Managed switch drops 802.1Q tagged frames on trunk when MTU set above 1500 to ESXi host
Environment
Small business managed switch (Netgear GS308E FW 2.5.0) connected via 1GbE trunk to an ESXi 7.0 host running multiple VLAN‑tagged VMs and a distributed vSwitch; default MTU 1500 previously, changed to 9000 (jumbo frames) for storage traffic and some VLANs.
Symptoms
After enabling MTU 9000 on the physical switch trunk port and the ESXi physical NIC, VMs on VLAN 100 intermittently lose connectivity to other VLANs and to storage targets; ping with large ICMP payloads (>1500) between hosts on different switches fails with no fragmentation. Smaller packets succeed. The issue is limited to traffic that crosses the managed switch trunk — VMs on the same host can talk at jumbo MTU without problem.
Observations
Captures on the ESXi physical NIC show 1514‑byte frames being forwarded from the host, and the switch egress counters on the trunk port increment but receiving ports drop frames; 'show interfaces status' on the switch reports no link flaps. When the switch MTU is left at 1500 and only the ESXi host uses 9000, hosts can still send large frames internally with offloading, but inter‑switch traffic fails. Lowering MTU on ESXi to 1500 restores full connectivity. Switch firmware is not the latest; vendor documentation warns of limited jumbo frame support when VLAN tagging is enabled on some models.
Already attempted
Reverted MTU to 1500 — problem resolved. Adjusted MTU on all intermediary devices (switch, ESXi pNIC, vmkernel) to 9000 one at a time and tested. Enabled/disabled VLAN tagging on trunk to isolate; issues persist only when both VLAN tagging and MTU >1500 active. Updated NIC driver on ESXi host to latest supported. Performed capture on switch mirror port — saw tagged frames truncated when larger than 1500. Contacted switch vendor support; they noted older switch models may mishandle jumbo frames with 802.1Q tags.
Question
Is this a configuration/MTU mismatch across devices or a known firmware limitation for certain managed switches handling 802.1Q tagged jumbo frames, and what safe diagnostics and mitigations should I use to restore connectivity without disabling VLAN separation or jumbo frames globally?
Verified community troubleshooting
No answers yet..
Incident analysis
Most likely causes
Summary
Suggested Fixes
Continue troubleshooting
Related troubles
Matched from the title, description and tags
- Virtualization27% match
KVM guest interface MTU drops to 1500 and large TCP transfers stall after enabling 802.1Q VLAN tagging on host bridge
Open0 answers - Networking23% match
LAN devices intermittently lose connectivity after enabling jumbo frames on mixed‑vendor switches
Open0 answers - Networking23% match
Managed switch starts ARP storm after adding VLAN: likely misconfigured PIM/IGMP snooping combined with default VLAN flooding
Open0 answers - Networking21% match
SMB shares become unreachable after enabling jumbo frames on mixed‑vendor switch
Open0 answers