Re: Asymmetric forwarded TCP TX collapse over one GRE path on FreeBSD 14.3-RELEASE-p8 (PF independent, offloads disabled)

> On 25. Feb 2026, at 10:08, Ivan K. <[email protected]> wrote:
> 
> The problem is that I've tried everything I know/found on the internet/was recommended by Codex and GPT, and I (and LLVMs) have run out of ideas :(
Could you capture the traffic of the connection experiencing bad throughput on the
sender and on the receiver side in a .pcap file and post them or prove a link to them?
You can send it privately, if preferred.

What TCP stack are you using?

Best regards
Michael
> 
>> 25 февр. 2026 г., в 10:55, Ivan K. <[email protected]> написал(а):
>> 
>> I don't quite understand your question.
>> iperf3 shows 200-300 Mbit through VPS0 and 100 kbit through VPS2 at the same time when I run them from my laptop. 
>> If I run them from the router, the speeds are good on both tunnels.
>> 
>>> 25 февр. 2026 г., в 01:12, Mark Saad <[email protected]> написал(а):
>>> 
>>> 
>>> 
>>> On Tue, Feb 24, 2026 at 3:35 PM Ivan K. <[email protected]> wrote:
>>> Yes, I wrote about this in my original letter. It is an Intel Gigabit server network card. I disabled everything possible on it with the command:
>>> 
>>> ifconfig igb0 -mextpg -vlanhwtag -vlanhwfilter -vlanhwtso -vlanhwcsum -rxcsum -txcsum -rxcsum6 -txcsum6 -tso -lro
>>> 
>>> The options that remained are: VLAN_MTU, JUMBO_MTU, WOL_MAGIC, HWSTATS
>>> 
>>>> 24 февр. 2026 г., в 23:23, Mark Saad <[email protected]> написал(а):
>>>> 
>>>> 
>>>> 
>>>> On Tue, Feb 24, 2026 at 3:02 PM Ivan K. <[email protected]> wrote:
>>>> Each real life - 1500, gre 1472.
>>>> I checked this. I tried setting mtu 1200 on the client device. I have mss clamping at 1400 in pf.
>>>> If it's mtu, it doesn't explain why the identical tunnel on the same machine functions normally.
>>>> 
>>>> 
>>>>> 24 февр. 2026 г., в 21:10, Mark Saad <[email protected]> написал(а):
>>>>> 
>>>>> 
>>>>> 
>>>>> On Tue, Feb 24, 2026 at 12:19 PM Ivan K. <[email protected]> wrote:
>>>>> Hello, everyone. The text below was written by me and “polished” by a neural network, which I have been using to research this problem for three days now.
>>>>> 
>>>>> I am investigating a reproducible GRE issue and would appreciate advice on further debugging.
>>>>> 
>>>>> Versions / hosts
>>>>> - Router: FreeBSD 14.3-RELEASE-p8
>>>>> - VPS0: FreeBSD 14.3-RELEASE-p8
>>>>> - VPS2: FreeBSD 14.3-RELEASE-p8
>>>>> 
>>>>> Topology
>>>>> - Router has WAN on VLAN interface `onp` (parent `igb0`) and LAN on `igb0`.
>>>>> - Two GRE tunnels from the same router:
>>>>>   - GRE to VPS0: healthy
>>>>>   - GRE to VPS2: problematic
>>>>> - There is also an ng tunnel to VPS2 (works better than bad GRE, but not full speed).
>>>>> - Clients behind router tested via both Wi-Fi and wired Ethernet (same problem).
>>>>> - Also tested traffic from a VM on the router: VM path does not reproduce the severe collapse.
>>>>> 
>>>>> Core symptom (asymmetric)
>>>>> - Only TX direction is affected (client uploads through GRE->VPS2).
>>>>> - RX/download direction is not affected.
>>>>> - Through GRE->VPS2, TCP upload collapses to about ~100 Kbit/s per flow.
>>>>> - 5 parallel TCP flows produce roughly 5 x ~100 Kbit/s.
>>>>> - iperf3 TCP shows many retransmissions and very small cwnd (~2.7 KB).
>>>>> - Same client via GRE->VPS0 is fast (normal high throughput).
>>>>> - Traffic generated locally on router through GRE->VPS2 is also fast.
>>>>> 
>>>>> Firewall / policy checks
>>>>> - PF disabled completely (`pfctl -d`) for tests: problem remains.
>>>>> - Therefore this is not PF rule/policy dependent in current testing.
>>>>> - Also checked for ipfw/dummynet:
>>>>>   - `kldstat | egrep 'ipfw|dummynet|pf'` shows only `pf.ko`.
>>>>>   - `ipfw show` returns “Protocol not available”.
>>>>>   - `net.inet.ip.fw.enable` OID not present.
>>>>> 
>>>>> Offload / NIC checks performed
>>>>> - Explicitly disabled offload features (multiple attempts), including:
>>>>>   - `-lro`
>>>>>   - `-tso` / tso-related toggles
>>>>>   - `-rxcsum` / `-txcsum` / v6 checksum toggles
>>>>>   - VLAN hardware offload related toggles where supported
>>>>> - No improvement.
>>>>> - Current interface snapshot during testing:
>>>>>   - `ifconfig igb0` shows options: `VLAN_MTU,JUMBO_MTU,WOL_MAGIC,HWSTATS` (no LRO/TSO flags active)
>>>>>   - `ifconfig onp` shows `options=0`
>>>>> - Reboots of router and VPS2 done after changes: no improvement.
>>>>> - Also tested with minimal manual GRE setup (without tunnel-management scripts, only one GRE up): no improvement.
>>>>> 
>>>>> Queue/counters/sysctl checks
>>>>> - `netstat -Q` shows no queue drops (`QDrops=0`).
>>>>> - `netstat -iW -I ...` during tests shows no obvious interface error growth.
>>>>> - `sysctl net.inet.ip.fastforwarding` does not exist on this release.
>>>>> 
>>>>> UDP vs TCP behavior
>>>>> - UDP over problematic GRE path can run at high throughput (similar order from client and from router), i.e., no analogous dramatic collapse.
>>>>> - TCP over same path collapses hard in TX direction from forwarded clients.
>>>>> 
>>>>> Packet capture findings
>>>>> - Collected captures at multiple points (router and VPS side).
>>>>> - In bad runs, client-side TCP packets are seen, but many segments are retransmitted repeatedly.
>>>>> - Comparing capture points suggests segments often appear on VPS GRE side only much later / after retransmission, consistent with loss on the GRE path for this forwarded TCP case.
>>>>> - In control runs (local router-generated traffic), packet timing/count alignment is normal.
>>>>> 
>>>>> What makes this puzzling
>>>>> - Same router, same OS version, same general config:
>>>>>   - GRE to VPS0 works well.
>>>>>   - GRE to VPS2 fails for forwarded TCP TX only.
>>>>> - PF disabled does not change behavior.
>>>>> - Offloads disabled does not change behavior.
>>>>> - No obvious drops in standard queue/interface counters.
>>>>> 
>>>>> Request
>>>>> Any suggestions on what to inspect next for FreeBSD 14.3-p8 in this scenario:
>>>>> - netisr / forwarding path diagnostics
>>>>> - GRE-specific kernel instrumentation
>>>>> - known igb/VLAN/GRE forwarding edge cases
>>>>> - any tunables or debug knobs that could explain “forwarded TCP TX collapse only on one GRE peer/path”
>>>>> 
>>>>> Thanks.
>>>>> 
>>>>> 
>>>>> Ivan
>>>>>  Just a guess but it sounds like you may have a MTU issue . What is the MTU settings on each of the real interfaces and the gre tunnels ?
>>>>> -- 
>>>>> mark saad | [email protected]
>>>> 
>>>> Ivan
>>>>   What sort of network cards are you using? This is another long shot but do you have anything that does GRE offload? Some of the intel
>>>> and mellanox cards support some sort of checksum offloads. 
>>>> 
>>>> -- 
>>>> mark saad | [email protected]
>>> 
>>> Ivan
>>>   So back to how you are testing. Do you have iperf / iperf3 blasting traffic from router to target VPS0 and VPS2 at the same time ?
>>> 
>>> 
>>> -- 
>>> mark saad | [email protected]
>> 
>