XvMotion runs very slow and eventually fails

Products

• VMware vSphere ESXi
• VMware vCenter Server

Issue/Introduction

This issue manifests with the following conditions:
- several vCenter Server instances spread out geographically
- Provisioning is enabled on vmk0 (Management vmkernel interface)
- XvMotion from one site to another starts off fast and then slows down and eventually fails
- The following log lines are seen in vmkernel.log:
2026-01-19T16:37:58.241Z Wa(180) vmkwarning: cpu60:2106172)WARNING: XVMotion: 3064: 52fa5f6a-####-####-####-689e####b0e S: timed out while waiting for disk 1's queue count to drop below the maximum limit of 32768 blocks. This could indicate either network or stora$
2026-01-19T16:37:58.241Z Wa(180) vmkwarning: cpu60:2106172)WARNING: Nfc: 99: 52fa5f6a-####-####-####-689e####b0e S: Failed to allocate the XvMotion blocks to back scatter gather array: Timeout
2026-01-19T16:37:58.241Z In(182) vmkernel: cpu60:2106172)Migrate: 101: 552fa5f6a-####-####-####-689e####b0e S: MigrateState: Failed
2026-01-19T16:37:58.241Z Wa(180) vmkwarning: cpu60:2106172)WARNING: Migrate: 257: 52fa5f6a-####-####-####-689e####b0e S: Failed: Out of memory (0xbad0014) @0x42########8f
2026-01-19T16:37:58.241Z In(182) vmkernel: cpu60:2106172)Nfc: 427: 52fa5f6a-####-####-####-689e####b0e S: Exiting
2026-01-19T16:37:58.256Z In(182) vmkernel: cpu102:2100676 opID=########)Nfc: 770: 52fa5f6a-####-####-####-689e####b0e S: Waiting for 0 pending disk IOs to complete.
2026-01-19T16:37:58.256Z In(182) vmkernel: cpu102:2100676 opID=########)Nfc: 837: 52fa5f6a-####-####-####-689e####b0e S: Flushing all pending IOs to destination
2026-01-19T16:38:00.251Z Wa(180) vmkwarning: cpu74:2106170)WARNING: VMotionUtil: 6190: 52fa5f6a-####-####-####-689e####b0e S: Stream restart aborted after a soft barrier due to migration failure.

- Additionally, testing with different types of storage (eg local disk, NFS etc) makes no difference to behaviour, ruling out any storage related issues

Environment

VMware vCenter Server 8.x
VMware ESXi 8.0.3

Cause

This happens as a result of network throttling on the WAN network between sites

Resolution

Continue troubleshooting further with your firewall vendor and ISP