VMware ESXi 8.0 Update 3e Release Notes - VMware ESXi free again?

VMware ESXi 8.0 Update 3e (Build 24674464) was released on 10 April 2025. The release notes are available here.

When I went through these release notes, I saw a very interesting statement ...

Broadcom makes available the VMware vSphere Hypervisor version 8, an entry-level hypervisor. You can download it free of charge from the Broadcom Support Portal - here.    

To be honest, I don't know if VMware community and home labbers will stay or migrate back to ESXi, and how long Broadcom will provide ESXi for free. To be honest, it seems that the relationship between Broadcom and the VMware community was broken, and trust is a very important factor to invest time in some technology.

Besides the statement about making ESXi free again, I went through the fixes and categorized improvements/fixes, and the majority of them fell into the STORAGE category, followed by the COMPUTE/Hypervisor category, and the NETWORKING category.

Full categorization is ...

• 34 fixes in the storage category
• 12 fixes in the compute category
• 7 fixes in the networking category
• 6 fixes in the manageability category
• 2 fixes in the security category

The majority of storage improvements/fixes do not surprise me. The storage is the most critical component of the data center.

Below are listed improvements/fixes within categories and sub-categories. 

STORAGE (34)

NVMeoF/TCP improvements/fixes (5)

PR 3482259: NVMe over TCP controllers might stay offline and not recover even when the network and the target port have no issues

PR 3473257, PR 3469074: ESXi NVMe/TCP controllers do not recover connectivity when a target upgrades or restarts

PR 3454941: You cannot create datastores on namespaces backed by some kind of NVMe/TCP storage arrays in vSphere 8.0 Update 3b environments

PR 3444303: A rare race condition when connecting to a target NVMe over TCP controller might cause ESXi failure with a purple diagnostic screen 

PR 3488415: ESXi host fails with a purple diagnostic screen with messages such as #PF Exception 14 in world XXXXXXX:vmknvmeGener

iSCSI improvements/fixes (4)

PR 3459526: ESXi boot might take longer on hosts with iSCSI configuration

PR 3469036: ESXi installation on an iSCSI LUN fails as the ESXi iSCSI Boot Firmware Table (iBFT) module fails to discover the LUN

PR 3470385: Restoring the iSCSI configurations on an ESXi host might fail during boot due to stale entries in the ConfigStore database

PR 3481507: Discovery of iSCSI dynamic targets, LUNs, and datastores fails on ESXi host reboot

Fibre Channel improvements/fixes (2)

PR 3445276, PR 3486955: Storage path unclaiming requests might be blocked due to a Fabric Performance Impact Notification (FPIN) issue

PR 3469014: ESXi host might fail with a purple diagnostic screen if the Multipath Plug-in (MPP) does not implement the event handler for Fabric Performance Impact Notification (FPIN)

vSAN improvements/fixes (8)

PR 3477062: vSAN hosts might become unresponsive due to high latencies and high CPU usage during peaks of guest unmap traffic

PR 3473626: If a command issued to a vSAN disk group complete after 120 seconds, the vSAN host might fail with a purple diagnostic screen

PR 3484669: You see intermittent gaps in vSAN performance statistics

PR 3441150: In the vSphere Client, you see unusually high storage utilization of vSphere Replication persistent state files (.psf) in a vSAN datastore

PR 3463364: Storage policy change in a vSAN ESA cluster with less than six hosts might cause performance issues during vSAN resync

PR 3438291: ESXi hosts on a vSAN cluster might fail with purple diagnostic screen due to a rare timing issue

PR 3479340: You see checksum errors for some objects in a stretched cluster after a network outage

PR 3486647: Copying files with the cp command creates files with unexpected access timestamp (atime) and modification timestamp (mtime) values (this is about vSAN File Services NFS 4.1)

SCSI improvements/fixes (2)

PR 3483589: In very rare cases, ESXi hosts might fail with a purple diagnostic screen due to a SCSI-3 reservation failure

PR 3479464 - SCSI commands from GuestOS might fail with PSOD.

Raw Device Mapping (RDM) improvements/fixes (1)

PR 3470732: Virtual machines using Raw Device Mapping (RDM) might become unresponsive due storage heap exhaustion

VMFS improvements/fixes (2)

PR 3480154: High VMFS heap memory consumption might cause virtual machines to become unresponsive

PR 3450501: ESXi hosts working with both VMFS5 and VMFS6 datastores might fail with a purple diagnostic screen due to rare issue

VVOLs improvements/fixes (6)

PR 3433295: Virtual machines residing on a vSphere Virtual Volumes datastore might become inaccessible due to a size reporting issue

PR 3431433: Storage tests might fail when using NVMe as Guest Controllers with SCSI vSphere Virtual Volumes on Windows Server Failover Clusters (WFSC)

PR 3451651: ESXi host fails with a purple diagnostic screen after unsuccessful rebind operation of vSphere Virtual Volumes

PR 3450374: A rare race between rebind and lazy unbind tasks in vSphere Virtual Volume might cause an ESXi host to fail with a purple diagnostic screen

PR 3476179: A memory leak might prevent the Object Storage File System Daemon (osfsd) from processing requests for vSphere Virtual Volumes

PR 3459100: Virtual machine migration in a vSphere Virtual Volumes datastore might cause an ESXi host to fail with a purple diagnostic screen

VADP (vSphere Storage APIs - Data Protection) improvements/fixes (2)

PR 3477772: Encrypted virtual machines with active Change Block Tracking (CBT) might intermittently power off after a rekey operation

PR 3460312: Virtual machines on NFSv3 datastores might fail during snapshot consolidation if a third-party backup is also active

UNMAP improvements/fixes (1)

PR 3465047: You see no UNMAP commands sent to NVMe 2.0 targets or targets supporting TP4040

Storage I/O stack improvements/fixes (1)

PR 3444408: Allocation failure in Physical Region Pages (PRP) during I/O split might cause stuck I/O

COMPUTE (12)

Hypervisor improvement/fixes (12)

PR 3467137: A race condition during heap cleanup might cause an ESXi host to fail with a purple diagnostic screen

PR 3341273: Migration of virtual machines fails with an error "Failed to allocate migration heap" on the source ESXi host

PR 3483610: If you insert an invalid or blank DVD on an ESXi host and configure it as a host device on virtual machines, the VM performance degrades

PR 3454942: In the vSphere Client, you cannot reconfigure a vGPU virtual machine if more than 128 vCPUs are active

PR 3457744: Migration of virtual machines to 8.0 Update 3 and later ESXi host might result in TPM null hierarchy issues

PR 3447505: Nonuniform Sub-NUMA Clustering (SNC) sizes might cause ESXi hosts to fail with a purple diagnostic screen post upgrade to ESXi 8.0 Update 3

PR 3433610: ESXi hosts fail with purple diagnostic screen when connecting to the network I/O scheduler

PR 3437835: AMD Turin processors with PCI Multi-segment Support preset in the BIOS might cause ESXi host failures

PR 3477409: vSphere vMotion operations might fail due to a rare issue with no namespaces present in the namespacemgr.db file

PR 3456018: ESXi hosts fail with purple diagnostic screen and message such as "PCI bus error, undiagnosed"

PR 3470500: After update to ESXi 8.0 Update 3b, you see an alarm requiring a hardware reset due to AMD erratum 1474 on all updated hosts

PR 3442088: vSphere vMotion tasks fail with an error NamespaceMgr could not lock the db file

NETWORKING (7)

vSwitch improvements/fixes (4)

PR 3461166: In the vSphere Client, you cannot change a VMNIC on the virtual switch of an ESXi host

PR 3481216: "Route based on physical NIC load" policy does not work as expected

PR 3475015: Duplicate traffic on multiple uplinks might cause traffic outage

PR 3434172: You cannot apply a reverse path filter for virtual machines in promiscuous mode even /Net/ReversePathFwdCheckPromisc is active

TCP improvements/fixes (1)

PR 3466538: A race condition between the TCP control path and the keep alive timer might cause an ESXi host to fail with a purple diagnostic screen

NVIDIA nmlx5 NIC driver improvements/fixes (2)

PR 3453042: An ESXi host fails with a purple diagnostic screen due to uplink watchdog resets on nmlx5 driver NICs

PR 3453046: You might see low throughput with IPv6 UDP traffic in Enhanced Datapath mode for the nmlx5 driver

SECURITY (2)

Firewall improvement/fixes (2)

PR 3450290: You cannot manually deactivate the DVSSync firewall ruleset

PR 3437903: The intrusion detection and prevention system (IDS/IPS) might report false positives for dropped packets in vSphere reporting tools

MANAGEABILITY (6)

Logging improvements/fixes (2)

PR 3484787: If you reconfigure the scratch location and reboot an ESXi host, all logging stops and the host becomes unresponsive

PR 3469652: Logging stops on all log files and configured remote host targets on ESXi hosts

PR 3433436: In the vSphere Client, you see multiple alarms "Cannot login user root@xxxxx: no permission" for ESXi hosts in lockdown mode

Identity Management improvements/fixes (2)

PR 3486556: You cannot retrieve Active Directory domain accounts from ESXi hosts by using PowerCLI

PR 3469107: ESXi hosts intermittently disconnect from the Active Directory domain or from vCenter

Monitoring improvements/fixes (2)

PR 3484669: You see intermittent gaps in vSAN performance statistics

PR 3410296: You see gaps in the collection of metrics for some virtual machines