PSOD on the NSX Host - VDL2_ProcessPortDBEvent
search cancel

PSOD on the NSX Host - VDL2_ProcessPortDBEvent

book

Article ID: 393372

calendar_today

Updated On:

Products

VMware NSX

Issue/Introduction

  • The ESXi host in NSX environment crashes with a PSOD
  • The PSOD backtrace may contain errors like the following:

LogEFI[2098810]: Panic from another CPU (cpu 92, world 42400331): ip=0x######## randomOff=0x2de00000:Spin count exceeded - possible deadlock with PCPU 66Halting PCPU 92.Panic from another CPU (cpu 2, world 3715716): ip=0x######## randomOff=0x2de00000:Spin count exceeded - possible deadlock with PCPU 66Halting PCPU 2.Panic from another CPU (cpu 79, world 2098221): ip=0x######## randomOff=0x2de00000:Spin count exceeded - possible deadlock with PCPU 112Halting PCPU 79.Panic from another CPU (cpu 112, world 3718690): ip=0x######## randomOff=0x2de00000:NMI IPI: Panic requested by another PCPU. RIPOFF(base):RBP:CS [0x11f53f(0x42002de00000):0x42005c0016c0:0x748] (Src 0x4, CPU112)Halting PCPU 112.Panic from another CPU (cpu 87, world 2098178): ip=0x######## randomOff=0x2de00000:Spin count exceeded - possible deadlock with PCPU 112Halting PCPU 87.Panic from another CPU (cpu 72, world 42401223): ip=0x######## randomOff=0x2de00000:Spin count exceeded - possible deadlock with PCPU 112Halting PCPU 72.Panic from another CPU (cpu 82, world 42401222): ip=0x######## randomOff=0x2de00000:Spin count exceeded - possible deadlock with PCPU 112Halting PCPU 82.2025-03-24T20:16:51.374Z cpu66:2097874)ESC[45mESC[33;1mVMware ESXi 8.0.2 [Releasebuild-23305546 x86_64]ESC[0m
LogEFI[2098810]: NMI IPI: Panic requested by another PCPU. RIPOFF(base):RBP:CS [0x11d1d4(0x42002de00000):0x1:0x748] (Src 0x4, CPU66)
LogEFI: cpu66:2097874)cr0=0x8001003d cr2=0x7fa532626748 cr3=0x23c000 cr4=0x14216c
LogEFI: cpu66:2097874)FMS=06/6a/6 uCode=0xd0003e7
LogEFI[2098810]: *PCPU66:2097874/PSEventHelper
LogEFI[2098810]: PCPU  0: UUSUSUUSSUSUUUSSUUUUSUSUUUUUUUUUUSSSUSSUSUSSUSSUUUUUSUUSUUUUSUUS
LogEFI[2098810]: PCPU 64: SSSSVUSSSVVVSVUSVVSSVVISVVVVUSVVVVVVVSVSVUVSIVIVSSIISISISSUSVSSS
LogEFI: cpu66:2097874)Code start: 0x42002de00000 VMK uptime: 116:08:10:44.932
LogEFI: cpu66:2097874)Saved backtrace from: pcpu 66 SpinLock spin out NMI
LogEFI: cpu66:2097874)0x453a9691be00:[0x42002df1d1d3]RefCountBlock@vmkernel#nover+0x3c stack: 0x600004a
LogEFI: cpu66:2097874)0x453a9691be10:[0x42002e0a4554]Port_AcquireExcl@vmkernel#nover+0x201 stack: 0x250
LogEFI: cpu66:2097874)0x453a9691be70:[0x42002ff7410f][email protected]#1.1.8.0.23382415+0xa8 stack: 0x432d3c80205c

Environment

NSX 4.x

Cause

This is a VDL2 issue where the port-exclusive lock is being acquired without the portset write lock. It occurs due to a race condition in the lock ordering, typically happening when a new port (VM or container) is created or during a vMotion event.

Resolution

This issue is resolved in VMware NSX 4.2.1.3 available at Broadcom Downloads.
If you are having difficulty finding and downloading software, please review the Download Broadcom products and software KB.