netdev - Re: stuck tcp connections 2.6.26-rc7

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] [day] [month] [year] [list]

Message-ID: <dd18b0c30806261540l56d68868p1d32597b2f5cb171@mail.gmail.com>
Date:	Thu, 26 Jun 2008 22:40:14 +0000
From:	"Justin Mattock" <justinmattock@...il.com>
To:	"David Miller" <davem@...emloft.net>
Cc:	ray-lk@...rabbit.org, didier@...oud.com,
	linux-kernel@...r.kernel.org, netdev@...r.kernel.org
Subject: Re: stuck tcp connections 2.6.26-rc7

On Thu, Jun 26, 2008 at 7:54 AM, Justin Mattock <justinmattock@...il.com> wrote:
> On Wed, Jun 25, 2008 at 11:29 PM, Justin Mattock
> <justinmattock@...il.com> wrote:
>> On Wed, Jun 25, 2008 at 10:24 PM, David Miller <davem@...emloft.net> wrote:
>>> From: "Ray Lee" <ray-lk@...rabbit.org>
>>> Date: Wed, 25 Jun 2008 08:16:20 -0700
>>>
>>>> On my system, I've noticed that the flashplugin can manage to starve
>>>> other processes in a pure scheduler sort of way, so you might try
>>>> running latencytop to see if there's some sort of issue there. OTOH,
>>>> I've noticed firefox itself hang for 3-15 seconds in the middle of a
>>>> read (just as you describe), with nothing else going on in the system,
>>>> so <shrug>. I've been blaming firefox, but I could be wrong.
>>>
>>> See if firefox is stuck in disk wait when things get wedged like
>>> this.
>>>
>>
>> Hello; Thanks for the response. I must admit I need to figure out how
>> to do that(disk wait mode with firefox).
>> Aside from that, it seems there are reports of this with the latest
>> and so forth. As a note
>> This explains why and when I plugged in an older kernel that had no
>> problems at all,
>> was suddenly giving me problems, leading me to believe  that my
>> service provider(was bandwidth limiting), or a script kiddie
>> was messing with the line, but this did not seem to be the case; Thankfully.
>> I'll look into it and let you know.
>> regards;
>>
>> --
>> Justin P. Mattock
>>
>
> Hello;
> I must be mature too admit when I'm wrong, (I'm very wrong, and stupid);
>  I was too quick to judge. : (          (apologize firefox)
> I had overlooked one aspect of the situation: "wpa_supplicant",
> After watching wpa_supplicant do it's job, I had noticed that it keeps
> CTRL-EVENT-DISCONNECT - Disconnect event - remove keys,
> and CTRL-EVENT-SCAN-RESULTS and trying to associate with
> XX:XX:XX:XX:XX (SSID= XX freq= 2462 MHz).
> anyways despite the manual jargon of wpa_supplicant; basically
> whenever wpa_supplicant disconnects  and reconnects  the keys
> I had noticed flash player, and any other connection's to stop or even
> pause during the playback, depending on how well the transmission was.
> So for now as for what I can see: wpa_supplicant is the main cause of
> my troubles.
> regards,
>
>
>
> --
> Justin P. Mattock
>

Yep that resolved the issue. i.e. madwifi latest svn and previous
versions has a bug in it to where it is causing wpa_supplicant
to constantly disconnect and connect (re keying event) under heavy
loads resulting in horrible internet connections.
loaded an older version prior to march 08 and voila no problems(knock on wood)
I'll monitor this for a few days, and see what results of it.
-- 
Justin P. Mattock
--
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