"RSSPlugCleanupRSSEngine@(lb_netqueue_bal)" purple diagnostic screen on ESXi 6.7
search cancel

"RSSPlugCleanupRSSEngine@(lb_netqueue_bal)" purple diagnostic screen on ESXi 6.7

book

Article ID: 317834

calendar_today

Updated On:

Products

VMware vSphere ESXi

Issue/Introduction

Symptoms:
  • The ESXi 6.7 host experiences a purple diagnostic screen.
  • You see a backtrace similar to: 

    yyyy-mm-dd <time> </time> cpu9:2097316)0x451a8521bc70:[0x41801a68ad30]RSSPlugCleanupRSSEngine@(lb_netqueue_bal)#<None>+0x7d stack: 0x4305e4c1a8e0, 0x41801a68af2b, 0x430db8c6b188, 0x4305e4c1a8e0, 0x0
    yyyy-mm-dd <time> </time> cpu9:2097316)0x451a8521bc90:[0x41801a68af2a]RSSPlugInitRSSEngine@(lb_netqueue_bal)#<None>+0x127 stack: 0x0, 0x20c49ba5e353f7cf, 0x4305e4c1a930, 0x4305e4c1a800, 0x4305e4c1a9f0
    yyyy-mm-dd <time> </time> cpu9:2097316)0x451a8521bcd0:[0x41801a68b21c]RSSPlug_PreBalanceWork@(lb_netqueue_bal)#<None>+0x1cd stack: 0x32, 0x32, 0x0, 0xe0, 0x4305e4c1a800
    yyyy-mm-dd <time> </time>cpu9:2097316)0x451a8521bd30:[0x41801a687752]Lb_PreBalanceWork@(lb_netqueue_bal)#<None>+0x21f stack: 0x4305e4c1a800, 0xff, 0x0, 0x43057f38ea00, 0x4305e4c1a800
    yyyy-mm-dd <time> </time> cpu9:2097316)0x451a8521bd80:[0x418019c13b18]UplinkNetqueueBal_BalanceCB@vmkernel#nover+0x6f1 stack: 0x4305e4bc6088, 0x4305e4c1a840, 0x4305e4c1a800, 0x4305e435fb30, 0x4305e4bc6088
    yyyy-mm-dd <time> </time> cpu9:2097316)0x451a8521bf00:[0x418019cd1c89]UplinkAsyncProcessCallsHelperCB@vmkernel#nover+0x116 stack: 0x430749e60870, 0x430196737070, 0x451a85223000, 0x418019ae832b, 0x430749e60088
    yyyy-mm-dd <time> </time> cpu9:2097316)0x451a8521bf30:[0x418019ae832a]HelperQueueFunc@vmkernel#nover+0x157 stack: 0x430749e600b8, 0x430749e600a8, 0x430749e600e0, 0x451a85223000, 0x430749e600b8
    yyyy-mm-dd <time> </time> cpu9:2097316)0x451a8521bfe0:[0x418019d081f2]CpuSched_StartWorld@vmkernel#nover+0x77 stack: 0x0, 0x0, 0x0, 0x0, 0x0


Environment

VMware ESXi 6.7.x

Cause

This issue occurs when the load-based netqueue balancer module, misses or fails to cleanup the RSS engine private data, as a result dedicated heap gets full and causes the crash during subsequent load-balancing.

Resolution

This is a known issue affecting ESXi 6.7.x.

This issue is resolved in ESXi 6.7 Update 2 .

Workaround:
To workaround this issue, run this command on the ESXi host command line: esxcli network nic queue loadbalancer set --rsslb=false -n vmnicX
Where in vmnicX 'X' specifies the NIC number from the ESXi host to which we need to set the parameter.

Powered by Wolken Software