The table lists the known issues in VMware vSphere ESXi 8.x.
Issue ID | Functional area | Description | Workaround |
---|---|---|---|
JIT-246266 | Coredump file | On installing and uninstalling of OMSA component on any version of ESXi, a coredump file is generated at /var/core/. File name example - sfcb-ProviderMa-zdump.000. This can be safely ignored and there is no functionality loss. | N/A |
JIT-276570 | NVMe or TCP | Changing the speed of the data paths of the existing nvmetcp luns, on ESXi 8. 0 U1c or ESXi 8.0, and rebooting the physical switches, few of the ESXi hosts with NVMe TCP luns might experience PSOD. This has been addressed in ESXi 8.0 U2. | Migrate to ESXi 8.0 U2. |
JIT-234520 | Networking | The Pensando DPU port might be marked as down if the adapter port speed and physical port switches are not matched. | Ensure to configure both the adapter port setting and switch port setting to auto negotiation. |
JIT-282881 | Storage devices | On attaching the virtual image in the iDRAC virtual console, the ESXi host shows the virtual image or iso details in the ESXi host web client under the storage adapter tab. However, on disconnecting the virtual image, the storage adapter tab still reports the size of the image and this value does not get refreshed. This does not impact functionality, and this is a cosmetic issue. | Restart the host service on the ESXi host to get the values refreshed. or reset the usb on iDrac virtual console. |
JIT-240816 & JIT-242713 | Networking | Using vCenter web client, we cannot enable more than 36 virtual functions (VFs) on Data Processing Unit (DPU) configuration. |
Use cli to enable the VFs. The command to enable 64 VFs on dual ports of Mellanox DPU : esxcli system module parameters set -m nmlx5_core -p max_vfs=64,64. It is recommended to enable the VFs DPUs using esxcli rather than using both the methods, in order to avoid enabling few VFs using web client and few using CLI. |
JIT-277181 | VM Migration | While manually migrating a VM in a cluster of 20 nodes or more, the VMotion wizard might not list all the nodes in the page. This is a cosmetic defect with the UI. | Use a quick filter to search the node to select a particular host. |
JIT-279119 | ESXi Deployment | For the PowerEdge system with Intel Saphir Rapids, the CPU processor information is not populated in the operating system boot DCUI screen. | N/A |
JIT-280057 | ESXi Upgrades | ESXi operating system upgrades fail on PowerEdge XR4000 with GPIO driver installed. | To uninstall the GPIO driver prior to ESXi 8.x operating system upgrade #esxcli software vib remove -v dellgpio. |
JIT-261233 | SRIOV | Host that is installed with ESXi 8.0 and above might hit the PSOD post powering on the virtual machine that is assigned with VFs from Intel NIC devices. To see this failure, while assigning the VFs to a virtual machine, the virtual machine should not be having the guest operating system installed and secure boot should be enabled. | On a failing system, run the command esxcfg-advcfg --set-kernel "FALSE" iovEnablePostedIntr and reboot the system. |
JIT-247089 | vCenter Monitor | The storage adapter page of NVMe devices for ESXi host in vCenter, the namespace tab does not reflect the correct value. | N/A |
JIT-265924 | NVMe over TCP Lun size | On a system with lun configured using NVMeTCP, the lun size might be in correctly listed in the namespace tab for NVMe storage adapters section. | N/A |
JIT-272873 | Data Processing Unit (DPU) | ESXi on DPU fails to come online when ESXi gets in to the PSOD state. It is a known issue in the current iDRAC firmware release (6.10.85.00) specially to PowerEdge R760 & R660 platforms. | Invoke a manual cold boot of x86 after the PSOD. |
JIT-252359 | ESXi deployment | PERC LUN fails to enumerate due to the interruption of cookies depletion on certain hardware configurations during ESXi deployment. See VMware Configuration Maximum Guide to determine the recommended configuration limits for VMware products. | Add the boot parameter maxIntrCookies=4096 while booting the ESXi installer. |
JIT-273110 | ESXi Storage | On ESXi 8.0 U1 A04 (based on ESXi 8.0 U1c) host, post creating NVMe over TCP storage adapters and while configuring them the auto discovery of CDC may not work on certain configurations. | Run the command esxcli nvme fabrics discover -D -a vmhba <ID> |
JIT-238057 | Core files | Post performing the NVMe drive hot plug task from open manage (OM), and physically removing the drive from the slot, core files with file name sfcb-vmware_bas-zdump.000 is created. These files can be ignored and no functionality loss is seen. Contact Dell Tech Support if there is any functionality loss. | N/A |
JIT-243012, JIT-243319, and JIT-243604 | Monitoring or Management | On ESXi host with Data Processing Unit (DPU) we see empty performance charts or incorrect metrics for DPU-related metrics. This is a cosmetic issue. | We can get the relevant DPU utilization using esxtop tool with ESXi. |
JIT-260679 | CLI utility | On ESXi host, esxtop shows abnormal DAVG or KAVG readings for workloads with block size higher than 256 kb. | N/A |
JIT-261233 | SRIOV | ESXi host presses PSOD if virtual functions of Intel E810 devices are added to the virtual machine and powered-on. The PSOD is only seen if the virtual machine is not installed with guest operating systems. | Assign virtual functions to virtual machine only post installing guest operating system in virtual machine. |
JIT-263147 | Inventory or status reporting | iDRAC Redfish reports incorrect DPU boot status. At times, iDRAC Redfish report DPU Boot status as operating system booting even when the DPU operating system is booted. The expected boot status is the operating system running when the DPU operating system is up and running. This is because the communication between the DPU operating system and iDRAC is lost. | Attempt a cold boot of the x86 host if this happens. Contact Dell Tech support if the issue is not resolved after a cold boot. |
JIT-265183 | Storage | The user sees that an exception is reported when the Datastore browse option is used without any Datastore available in the system using host web client. | This is a known issue, the user has to add a valid datastore either local or remote. |
JIT-245199 | NIC firmware version that is displayed in esxi or vCenter is incorrect. | On ESXi host with the device BRCM NetXtreme Gigabit Ethernet 5720, the command esxcli network nic get -n vmnicX; where X is the vmnic number; displays the firmware version. However, this version does not match the firmware that is displayed in the iDRAC page under device. This is a cosmetic issue. | See the iDRAC Firmware inventory page for this network device firmware version. |
JIT-219523 | vCenter UI or hardware monitoring | Fan redundancy status is incorrect for ESXi host with system fans active and running, the vCenter server web client, or ESXi host web client lists the fan redundancy as redundancy lost. This is a cosmetic issue and can be ignored. See the fan redundancy status shown in the iDRAC UI to observe the correct status. | N/A |
JIT-239334 | vCenter UI or Data Processing Unit (DPU) | On vCenter host web client or ESXi host web client, an alarm stating Esxtokend dpu-trust failed alarm is seen on ESXi hosts with DPU's post reboot of ESXi host. This alarm is not critical and can be ignored. | N/A |
JIT-245199 | vCenter UI or Host Hardware Inventory | NIC firmware mismatch on vCenter web client, the ESXi host displays incorrect firmware for NIC. This is a known issue. | See the firmware inventory for devices in iDRAC. |
JIT-231337 | ESXi upgrade | Upgrading to ESXi 8.0 from an earlier ESXi version fails due to system management software and tools incompatibility. | Uninstall the impacted vib using #esxcli software vib remove -n <vib name> before upgrading to ESXi 8.0. |
JIT-246436 | Virtual Machine (VM) power operations | VM fails to power-on VM configured with UPT (Unified Pass Through) interface fails to power-on the ESXi host with Pensando DPU. | Ensure to enable SRIOV for the Pensando DPU. |
JIT-236014 | NIC driver version | On ESXi 8.0 host with Mellanox NIC devices, the iDRAC page shows an incorrect driver version for Mellanox NIC devices. | N/A |
JIT-242170 | Log collection | On an ESXi 8.0 host with DPU installed and iSM installed, the support assist log collection fails. | Add the USBNIC end-point IP address to esxioComm.
Allowed IPs for allowing the same channel for iSM to interact with iDRAC RCE server:
|
JIT-246511 | vCenter deployment | vCenter Server 8.0 appliance deployment fails while performing vCenter server 8.0 deployment using a UI installer on the RHEL 9 operating system, the deployment wizard fails with an error message: A problem occurred while reading the OVA File: TypeError: Cannot read properties of undefined reading 'length'. | On the RHEL operating system, install the libnsl package using the command
dnf install libnsl.
Ensure to configure the required repositories prior to execution of the command. |
JIT-245046 | vCenter server or device name discrepancy | On ESXi 8.0, the Quad port Intel NIC name is displayed incorrectly. Instead of displaying Intel(R) Ethernet 25G 4P E810-XXV OCP, the nic name is displayed as Intel(R) Ethernet Controller E810-C for SFP. This is a cosmetic defect and can be safely ignored. | See the iDRAC page for the device name. |