kern/163208: [pf] PF state key linking mismatch
Ermal Luçi
eri at freebsd.org
Mon Jan 23 11:50:07 UTC 2012
The following reply was made to PR kern/163208; it has been noted by GNATS.
From: =?ISO-8859-1?Q?Ermal_Lu=E7i?= <eri at freebsd.org>
To: =?ISO-8859-1?Q?Tilman_Keskin=F6z?= <arved at freebsd.org>
Cc: bug-followup at freebsd.org, freebsd-pf at freebsd.org
Subject: Re: kern/163208: [pf] PF state key linking mismatch
Date: Mon, 23 Jan 2012 12:16:38 +0100
--90e6ba6e89dc108bd804b7302abd
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
On Sun, Jan 22, 2012 at 11:41 AM, Tilman Keskin=F6z <arved at freebsd.org> wro=
te:
> * Bjoern A. Zeeb [Sat, 21 Jan 2012 21:01:41 +0000]:
> >
> > On 21. Jan 2012, at 20:52 , Tilman Keskin=F6z wrote:
> >
> >>
> >> On Jan 21, 2012, at 21:01 , Fabian Keil wrote:
> >>
> >>> Tilman Keskin=F6z <arved at FreeBSD.org> wrote:
> >>>
> >>>> Same here.
> >>>>
> >>>> Also Fabian Keil reported this in
> >>>>
> http://lists.freebsd.org/pipermail/freebsd-current/2011-July/025696.html
> >>>
> >>> This has been fixed in CURRENT shortly thereafter:
> >>> http://lists.freebsd.org/pipermail/freebsd-pf/2011-July/006199.html
> >>>
> >>> Maybe the fix hasn't been MFC'd.
> >>
> >> Hm, r223765 happend before the RELENG_9 Branchpoint.
> >> So maybe the Fix was not complete?
> >
> > See thread from earlier this month on freebsd-pf
> >
>
> The Thread suggests:
>
> * Matt Lager [Thu, 12 Jan 2012 15:48:23 -0700]:
> > So it looks likeI can comment out this code in
> > /usr/src/sys/contrib/pf/net/pf.c:
> >
> > /* mismatch. must not happen. */
> > printf("pf: state key linking mismatch! dir=3D%s, "
> > "if=3D%s, stored af=3D%u, a0: ",
> > dir =3D=3D PF_OUT ? "OUT" : "IN", kif->pfik_name, a=
->af);
> >
> > When this error occurs, I guess for valid reasons, does PF drop packets
> > or do something else with them, or is this purely an information notic=
e?
>
> I can confirm that removing this printf, brings back the performance for
> me.
>
>
Probably a sysctl to disable this should be provided.
There might be unexpected consequences from this and the better fix is to
find the section where the mbuf is being reused.
> Please fix :)
>
> _______________________________________________
> freebsd-pf at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-pf
> To unsubscribe, send any mail to "freebsd-pf-unsubscribe at freebsd.org"
>
--=20
Ermal
--90e6ba6e89dc108bd804b7302abd
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
<br><br>
<div class=3D"gmail_quote">On Sun, Jan 22, 2012 at 11:41 AM, Tilman Keskin=
=F6z <span dir=3D"ltr"><<a href=3D"mailto:arved at freebsd.org">arved at freeb=
sd.org</a>></span> wrote:<br>
<blockquote style=3D"BORDER-LEFT:#ccc 1px solid;MARGIN:0px 0px 0px 0.8ex;PA=
DDING-LEFT:1ex" class=3D"gmail_quote">
<div class=3D"im">* Bjoern A. Zeeb [Sat, 21 Jan 2012 21:01:41 +0000]:<br>&g=
t;<br></div>
<div class=3D"im">> On 21. Jan 2012, at 20:52 , Tilman Keskin=F6z wrote:=
<br>><br>>><br>>> On Jan 21, 2012, at 21:01 , Fabian Keil wr=
ote:<br>>><br>>>> Tilman Keskin=F6z <arved at FreeBSD.org>=
; wrote:<br>
>>><br>>>>> Same here.<br>>>>><br>>>=
>> Also Fabian Keil reported this in<br>>>>> <a href=3D"h=
ttp://lists.freebsd.org/pipermail/freebsd-current/2011-July/025696.html" ta=
rget=3D"_blank">http://lists.freebsd.org/pipermail/freebsd-current/2011-Jul=
y/025696.html</a><br>
>>><br>>>> This has been fixed in CURRENT shortly thereaf=
ter:<br>>>> <a href=3D"http://lists.freebsd.org/pipermail/freebsd-=
pf/2011-July/006199.html" target=3D"_blank">http://lists.freebsd.org/piperm=
ail/freebsd-pf/2011-July/006199.html</a><br>
>>><br>>>> Maybe the fix hasn't been MFC'd.<br>&g=
t;><br>>> Hm, r223765 happend before the RELENG_9 Branchpoint.<br>=
>> So maybe the Fix was not complete?<br>><br>> See thread from=
earlier this month on freebsd-pf<br>
><br><br></div>
<div class=3D"im">The Thread suggests:<br><br>* Matt Lager [Thu, 12 Jan 201=
2 15:48:23 -0700]:<br>> So it looks likeI can comment out this code in<b=
r>> /usr/src/sys/contrib/pf/net/pf.c:<br>><br>> =A0 =A0 =A0 =A0 =
=A0 =A0 =A0 =A0 /* mismatch. must not happen. */<br>
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 printf("pf: state key linking mis=
match! dir=3D%s, "<br>> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 &qu=
ot;if=3D%s, stored af=3D%u, a0: ",<br>> =A0 =A0 =A0 =A0 =A0 =A0 =A0=
=A0 =A0 =A0 dir =3D=3D PF_OUT ? "OUT" : "IN", kif->=
pfik_name, a->af);<br>
><br>> When this error occurs, I guess for valid reasons, does PF dro=
p packets<br>> or do something else with them, or is this purely an info=
rmation =A0notice?<br><br>I can confirm that removing this printf, brings b=
ack the performance for me.<br>
<br></div></blockquote>
<div>=A0</div>
<div>=A0</div>
<div>Probably a sysctl to disable this should be provided.</div>
<div>There might be unexpected consequences from this and the better fix is=
to find the section where the mbuf is being reused.</div>
<div>=A0</div>
<blockquote style=3D"BORDER-LEFT:#ccc 1px solid;MARGIN:0px 0px 0px 0.8ex;PA=
DDING-LEFT:1ex" class=3D"gmail_quote">
<div class=3D"im">Please fix :)<br><br></div>
<div>
<div></div>
<div class=3D"h5">_______________________________________________<br><a hre=
f=3D"mailto:freebsd-pf at freebsd.org">freebsd-pf at freebsd.org</a> mailing list=
<br><a href=3D"http://lists.freebsd.org/mailman/listinfo/freebsd-pf" target=
=3D"_blank">http://lists.freebsd.org/mailman/listinfo/freebsd-pf</a><br>
To unsubscribe, send any mail to "<a href=3D"mailto:freebsd-pf-unsubsc=
ribe at freebsd.org">freebsd-pf-unsubscribe at freebsd.org</a>"<br></div></d=
iv></blockquote></div><br><br clear=3D"all"><br>-- <br>Ermal<br>
--90e6ba6e89dc108bd804b7302abd--
More information about the freebsd-pf
mailing list