ESXi pathing degradation with iSCSI and LACP
Applies to
- ONTAP
- Interface Groups (ifgrps)
- LACP
Issue
In an environment utilizing iSCSI and interface groups with LACP enabled on the switch, the esxi hosts are reporting near constant pathing loss and pathing regained events happening on their hosts along with heartbeat timeouts.
Lost access to volume 69456395-6af690f5-a6e8-405b7f3945a0 (AFF_A_ISCSI01) due to connectivity issues. Recovery attempt is in progress and outcome will be reported shortly.2026-05-26T11:57:37.947Z In(14) vobd[2097956]: [vmfsCorrelator] 6549668914423us: [esx.problem.vmfs.heartbeat.timedout] 69695a51-e5ac77ad-736e-405b7f3945a0 AFF_A_ISCSI012026-05-26T11:57:57.852Z In(14) vobd[2097956]: [vmfsCorrelator] 6549674101479us: [vob.vmfs.heartbeat.recovered] Reclaimed heartbeat for volume 69695a51-e5ac77ad-736e-405b7f3945a0 (AFF_A_ISCSI01): [Timeout] [HB state abcdef02 offset 3506176 gen 103 stampUS 6549674101297 uuid 69b19a2a-e0a30c4d-6d0a-405b7f39ee60 jrnl <FB 16777226> drv 24.82]On the storage array, no errors are present on the individual ethernet ports, however, there is a near constant flow of initiators logging back into storage, even though a session already exists:
[?] Fri May 29 11:36:52 -0500 [AFFC30-A: iswt_admin_thread: iscsi.notice:notice]: ISCSI: New session from initiator iqn.1998-01.com.vmware:esxi01:cii_encrypt/AK1MHP6KDheprDaYkpztCFv8nkUQBBYNMYJ7+E2fjZc=/cii_encrypt at IP addr xxx.xxx.xxx.xx [?] Fri May 29 11:42:12 -0500 [AFFC30-A: iswt_admin_thread: iscsi.warning:notice]: ISCSI: New session request from initiator iqn.1998-01.com.vmware:esxi01, a session from this initiator already exists.