Re: [PATCH mptcp-next 0/3] mptcp: blackhole: sysctl SYN retrans + fix conditions

[Matthieu Baerts <matttbe@kernel.org>]

H Mat,

On 17/01/2025 01:13, Mat Martineau wrote:
> On Tue, 14 Jan 2025, Matthieu Baerts (NGI0) wrote:
> 
>> Recently, I found a network having weird behaviours with MPTCP packets:
>>
>> - The first connection to a server had a successful 3WHS, then MPTCP
>>  options got stripped off.
>>
>> - The next one had the first SYN (with or without MPTCP) and 5
>>  retransmissions dropped, before being apparently intercepted and
>>  proxied to the end server.
>>
>> - (The next ones were sometimes intercepted, sometimes not, or dropped
>>   at the beginning. I'm trying to find out which kind of "optimiser" is
>>   causing this.)
>>
>> The result of this was a blackhole being "wrongly" detected, and no ways
>> to force connections with quite a few SYN drops to finally use MPTCP at
>> the end.
>>
>> In this series, we have:
>>
>> - A small fix for the doc.
>>
>> - A new sysctl to change the number of SYN retransmitted with MPTCP
>>  options before falling back to TCP. The modification looks simple
>>  enough to still be sent to netdev before the closure I think.
>>
> 
> Hi Matthieu -
> 
> Patches 1 and 2 LGTM:
> 
> Reviewed-by: Mat Martineau <martineau@kernel.org>

Thank you for the review!

I already applied these two patches:

New patches for t/upstream-net and t/upstream:
- 34dc1c307f30: doc: mptcp: sysctl: blackhole_timeout is per-netns
- Results: 1b76b0bd2522..551c1dba2a68 (export-net)
- Results: 2a6dfa001e66..2d16bea74b90 (export)

New patches for t/upstream:
- 5fb87b001602: mptcp: sysctl: add syn_retrans_before_tcp_fallback
- Results: 2d16bea74b90..b7be33140cc3 (export)

Tests are now in progress:

- export-net:
https://github.com/multipath-tcp/mptcp_net-next/commit/70769bbce0f766c3583cb1caa791722685884f19/checks
- export:
https://github.com/multipath-tcp/mptcp_net-next/commit/caa6490e85f954f97ad955a2496c1b1297efee89/checks

Cheers,
Matt
-- 
Sponsored by the NGI0 Core fund.