Re: 9.0-RC1 panic in tcp_input: negative winow.
--NP+fiIeNjDlB/E6h
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
On Sun, Oct 23, 2011 at 08:10:38AM +0200, Pawel Jakub Dawidek wrote:
> On Sun, Oct 23, 2011 at 12:35:15PM +1100, Lawrence Stewart wrote:
> > On 10/22/11 19:49, Pawel Jakub Dawidek wrote:
> > > The panic message says:
> > >
> > > panic: tcp_input negative window: tp 0xfffffe007763e000 rcv_nxt 3718=
269252 rcv_adv 3718268291
> > >
> > > I only have picture of the backtrace:
> > >
> > > http://people.freebsd.org/~pjd/misc/panic_negative_window.jpg
> > >
> >=20
> > ewww that is not good. Can you give us any more information about the=
=20
> > machine and what it's doing? Is it terminating TCP connections from the=
=20
> > internet at large or only local LAN (i.e. is there likely to be packet=
=20
> > loss happening)? Are you doing TSO or LRO? Do you have any non-default=
=20
> > tuning in place?
>=20
> It is my local file server. It is doing NFS and AFP over LAN and also
> downloads files from the internet. It is triggered after few hours.
> I changed the KASSERT() into printf() and added printing 'win' variable
> and this is what got logged during the night:
>=20
> 05:16:24 tcp_input negative window: tp 0xfffffe0026772b70 rcv_nxt 1107827=
269 rcv_adv 1107826256 win=3D242
> 05:16:29 tcp_input negative window: tp 0xfffffe0026772b70 rcv_nxt 1107833=
451 rcv_adv 1107832977 win=3D880
> 05:16:41 tcp_input negative window: tp 0xfffffe0026772b70 rcv_nxt 1107849=
563 rcv_adv 1107848860 win=3D639
> 05:20:02 tcp_input negative window: tp 0xfffffe0026772b70 rcv_nxt 1108108=
230 rcv_adv 1108107331 win=3D567
> 05:24:30 tcp_input negative window: tp 0xfffffe0026772b70 rcv_nxt 1108433=
302 rcv_adv 1108432272 win=3D974
> 05:24:46 tcp_input negative window: tp 0xfffffe0026772b70 rcv_nxt 1108450=
385 rcv_adv 1108450060 win=3D751
> 05:26:44 tcp_input negative window: tp 0xfffffe0026772b70 rcv_nxt 1108574=
818 rcv_adv 1108573851 win=3D71
> 05:28:03 tcp_input negative window: tp 0xfffffe0026772b70 rcv_nxt 1108654=
103 rcv_adv 1108653166 win=3D0
> 05:28:43 tcp_input negative window: tp 0xfffffe0026772b70 rcv_nxt 1108692=
396 rcv_adv 1108691451 win=3D0
> 05:30:06 tcp_input negative window: tp 0xfffffe0026772b70 rcv_nxt 1108781=
258 rcv_adv 1108780372 win=3D235
> 05:35:05 tcp_input negative window: tp 0xfffffe0026772b70 rcv_nxt 1109067=
578 rcv_adv 1109067335 win=3D663
> 05:37:03 tcp_input negative window: tp 0xfffffe0026772b70 rcv_nxt 1109180=
403 rcv_adv 1109179411 win=3D0
> 05:41:03 tcp_input negative window: tp 0xfffffe0026772b70 rcv_nxt 1109428=
265 rcv_adv 1109427375 win=3D170
>=20
> And the systems seems to be fine.
>=20
> I'm happy to test patches, but one round would take 24h.
>=20
> My suggestion would be that if we won't be able to fix it before 9.0,
> we should turn this assertion off, as the system seems to be able to
> recover.
Shipped kernels have all assertions turned off.
--NP+fiIeNjDlB/E6h
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.18 (FreeBSD)
iEYEARECAAYFAk6j0/0ACgkQC3+MBN1Mb4h5iwCfQa26LAP0gzvVdmSIiR9rLNvj
5UsAnRPP8tZxdYYn9jOXOHo2pvnPM0bJ
=/lze
-----END PGP SIGNATURE-----
--NP+fiIeNjDlB/E6h--
討論串 (同標題文章)
完整討論串 (本文為第 6 之 21 篇)：