Date: Fri, 10 Aug 2012 12:18:54 -0700 From: "EXT-Glatting, Dennis P" <Dennis.P.Glatting@boeing.com> To: Karl Pielorz <kpielorz_lst@tdx.co.uk>, "freebsd-net@freebsd.org" <freebsd-net@freebsd.org> Subject: RE: FreeBSD 9.0-R em0 issues? Message-ID: <0457799DC5A3A5479336B9A1F2E7FC6450E4FF2F5F@XCH-NW-21V.nw.nos.boeing.com> In-Reply-To: <26ABB168B4C8E32E7D127EF9@Octca64MkIV.tdx.co.uk> References: <26ABB168B4C8E32E7D127EF9@Octca64MkIV.tdx.co.uk>
next in thread | previous in thread | raw e-mail | index | archive | help
I've seen this too on a H8DG6-F but changed to the igb interfaces. I am als= o seeing a problem under ESXi 5.0 on an unpatched system but haven't had th= e opportunity to patch (ESXi) it. There I will see these messages under a F= reeBSD instance: nms# uname -a FreeBSD nms 9.1-PRERELEASE FreeBSD 9.1-PRERELEASE #9: Mon Aug 6 22:02:46 P= DT 2012 root@nms:/sys/amd64/compile/SMUNI amd64 nms# grep em0 /var/log/messages Jul 9 15:50:39 nms kernel: em0: <Intel(R) PRO/1000 Legacy Network Connecti= on 1.0.4> port 0x2000-0x203f mem 0xd1020000-0xd103ffff,0xd1000000-0xd100fff= f irq 18 at device 0.0 on pci2 Jul 9 15:50:39 nms kernel: em0: Memory Access and/or Bus Master bits were = not set! Jul 9 15:50:39 nms kernel: em0: Ethernet address: 00:0c:29:a2:36:c2 Jul 10 09:28:49 nms kernel: em0: Watchdog timeout -- resetting Jul 18 16:53:30 nms kernel: em0: Watchdog timeout -- resetting Jul 26 22:37:16 nms kernel: em0: Watchdog timeout -- resetting Aug 6 20:55:01 nms kernel: em0: <Intel(R) PRO/1000 Legacy Network Connecti= on 1.0.4> port 0x2000-0x203f mem 0xd1020000-0xd103ffff,0xd1000000-0xd100fff= f irq 18 at device 0.0 on pci2 Aug 6 20:55:01 nms kernel: em0: Memory Access and/or Bus Master bits were = not set! Aug 6 20:55:01 nms kernel: em0: Ethernet address: 00:0c:29:a2:36:c2 Aug 6 22:16:34 nms kernel: em0: <Intel(R) PRO/1000 Legacy Network Connecti= on 1.0.4> port 0x2000-0x203f mem 0xd1020000-0xd103ffff,0xd1000000-0xd100fff= f irq 18 at device 0.0 on pci2 Aug 6 22:16:34 nms kernel: em0: Memory Access and/or Bus Master bits were = not set! Aug 6 22:16:34 nms kernel: em0: Ethernet address: 00:0c:29:a2:36:c2 Aug 6 22:58:35 nms kernel: em0: <Intel(R) PRO/1000 Legacy Network Connecti= on 1.0.4> port 0x2000-0x203f mem 0xd1020000-0xd103ffff,0xd1000000-0xd100fff= f irq 18 at device 0.0 on pci2 Aug 6 22:58:35 nms kernel: em0: Memory Access and/or Bus Master bits were = not set! Aug 6 22:58:35 nms kernel: em0: Ethernet address: 00:0c:29:a2:36:c2 Aug 8 12:51:50 nms kernel: em0: Watchdog timeout -- resetting Aug 8 22:49:58 nms kernel: em0: <Intel(R) PRO/1000 Legacy Network Connecti= on 1.0.4> port 0x2000-0x203f mem 0xd1020000-0xd103ffff,0xd1000000-0xd100fff= f irq 18 at device 0.0 on pci2 Aug 8 22:49:58 nms kernel: em0: Memory Access and/or Bus Master bits were = not set! Aug 8 22:49:58 nms kernel: em0: Ethernet address: 00:0c:29:a2:36:c2 -----Original Message----- From: owner-freebsd-net@freebsd.org [mailto:owner-freebsd-net@freebsd.org] = On Behalf Of Karl Pielorz Sent: Friday, August 10, 2012 11:55 AM To: freebsd-net@freebsd.org Subject: FreeBSD 9.0-R em0 issues? Hi, Apologies for posting to -net as well - I originally posted this to=20 -hackers, but was advised to re-post it here... A FreeBSD 9.0-R amd64 box - based on a SuperMicro X8DTL-IF Rev. 2.01=20 w/Intel L5630 & 6Gb of RAM seems to have issues with it's onboard NIC (em= =20 driver based - i.e. em0). The machine runs fine - but then suddenly loses all network connectivity.=20 Nothing is logged on the console, or /var/log/messages. Doing an 'infconfig em0 down' then up, doesn't fix it. Rebooting the box=20 does fix it "for a while". Having dug around Google - I've now set=20 "hw.em.enable_msix=3D0" - the box ran the whole of the day with that set,=20 before again - having em0 wedge up. When it does this 'netstat -n -i' returns "silly" figures - i.e. if I catch= =20 it even moments after it's done it - it'll claim to have suffered billions= =20 of input/output and collision errors (huge amounts more than the amount of= =20 traffic that machine would have handled) - e.g. " Name Mtu Network Address Ipkts Ierrs Idrop Opkts=20 Oerrs Coll em0 1500 <Link#5> 00:25:90:31:82:46 355482 10612864185945 0=20 291109 3032246910270 1516123455135 " Before it locks up, all ierr,oerr,coll are zero (i.e. only ipkts/opkts have= =20 incremented). Running sysctl dev.em.0.debug=3D1 spits out on the console: Interface is RUNNING and INACTIVE em0: hw tdh =3D -1, hw tdt =3D -1 em0: hw rdh =3D -1, hw rdt =3D -1 em0: Tx Queue Status =3D 1 em0: Tx descriptors avail =3D 986 em0: Tx Descriptors avail failure =3D 0 em0: RX discarded packets =3D 0 em0: RX Next to Check =3D 844 em0: RX Next to Refresh =3D 843 (I don't like the look of the -1's but I don't know enough about what that= =20 sysctl dumps out to know if that's "bad"). The HP switch it's connected to doesn't seem to log any errors for the port= . Any suggestions on how I can debug this further, or any ideas to try and=20 fix it? You can find the dmesg output here: <http://www.tdx.com/dmesg.txt>; (ifconfig -a is tacked on the end) Regards, -Karl _______________________________________________ freebsd-net@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "freebsd-net-unsubscribe@freebsd.org"
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?0457799DC5A3A5479336B9A1F2E7FC6450E4FF2F5F>