Two VMs were rebooted on a host abruptly.
You may see these messages 'NFS: 6284: Status: No connection. Retrying synchronous write I/O 4 of 25 times' on /var/run/log/vmkernel.log
VMware ESXi 7.x
The issue was primarily caused by the simultaneous failure of vmnic1 and vmnic4, which resulted in a network disruption.
This disruption triggered NFS storage I/O errors, ultimately leading to a loss of connectivity for the VMs and forcing them to reboot.
/var/run/log/vmkernel.log shows the below:
2025-02-19T06:42:45.017Z cpu0:2097726)NFS: 6284: Status: No connection. Retrying synchronous write I/O 4 of 25 times
/var/run/log/vmkwarning.log:
2025-02-19T06:42:26.016Z cpu50:2100636 opID=79d630c1)WARNING: NFS: 2581: Failed to get attributes (I/O error)2025-02-19T06:42:33.016Z cpu40:3915681 opID=218e57b4)WARNING: NFS: 2581: Failed to get attributes (I/O error)
2025-02-19T14:12:13.214Z: [netCorrelator] 16175198449711us: [vob.net.dvport.uplink.transition.down] Uplink: vmnic4 is down. Affected dvPort: 5672/50 09 89 6a 88 23 23 42-08 d6 82 fb 61 3b 97 75. 1 uplinks up. Failed criteria: 1282025-02-19T14:12:13.214Z: [netCorrelator] 16175198449805us: [vob.net.vmnic.linkstate.down] vmnic vmnic4 linkstate down2025-02-19T14:12:13.220Z: [netCorrelator] 16175198456706us: [vob.net.dvport.uplink.transition.down] Uplink: vmnic1 is down. Affected dvPort: 56/50 09 7e d3 68 46 0a 9f-6d f5 bf 34 0a 90 e6 09. 1 uplinks up. Failed criteria: 128
vmnic4 0000:af:00.0 Up 25000 Full 9000 nmlx5_core 4.22.73.1004 14.28.4512 b8:xx:xx:xx:xx:d8 15b3 1015 15b3 0016 Mellanox Technologies MT27710 Family [ConnectX-4 Lx]vmnic5 0000:af:00.1 Up 25000 Full 1500 nmlx5_core 4.22.73.1004 14.28.4512 b8:xx:xx:xx:xx:d9 15b3 1015 15b3 0016 Mellanox Technologies MT27710 Family [ConnectX-4 Lx]
Ensure the network adapters are running the recommended firmware version as specified in the compatibility guide Driver and firmware compatibility
Investigate the physical network infrastructure for any issues that could have caused the simultaneous failure of vmnic1 and vmnic4.
Involve the NFS storage vendor to check for any issues related to the I/O errors and ensure that the NFS configuration is optimal.