Host crashing/PSOD with error "Spin count exceeded - possible deadlock".

Products

• VMware vSphere ESXi

Issue/Introduction

• An ESX Host crashes with a PSOD that has the error: Spin count exceeded - possible deadlock


• During the PSOD ,the host crashes, and all services and VMs running on the host are terminated. The VMs do not get a chance to gracefully shut down, but instead are powered off abruptly.
  
  
• Below is an example of the backtrace :

Panic Details: Crash at 2023-08-28T02:05:28.046Z on CPU 96 running world 558792334 - python. VMK Uptime:471:12:50:00.699

Panic Message: @BlueScreen: Spin count exceeded - possible deadlock

Backtrace:

0x453c10e1bac0:[0x4200308fec3b]PanicvPanicInt@vmkernel#nover+0x327 stack: 0x453c10e1bb98, 0x0, 0x4200308fec3b, 0x453c10e1bc00, 0x453c10e1bac0
0x453c10e1bb90:[0x4200308ff194]Panic_NoSave@vmkernel#nover+0x4d stack: 0x453c10e1bbf0, 0x453c10e1bbb0, 0x453c10e1bc48, 0x453c10e1bc48, 0x60
0x453c10e1bbf0:[0x42003081b976]Lock_CheckSpinCount@vmkernel#nover+0x26f stack: 0xfffffffffffffff8, 0x15a8f9ed9da6c33, 0x3ff, 0x43201ac1efd8, 0x1
0x453c10e1bc40:[0x420030908c8d]SP_WaitReadLock@vmkernel#nover+0xba stack: 0x43201ac1efd8, 0x43201ac1efdc, 0x0, 0x0, 0x41ffd380b7a0
0x453c10e1bc80:[0x420030908d2a]SPAcqWriteLockWork@vmkernel#nover+0x33 stack: 0x43201ac10f60, 0x420031ce1595, 0x43201cac3b80, 0x420031d4da3b, 0x43201cac3b80
0x453c10e1bca0:[0x420031ce1594]rw_wlock@(tcpip4)#<None>+0x8d stack: 0x43201cac3b80, 0x41ffd38030b0, 0x6, 0x41ffd380b7a0, 0x43201ac10f60
0x453c10e1bcb0:[0x420031d4da3a]tcp_usr_attach@(tcpip4)#<None>+0x7b stack: 0x6, 0x41ffd380b7a0, 0x43201ac10f60, 0x43201ac1aff0, 0x453c10e1bd58
0x453c10e1bcf0:[0x420031cff0a6]socreate@(tcpip4)#<None>+0x1fb stack: 0x58, 0x43201ac1aff0, 0x41ffd380b7a0, 0x1, 0x6
0x453c10e1bd40:[0x420031ceb2b6]vmk_socket@(tcpip4)#<None>+0xa3 stack: 0x1, 0x420031ccf047, 0x45dbc1758950, 0x4313d2803210, 0x43201ac01700
0x453c10e1bda0:[0x420030cef385]UserSocketInetCreateInt@vmkernel#nover+0x126 stack: 0x41fff0c25bb4, 0x420000000000, 0x45dbc1758958, 0x0, 0x45dbc1758950
0x453c10e1be10:[0x420030cf1bbb]UserSocketInet_Create@vmkernel#nover+0x1b0 stack: 0x453c10e1befc, 0x45dbc1758950, 0x434488602010, 0x420030cb6e41, 0x1
0x453c10e1be80:[0x420030ceac9b]UserSocket_NewSocket@vmkernel#nover+0x18c stack: 0x0, 0x100000130ccafdc, 0x45dbc1758950, 0x420030cb196e, 0x0
0x453c10e1bef0:[0x420030d0b5cd]LinuxSocket_NewSocket64@vmkernel#nover+0xe stack: 0x0, 0x420030cb1020, 0xffffffffffffffef, 0x453c10e1bfd0, 0x0
0x453c10e1bf10:[0x420030cb101f]User_LinuxSyscallHandler@vmkernel#nover+0x1a4 stack: 0x0, 0x0, 0x0, 0x42003094d068, 0x10b
0x453c10e1bf40:[0x42003094d067]gate_entry@vmkernel#nover+0x68 stack: 0x0, 0x29, 0x71c50569f7, 0x6, 0x71cbb9c0a0

Cause

There is an issue getting a write lock and this causes the spin lock crash. 

Resolution

This issue is fixed in:

vSphere ESXi 7.0 U3q
vSphere ESXi 8.0 U1

Additional Information

Interpreting an ESXi host purple diagnostic screen