Resolved a "no lock" error on vSAN NFS v3 volumes caused by internal network timeout issues

VMware ESXi 8.0 Update 3g | 29 JUL 2025 | ISO Build 24859861

What's New

• This release contains bugfixes.

Resolved Issues in VMware ESXi 8.0 Update 3g
This critical bugfix patch resolves numerous issues across various components of ESXi, improving stability, performance, and functionality. Here's a summary of the resolved issues:

vSAN and Storage Issues:

• Resolved a "no lock" error on vSAN NFS v3 volumes caused by internal network timeout issues.
• Fixed a timing issue that caused vSAN ESA clusters to incorrectly show as non-compliant with SPBM policy.
• Addressed an issue preventing new virtual machine creation in vSAN clusters due to an excessive number of witness component objects.
• Fixed an issue where vSAN iSCSI targets lost connectivity after a syslog configuration change.
• Resolved an issue where applying a vSAN license on 4Kn Software Emulated (SWE) disks failed due to incorrect capacity calculation.
• Fixed UNMAP operation failures on NVMe 2.0 devices with specific unmap limits.
• Resolved occasional failures of NVME over TCP adapter to reconnect to storage array ports during upgrades/failovers.
• Addressed slow VM creation and deployment on Hitachi NVMe-oF storage targets using NVMe/TCP.

Virtual Machine and Host Management:

• Fixed an issue where VMs configured with a PCI passthrough device failed to boot.
• Resolved an issue where TLS 1.3 activation in ClusterStore did not persist across ESXi host reboots.
• Fixed an issue where a VM might fail to power on if its .vmx file was changed externally, leading to stale handle errors.
• Addressed issues with Kubernetes control plane containers on suspended/resumed VMs leading to unresponsiveness or resets.
• Resolved hostd service failures when using ESXi host keys to encrypt VMs and then deleting/changing the key.
• Fixed ESXi host failures with a purple diagnostic screen during VM migration from Tintri NFSv4.1 datastores.
• Resolved an issue preventing SSH connections using public/private key pairs with HTTP PUT via ConfigManager APIs.
• Fixed vSphere Virtual Volumes daemon (vvold) failures after converting a VASA provider from external to internal.
• Addressed ClusterStore service crashloops due to excessive memory usage and network overload.
• Resolved DNS traffic overload caused by the authentication service of ClusterStore continuously trying to connect to old host groups.
• Fixed intermittent reporting of hidden portsets (DVFilter Coalesce Portset and pps) in IF-MIB SNMP data.
• Resolved VMX service failures during snapshot operations of VMs with IOFilter attached due to a race condition.
• Fixed false uplink down alarms in vSphere Client after ESXi host reboots due to incorrect event ordering.
• Resolved ESXi host failures with a purple diagnostic screen when running Linux fstrim due to VMFS heap memory exhaustion.
• Fixed VM failures with fixed passthrough hot-plugging enabled due to misallocation of interrupts after hot-adds of NVMe drives.
• Addressed rsyslogd CPU usage at 100% during restart, preventing share mounting.
• Resolved indefinite delays in vSAN CMMDS shutdown during ESXi host reboots.
• Fixed delays in power-on of vGPU virtual machines leading to vpxd overload and vCenter unresponsiveness.
• Addressed a routing issue with Cisco ACI switches causing vSAN traffic loss due to incorrect TCP socket binding.
• Fixed excessive logs in nsxdavim.log leading to syslog server overload due to frequent updates from hostd.
• Resolved an issue where server firmware did not automatically activate PCIe Atomic Operations for AMD GPU passthrough devices.
• Fixed ESXi host failures on debug builds due to a race condition between VM power-off and memory scrubbing.
• Addressed failures to add new disks external to a VM's home directory due to reaching security policy limits.
• Resolved multiple VM resets or unresponsiveness in case of uncorrectable memory errors causing machine check exceptions.
• Fixed VM migration failures during NSX upgrade or DFW reconfiguration.
• Addressed VMX service failures during replication-based snapshot operations due to storage system failures.
• Resolved VM failures with Signal 11 (segmentation violation) during power on due to VM statistics initialization issues.
• Fixed "Route based on physical NIC load" teaming policy inconsistencies during ToR switch upgrades.
• Resolved issues where PCIe devices were reported as non-passthrough capable or SR-IOV could not be activated after upgrading to ESXi 8.0 Update 3 or later due to overly strict sanity checks.
• Fixed ESXi boot failures on systems with more than 512 logical CPUs in one NUMA node.
• Resolved issues where no vGPU device groups were visible for NVIDIA HGX H100 systems due to incorrect SR-IOV virtual function checks.
• Fixed booting failures for vGPU virtual machines on NVIDIA Blackwell GPU.
• Addressed ESXi performance degradation with Hitachi NVMe-oF storage targets due to slow VM creation/deployment.
• Resolved VM vNIC activation failures after vSphere vMotion due to highly fragmented system memory.
• Fixed multiple hardware sensor alerts caused by a Baseboard Management Controller (BMC) issue.
• Resolved an issue where VASA API version did not automatically refresh after upgrade to vCenter Server 8.0.
• Fixed an issue where Lockdown Mode remained active after host reboot despite deactivation.
• Addressed failures in vCenter Server upgrade/update using vCenter Lifecycle Manager with custom update repositories having untrusted certificates.
• Resolved an issue preventing updates to ESXi 8.0 Update 2b or later using esxcli software vib commands.
• Fixed an "Jumpstart plugin restore-networking activation failed" error during ESXi host boot if IPv6 was deactivated