Frequent FC disconnections on VMware host due to faulty XBAR on Cisco MDS switch
Applies to
- Ontap 9
- VMware
- Cisco MDS switch
Issue
- VMware host reports frequent disconnections with FC LUNs.
- Port are reporting
IO WQE
errors in EMS logs as below.
Mon Jan 15 15:12:23 +0000 [NetApp-01: fct_tpd_work_thread_0: fcp.io.status:debug]: STIO Adapter:1a IO WQE failure, Handle 0x0, Type 8, S_ID: 3xxxFC, VPI: 5, OX_ID: 4xxC, Status 0x3 Ext_Status 0x1d
Mon Jan 15 15:13:40 +0000 [NetApp-01: fct_tpd_work_thread_0: fcp.io.status:debug]: STIO Adapter:1c IO WQE failure, Handle 0x2, Type 8, S_ID: 3xxxEF, VPI: 6, OX_ID: 4xx8, Status 0x3 Ext_Status 0x16
- Ports on other clusters in the same environment are also reporting
IO WQE
errors in EMS logs. - These affected ports are connected to Cisco MDS switchA.
- Ports connected to Cisco MDS switchB are not reporting any errors.
- Rx & Tx values for all ports are in optimal range.
- Command timeouts are seen on VMware host end.