Date: Thu, 12 Jan 2017 10:38:32 -0700 From: Sean Bruno <sbruno@freebsd.org> To: "O. Hartmann" <ohartmann@walstatt.org>, freebsd-current <freebsd-current@freebsd.org> Subject: Re: CURRENT: em0 NIC freezes under heavy I/O on net Message-ID: <655c6935-1bcf-b54f-9298-ddacd184570b@freebsd.org> In-Reply-To: <20170111091643.1d45ab39@freyja.zeit4.iv.bundesimmobilien.de> References: <20170111091643.1d45ab39@freyja.zeit4.iv.bundesimmobilien.de>
next in thread | previous in thread | raw e-mail | index | archive | help
This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --6Rfx0wX8WgAHb1nInntdb2c8vHweCogUj Content-Type: multipart/mixed; boundary="fxtExBu6tsqRUlsEbG2fdKcgwu4WNcKHo"; protected-headers="v1" From: Sean Bruno <sbruno@freebsd.org> To: "O. Hartmann" <ohartmann@walstatt.org>, freebsd-current <freebsd-current@freebsd.org> Message-ID: <655c6935-1bcf-b54f-9298-ddacd184570b@freebsd.org> Subject: Re: CURRENT: em0 NIC freezes under heavy I/O on net References: <20170111091643.1d45ab39@freyja.zeit4.iv.bundesimmobilien.de> In-Reply-To: <20170111091643.1d45ab39@freyja.zeit4.iv.bundesimmobilien.de> --fxtExBu6tsqRUlsEbG2fdKcgwu4WNcKHo Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: quoted-printable On 01/11/17 01:27, O. Hartmann wrote: > Running recent CURRENT (FreeBSD 12.0-CURRENT #5 r311919: Wed Jan 11 08:= 24:28 > CET 2017 amd64), the system freezes when doing a rsync over automounted= > (autofs) NFSv4 filesystem, mounted from another CURRENT server (same re= vision, > but with BCM NICs). >=20 > The host in question is a Fujitsu Celsius M740 equipted with an Intel N= IC: >=20 > [...] > em0: <Intel(R) PRO/1000 Network Connection> port 0xf020-0xf03f mem > 0xfb300000-0xfb31ffff,0xfb339000-0xfb339fff at device 25.0 numa-domain = 0 on > pci1 em0: attach_pre capping queues at 1 em0: using 1024 tx descriptors= and > 1024 rx descriptors em0: msix_init qsets capped at 1 > em0: Unable to map MSIX table=20 > em0: Using an MSI interrupt > em0: allocated for 1 tx_queues > em0: allocated for 1 rx_queues > em0: netmap queues/slots: TX 1/1024, RX 1/1024 > [...] >=20 > The pciconf output reveals: >=20 > em0@pci0:0:25:0: class=3D0x020000 card=3D0x11ed1734 chip=3D0x153= a8086 rev=3D0x05 > hdr=3D0x00 vendor =3D 'Intel Corporation' > device =3D 'Ethernet Connection I217-LM' > class =3D network > subclass =3D ethernet > bar [10] =3D type Memory, range 32, base 0xfb300000, size 131072,= enabled > bar [14] =3D type Memory, range 32, base 0xfb339000, size 4096, e= nabled > bar [18] =3D type I/O Port, range 32, base 0xf020, size 32, enabl= ed > cap 01[c8] =3D powerspec 2 supports D0 D3 current D0 > cap 05[d0] =3D MSI supports 1 message, 64 bit enabled with 1 messag= e > cap 13[e0] =3D PCI Advanced Features: FLR TP >=20 > I have a customized kernel. The NIC has revealed itself all the time as= an > "emX" device (never as igbX). The kernel contains device netmap (if > relevevant). >=20 > The phenomenon: >=20 > Syncing a poudriere repository between to remote hosts, I use rsync on = a NGSv4 > exported filesystem, mounted via AUTOFS. So far, this work two days ago= > perfectly. Since yesterday, syncing brings down the network connection = - the > connection is simply dead. Terminating the rsync, bringing em0 down and= up > again doesn't help much, for short moments, the connection is establish= ed, but > dies within seconds. Restarting via "service netif restart" all network= > services have the same effect: after the desaster, it is impossible for= me to > bring back the NIC/connection to normal, I have to reboot. The same hap= pens > when having heavy network load, but it takes a time and even rsync isn'= t > "deadly" within the same timeframe - it takes sometimes a couple of sec= onds, > another takes only one or two seconds to make the connection die.=20 >=20 > I checked with dd'ing a large file over that connection, it takes sever= al > seconds then to make the connection freezing (so, someone could reprodu= ce iy > not ncessarily using rsync). >=20 > Kind regards, >=20 > oh If you have the time today or tomorrow. Can you please capture 'sysctl dev.em.0' and post it here? In addition, I would like to have this patch tested in your configuration= : https://people.freebsd.org/~sbruno/em_tx_limit.diff Finally, if you have any loader.conf entries for hw.em, please post them as well. sean --fxtExBu6tsqRUlsEbG2fdKcgwu4WNcKHo-- --6Rfx0wX8WgAHb1nInntdb2c8vHweCogUj Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- iQGTBAEBCgB9FiEEuq1GMucSHejSCZfdEgHvyh5yfmQFAlh3vxlfFIAAAAAALgAo aXNzdWVyLWZwckBub3RhdGlvbnMub3BlbnBncC5maWZ0aGhvcnNlbWFuLm5ldEJB QUQ0NjMyRTcxMjFERThEMjA5OTdERDEyMDFFRkNBMUU3MjdFNjQACgkQEgHvyh5y fmTbYQgAyuvoKTCc3dabNG3Nk71hKDWeMqq7Uu0jU1OnC168A7poZbxKM48NZ5Ae n/8yjHic3buqAA+EHRl7nvu781HyWjn5lAi3avn96q+AngkKHm0F4qS7/mYP3LfV gtf9wfKk3DwoDeMDJdbrwI2aQUvH09IYm3yr9JBMbHaFF4rkm04viMgaBVVdI95v gcITlBofaAjmmSc1bIfuykZvk/oBzpBICh++6gpY5tBC8uH637vdFzlpEZciONCE wz+18XmJ3siqTIqwEYDO2gdw73yt03gKGGadAiHNl3Z5zfX2ousFt1eby8IBvs+l xnfpwauYGqt83Zecyui8/QTaqhZEtA== =QvAS -----END PGP SIGNATURE----- --6Rfx0wX8WgAHb1nInntdb2c8vHweCogUj--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?655c6935-1bcf-b54f-9298-ddacd184570b>