Connection problem between HyperV and a 6860E stack

[fripouille]

Hello,

I explain my problem to you...

The customer has a server equipped with two 10Gbps network cards on which HyperV is installed in order to virtualize several servers.

On the switch side, we have 2 stacked 6860E-U28 switches (version 8.8.56.R02).
We are having problems with the stability of the 10Gbps copper link : if we disconnect and reconnect the RJ45 cord from the server, the link no longer wants to be established.

Here is information seen in the logs when connecting the cord:
2022 May 13 11:04:30.517 SW_EPGT_LT1_COEUR swlogd intfCmm Mgr EVENT: CUSTLOG CMM Link 2/1/32 operationally up
2022 May 13 11:04:30.527 SW_EPGT_LT1_COEUR swlogd intfCmm Mgr INFO: esmSetRateLimit: Txing limit=997, trafficType=0, limitType=1 to zslot=0,
2022 May 13 11:04:30.527 SW_EPGT_LT1_COEUR swlogd intfCmm Mgr INFO: esmSendConfMsg: chassis 2 zslot 0 zport 31 Txed conf msgId:393231
2022 May 13 11:04:30.527 SW_EPGT_LT1_COEUR swlogd intfCmm Mgr INFO: esmSetRateLimit: Txing limit=997, trafficType=1, limitType=1 to zslot=0,
2022 May 13 11:04:30.527 SW_EPGT_LT1_COEUR swlogd intfCmm Mgr INFO: esmSendConfMsg: chassis 2 zslot 0 zport 31 Txed conf msgId:393231
2022 May 13 11:04:30.527 SW_EPGT_LT1_COEUR swlogd intfCmm Mgr INFO: esmSetRateLimit: Txing limit=997, trafficType=2, limitType=1 to zslot=0,
2022 May 13 11:04:30.527 SW_EPGT_LT1_COEUR swlogd intfCmm Mgr INFO: esmSendConfMsg: chassis 2 zslot 0 zport 31 Txed conf msgId:393231
2022 May 13 11:04:30.527 SW_EPGT_LT1_COEUR swlogd intfCmm Mgr INFO: cmmEsmHandleNiMsg: Rx CMM_ESM_LINK_STATUS_CHG from chassis 2 NI 1
2022 May 13 11:04:30.551 SW_EPGT_LT1_COEUR swlogd intfNi Drv INFO: niEsmGetEvent:RX CMM_ESMDRV_CMD_CONFIG_MSGID
2022 May 13 11:04:30.551 SW_EPGT_LT1_COEUR swlogd intfNi Drv INFO: eniApplyConfig: cmd:21 zport:31 cmdHy:0, apMedia:1
2022 May 13 11:04:30.552 SW_EPGT_LT1_COEUR swlogd intfNi Drv INFO: niEsmGetEvent:RX CMM_ESMDRV_CMD_CONFIG_MSGID
2022 May 13 11:04:30.552 SW_EPGT_LT1_COEUR swlogd intfNi Drv INFO: eniApplyConfig: cmd:28 zport:31 cmdHy:0, apMedia:1
2022 May 13 11:04:30.553 SW_EPGT_LT1_COEUR swlogd intfNi Drv INFO: niEsmGetEvent:RX CMM_ESMDRV_CMD_CONFIG_MSGID
2022 May 13 11:04:30.553 SW_EPGT_LT1_COEUR swlogd intfNi Drv INFO: eniApplyConfig: cmd:34 zport:31 cmdHy:0, apMedia:1
2022 May 13 11:04:30.560 SW_EPGT_LT1_COEUR swlogd portMgrNi main INFO: : [pmnHALLinkStatusCallback:207] LINKSTS 2/1/32 DOWN (gport 0x2001f) Speed 0 Duplex HALF
2022 May 13 11:04:30.581 SW_EPGT_LT1_COEUR swlogd stpNi _SOKt INFO: stpnimsg_processMsgFromPM: PM_LINK_STATUS_MSGID gPort=x2001f linkStatus=0
2022 May 13 11:04:30.581 SW_EPGT_LT1_COEUR swlogd ^^ repeated 1 times
2022 May 13 11:04:30.581 SW_EPGT_LT1_COEUR swlogd intfNi Drv INFO: niEsmSendLinkStatusChgMsg(1066): linkstatus DOWN sent on peerId=1
2022 May 13 11:04:30.581 SW_EPGT_LT1_COEUR swlogd intfCmm Mgr EVENT: CUSTLOG CMM Link 2/1/32 operationally down
2022 May 13 11:04:30.601 SW_EPGT_LT1_COEUR swlogd intfCmm Mgr INFO: cmmEsmHandleNiMsg: Rx CMM_ESM_LINK_STATUS_CHG from chassis 2 NI 1

Sometimes this message also appears:

2022 May 13 11:19:18.377 SW_EPGT_LT1_COEUR swlogd intfNi Drv INFO: niEsmResetTxvrInfo:IGNORE: port:32 not in my chassis:2:1
2022 May 13 11:19:18.378 SW_EPGT_LT1_COEUR swlogd ChassisSupervisor SharedMem Sync INFO: sending shared sfp update 1/32 state 0

The two server cards are aggregated on the server side but not on the switch side, as Windows recommends. However, we have configured an aggregate on the switch side but the connection problems are identical...

Is there a known problem with HyperV or a particular setting to be made ?

Thank you in advance for your help.