and run the `install` command:<br> 1. Run the `netq bootstrap reset` command <br> 2. Run the `netq install` command. | -| Cluster vip is an invalid parameter for standalone upgrade. | Single server | Remove the `cluster-vip` option from the `netq upgrade bundle` command. | -| Please provide cluster-vip option and run command. | HA server cluster | Include the `cluster-vip` option in the `netq upgrade bundle` command. | -| Could not find admin app pod, please re-run the command. | | Re-run the `netq upgrade bundle <tarball>` command. | +| Cannot upgrade a non-bootstrapped NetQ server. Please reset the cluster and re-install.| | Only a server that has been bootstrapped and has a valid `/etc/app-release` file can be upgraded.<br> 1. Run the `netq bootstrap reset` command. <br> 2. Run the {{<link title="install" text="netq install">}} command according to your deployment type. | +| Unable to get response from admin app. | | Re-run the {{<link title="upgrade" text="netq upgrade bundle <text-bundle-url>">}} command. If the retry fails with same error, reset the server and run the `install` command:<br> 1. Run the `netq bootstrap reset` command. <br> 2. Run the {{<link title="install" text="netq install">}} command according to your deployment type. | +| Unable to get response from kubernetes api server. | | Re-run the {{<link title="upgrade" text="netq upgrade bundle <text-bundle-url>">}} command. If the retry fails with same error, reset the server and run the `install` command:<br> 1. Run the `netq bootstrap reset` command <br> 2. Run the {{<link title="install" text="netq install">}} command according to your deployment type. | +| Cluster vip is an invalid parameter for standalone upgrade. | Single server | Remove the `cluster-vip` option from the {{<link title="upgrade" text="netq upgrade bundle">}} command. | +| Please provide cluster-vip option and run command. | HA cluster | Include the `cluster-vip` option in the {{<link title="upgrade" text="netq upgrade bundle">}} command. | +| Could not find admin app pod, please re-run the command. | | Re-run the {{<link title="upgrade" text="netq upgrade bundle <text-bundle-url>">}} command. | | Could not upgrade server, unable to restore got exception: {} | On-premises | The backup/restore option is only applicable for on-premises deployments which use {{<link title="Install a Custom Signed Certificate" text="self-signed certificates">}}.| {{</tab>}} {{<tab "Installation Issues" >}} -| Error | Setup | Solution | +| Error Message | Deployment Type | Solution | | ---- | ---- | ---- | -| NetQ is operational with version: {}. Run the bootstrap reset before re-installing NetQ. | | 1. Run the `netq bootstrap reset` command. <br> 2. Run the `netq install` command. | +| NetQ is operational with version: {}. Run the bootstrap reset before re-installing NetQ. | | NetQ has previously been installed. To install a new version:<br>1. Run the {{<link title="bootstrap" text="netq bootstrap reset">}} command. <br> 2. Run the {{<link title="install" text="netq install">}} command. | | The Default interface was not found | | You must have a default route configured in your routing table, and the associated network interface must correspond to this default route. -| No default route found. Please set the default route via interface {} and re-run the installation. | | See above. | -| Default route set via a different interface {}. Please set the default route via interface {} and re-run the installation.| | See above. | +| No default route found. Please set the default route via interface {} and re-run the installation. | | You must have a default route configured in your routing table, and the associated network interface must correspond to this default route. | +| Default route set via a different interface {}. Please set the default route via interface {} and re-run the installation.| | You must have a default route configured in your routing table, and the associated network interface must correspond to this default route. | | Minimum of {} GB RAM required but {} GB RAM detected.| | Increase VM resources according to your {{<link title="Install the NetQ System" text="deployment model requirements">}}.| | Minimum of {} CPU cores required but {} detected.| | Increase VM resources according to your {{<link title="Install the NetQ System" text="deployment model requirements">}}.| | Please free up disk as {} is {}% utilised. Recommended to keep under 70%. | | Delete previous software tarballs in the `/mnt/installables/` directory to regain space. If you cannot decrease disk usage to under 70%, contact the NVIDIA support team. | -| Did not find the vmw_pvscsi driver enabled on this NetQ VM. Please re-install the NetQ VM on ESXi server. | VMware | The NetQ VM must have the `vmw_pvscsi` driver enabled. | +| Did not find the `vmw_pvscsi` driver enabled on this NetQ VM. Please re-install the NetQ VM on ESXi server. | VMware | The NetQ VM must have the `vmw_pvscsi` driver enabled. | | Error: Bootstrapped IP does not belong to any local IPs | | The IP address used for bootstrapping should be from the local network. | | ERROR: IP address mismatch. Bootstrapped with: {} kube_Config: {} Admin_kube_config: {}| | The bootstrap IP address must match the kube config and admin kube config IP addresses. | -| ERROR: Clock not synchronised. Please check timedatectl. | | The system clock must be synchronized. Verify synchronization using the `timedatectl` command. | +| ERROR: Clock not synchronised. Please check `timedatectl`. | | The system clock must be synchronized. Verify synchronization using the `timedatectl` command. | | {} does not have sse4.2 capabilities. Check lscpu. | | The CPU model used for the installation must support SSE4.2. | | NTP is installed. Please uninstall NTP as it will conflict with chrony installation.| | Uninstall NTP and any other NTP services, such as `ntpd` or SNTP.| | Netqd service is not running | | Verify that the `netqd` service is up and running prior to installation. | -| Found identical ip for {} and {}/ Please provide different ip for cluster vip/workers. | Cluster | The cluster virtual IP address (VIP) and worker node IP addresses must be unique. | -| Please provide worker nodes IPV6 addresses in order to have IPV6 support. | Cluster | IPv6 addresses must be provided for worker nodes if IPv6 support is required. | -| Master node is not initialised. Run “net install cluster master-init” on master node before NetQ Install/upgrade command. | Cluster | Initialize the master node with the `netq install cluster master-init` command.| -| Worker node Ip {} is not reachable| Cluster | Make sure worker nodes are reachable in your network. | -| Worker node {} is not initialised. Please initialise worker node and re-run the command.| Cluster | After initializing the cluster on the master node, initialize each worker nodes with the `netq install cluster worker-init` command. | -| Cluster VIP is not valid IP address | Cluster | Provide a valid cluster IP address. | -| All cluster addresses must belong to the same subnet. Master node net mask = {} | Cluster | Make sure all cluster IP addresses---the master, worker and virtual IP---belong to the same subnet.| -| Virtual IP {} is already used | Cluster | Provide a unique virtual IP address. | -| Package {} with version {} must be installed. | | Make sure `netq-apps` version is the same as the tarball version. | -| Master node is already bootstrapped | | Run the `netq bootstrap reset` command, followed by the `netq install` command to re-attempt the installation. | +| Found identical ip for {} and {}/ Please provide different ip for cluster vip/workers. | HA cluster | The cluster virtual IP address (VIP) and worker node IP addresses must be unique. | +| Please provide worker nodes IPV6 addresses in order to have IPV6 support. | HA cluster | Provide IPv6 addresses for each worker node. | +| Master node is not initialised. Run “net install cluster master-init” on master node before NetQ Install/upgrade command. | HA cluster | Initialize the master node with the `netq install cluster master-init` command.| +| Worker node Ip {} is not reachable| HA cluster | Make sure worker nodes are reachable in your network. | +| Worker node {} is not initialised. Please initialise worker node and re-run the command.| HA cluster | After initializing the cluster on the master node, initialize each worker node with the `netq install cluster worker-init` command. | +| Cluster VIP is not valid IP address | HA cluster | Provide a valid cluster IP address. | +| All cluster addresses must belong to the same subnet. Master node net mask = {} | HA cluster | Make sure all cluster IP addresses---the master, worker, and virtual IP---belong to the same subnet.| +| Virtual IP {} is already used | HA cluster | Provide a unique virtual IP address. | +| Package {} with version {} must be installed. | | Make sure the `netq-apps` version is the same as the tarball version. | +| Master node is already bootstrapped | | Run the {{<link title="bootstrap" text="netq bootstrap rest">}} command, followed by the {{<link title="install" text="netq install">}} command to re-attempt the installation. | {{</tab>}} {{</tabs>}} ---> + ## Installation and Upgrade Hook Scripts NVIDIA might provide hook scripts to patch issues encountered during a NetQ installation or upgrade. When you run the `netq install` or `netq upgrade` command, NetQ checks for specific hook script filenames in the `/usr/bin` directory. The expected filenames for NetQ 4.12.0 are: