netdev - Re: [PATCH] tcp: sysctl for initial receive window

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <alpine.LNX.2.01.1209250719120.25195@nerf07.vanv.qr>
Date:	Tue, 25 Sep 2012 07:29:19 +0200 (CEST)
From:	Jan Engelhardt <jengelh@...i.de>
To:	Jesper Dangaard Brouer <brouer@...hat.com>
cc:	netdev@...r.kernel.org, Nandita Dukkipati <nanditad@...gle.com>,
	Eric Dumazet <eric.dumazet@...il.com>
Subject: Re: [PATCH] tcp: sysctl for initial receive window



On Friday 2012-09-21 10:55, Jesper Dangaard Brouer wrote:
>diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
>index c7fc107..684131c 100644
>--- a/Documentation/networking/ip-sysctl.txt
>+++ b/Documentation/networking/ip-sysctl.txt
>@@ -257,6 +257,18 @@ tcp_frto_response - INTEGER
> 		  to the values prior timeout
> 	Default: 0 (rate halving based)
> 
>+tcp_init_recv_window - INTEGER
>+	Default initial advertised receive window.  Actual window size
>+	is this value multiplied by the MSS of the connection.  Its

	is this value multiplied by the MSS of the connection.  It is

>+	possible to control/override this value per route table entry
>+	via the iproute2 option initrwnd.
>+	Minimum value is 1, but 2 is the recommended minimum.
>+	The effective max value, is limited by the sockets receive

	The effective max value is limited by the sockets receive

>+	buffer size (default tcp_rmem[1], and possibly scaled by
>+	tcp_adv_win_scale), and can further be limited by window

	tcp_adv_win_scale) and can further be limited by window

>+	clamp.

	clamping.

>+	Default: 10
>+
> tcp_keepalive_time - INTEGER
> 	How often TCP sends out keepalive messages when keepalive is enabled.
> 	Default: 2hours.

The "recommended minimum" is somewhat strange from a language POV,
since the recommendation is actually to _not touch_ the option at all
(because the default works and there is potential abuse as Dave
mentions).

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html