Date: Fri, 12 Jun 2015 09:29:22 -0300 From: Christopher Forgeron <csforgeron@gmail.com> To: Cs <bimmer@field.hu> Cc: FreeBSD Net <freebsd-net@freebsd.org> Subject: Re: FreeBSD 10.1-REL - network unaccessible after high traffic Message-ID: <CAB2_NwA9i-wMXGH2%2BcP9SWxDMNomFRjoVP25hsGWaTDGjBxFTw@mail.gmail.com> In-Reply-To: <557AB1BB.60502@field.hu> References: <374339249.53058039.1433681874571.JavaMail.root@uoguelph.ca> <55744F28.5000402@field.hu> <CAB2_NwA-D7bH47=Qkf9QLF3=mZOQBVo81bUsQzQr02W9U4vHMA@mail.gmail.com> <557AB1BB.60502@field.hu>
next in thread | previous in thread | raw e-mail | index | archive | help
Well, even at low speed it could drop due to memory from what I've seen. What was the last line from vmstat 5 before it locked up? I find that the em driver isn't crap, but there is a deeper problem inside of FreeBSD that is being exposed now - For me it's due to faster network connections. Are you using rsync to move the files? On Fri, Jun 12, 2015 at 7:17 AM, Cs <bimmer@field.hu> wrote: > it seems it's not memory related. Server just died a few minutes ago > during transporting the backup (400GB) around 800Mbps speed.. > will disable remote backup, it's a shame that em driver is such a crap. > > > 2015.06.08. 5:01 keltez=C3=A9ssel, Christopher Forgeron =C3=ADrta: > >> You know what helped me: >> >> 'vmstat 5' >> >> Leave that running. If the last thing on the console after a crash/hang = is >> vmstat showing 8k of memory left, then you're in the same problem-park a= s >> me. >> >> My 10.1 96GiB RAM box is chewing ~8 GiB of RAM in less than 5 seconds, a= nd >> then crashing/panicking/hanging. >> >> There's others with this issues if you search for it; a sysctl >> to vm.v_free_min to double or triple that value may help, but first let = us >> know if that's what is bonking your sever. >> >> >> >> On Sun, Jun 7, 2015 at 11:03 AM, Cs <bimmer@field.hu> wrote: >> >> ok, just lowered it to 1500 but please also note that it was on 1500 fo= r >>> 2 >>> years >>> >>> 2015.06.07. 14:57 keltez=C3=A9ssel, Rick Macklem =C3=ADrta: >>> >>> Since disabling TSO didn't help, you could try dropping to 1500mtu >>>> on both interfaces. Some people run into problems when 9K jumbo cluste= rs >>>> fragment the kernel address space used to allocate mbufs. >>>> >>>> Good luck with it, rick >>>> >>>> ----- Original Message ----- >>>> >>>> Hi All, >>>>> >>>>> It worked fine for two weeks but I had a network outage 2 days ago >>>>> then >>>>> today. Tried to disable rxcsum and txcsum after the first one, didn't >>>>> help. Don't know what else to do it's a shame that I can't use this >>>>> card >>>>> with fbsd i REALLY don't want to install linux instead but my >>>>> production >>>>> servers outages are not welcomed by the customers.. >>>>> >>>>> 2015.05.26. 10:36 keltez=C3=A9ssel, Cs =C3=ADrta: >>>>> >>>>> Thanks Mark, good idea. I found this thread which is exactly the >>>>>> same >>>>>> problem as mine: >>>>>> >>>>>> >>>>>> https://forums.freebsd.org/threads/workaround-freebsd-10-1-sudden-ne= twork-down.49264/ >>>>>> >>>>>> Will see if it helps in a couple weeks. >>>>>> >>>>>> Regards, >>>>>> Csaba >>>>>> >>>>>> 2015.05.26. 10:30 keltez=C3=A9ssel, Mark Schouten =C3=ADrta: >>>>>> >>>>>> Oh, didn't see your lowest remark. Then, the next thing that comes >>>>>>> past here a few times per week is 'Try disabling TSO'. >>>>>>> >>>>>>> >>>>>>> Met vriendelijke groeten, >>>>>>> >>>>>>> -- >>>>>>> Kerio Operator in de Cloud? https://www.kerioindecloud.nl/ >>>>>>> Mark Schouten | Tuxis Internet Engineering >>>>>>> KvK: 61527076 | http://www.tuxis.nl/ >>>>>>> T: 0318 200208 | info@tuxis.nl >>>>>>> >>>>>>> >>>>>>> >>>>>>> Van: Cs <bimmer@field.hu> >>>>>>> Aan: Mark Schouten <mark@tuxis.nl> >>>>>>> Cc: <freebsd-net@freebsd.org> >>>>>>> Verzonden: 25-5-2015 11:12 >>>>>>> Onderwerp: Re: FreeBSD 10.1-REL - network unaccessible after >>>>>>> high >>>>>>> traffic >>>>>>> >>>>>>> It was on 1500 for ~3 years :) >>>>>>> Regards, >>>>>>> Csaba >>>>>>> On May 25, 2015, 10:30, at 10:30, Mark Schouten >>>>>>> <mark@tuxis.nl> >>>>>>> wrote: >>>>>>> >>>>>>> Try lowering your mtu to 1500, that worked miracles for me.. >>>>>>>> >>>>>>>> -- >>>>>>>> Mark Schouten >>>>>>>> Tuxis Internet Engineering >>>>>>>> mark@tuxis.nl / 0318 200208 >>>>>>>> >>>>>>>> On 25 May 2015, at 09:36, "Cs" <bimmer@field.hu> wrote: >>>>>>>> >>>>>>>>> Hi all, >>>>>>>>> I have two FreeBSd 10.1-RELEASE servers connected to each >>>>>>>>> other. >>>>>>>>> They >>>>>>>>> >>>>>>>>> were connected via cross link, but they are connected to a cisco >>>>>>>> switch >>>>>>>> now (the problem was the same with cross link too). When >>>>>>>> transferring >>>>>>>> huge files (50-500GB backup files) via Gigabit (it is important!) >>>>>>>> the >>>>>>>> network randomly dies. The backup runs every day/week and >>>>>>>> sometimes the >>>>>>>> connection is ok for months sometimes it happens twice a week. >>>>>>>> When the >>>>>>>> network dies I can log in to the server via IPMI and use the >>>>>>>> console >>>>>>>> everything is OK, but can't send anything out on the network. >>>>>>>> ifconfig >>>>>>>> em0 down/up doesn't help nor netif restart. The problem never >>>>>>>> occured >>>>>>>> when I used 100Mbit connection between them, but it was 3com NIC >>>>>>>> (xl), >>>>>>>> gigabit adapter is Intel (em0). When I limit the transfer rate >>>>>>>> (rsync >>>>>>>> bandwith limit or ipfw pipe) the problem is much more rare. >>>>>>>> >>>>>>>> I tried to set these tuning parameters on both servers with >>>>>>>>> different >>>>>>>>> >>>>>>>>> buffer size but nothing helped: >>>>>>>> >>>>>>>> # cat /etc/sysctl.conf >>>>>>>>> security.bsd.see_other_uids=3D0 >>>>>>>>> net.inet.tcp.recvspace=3D512000 >>>>>>>>> net.route.netisr_maxqlen=3D2048 >>>>>>>>> kern.ipc.nmbclusters=3D1310720 >>>>>>>>> net.inet.tcp.sendbuf_max=3D16777216 >>>>>>>>> net.inet.tcp.recvbuf_max=3D16777216 >>>>>>>>> kern.ipc.soacceptqueue=3D32768 >>>>>>>>> # cat /boot/loader.conf >>>>>>>>> geom_mirror_load=3D"YES" # RAID1 disk driver (see gmirror(8)) >>>>>>>>> ipfw_load=3D"YES" >>>>>>>>> net.inet.ip.fw.default_to_accept=3D1 >>>>>>>>> kern.maxusers=3D4096 >>>>>>>>> accf_data_load=3D"YES" >>>>>>>>> The duplex settings are identical on both servers. >>>>>>>>> Server A: >>>>>>>>> em1: flags=3D8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric = 0 >>>>>>>>> mtu >>>>>>>>> >>>>>>>>> 9000 >>>>>>>> >>>>>>>> >>>>>>>> options=3D4219b<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,VLAN_HWCSUM,= TSO4,WOL_MAGIC,VLAN_HWTSO> >>>>>>>> >>>>>>>> >>>>>>>> ether 00:25:90:24:52:66 >>>>>>>> >>>>>>>>> inet x.x.x.x netmask 0xfffffe00 broadcast x.x.x.x >>>>>>>>> nd6 options=3D29<PERFORMNUD,IFDISABLED,AUTO_LINKLOCAL> >>>>>>>>> media: Ethernet autoselect (1000baseT <full-duplex>) >>>>>>>>> status: active >>>>>>>>> Server B: >>>>>>>>> em0: flags=3D8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric = 0 >>>>>>>>> mtu >>>>>>>>> >>>>>>>>> 9000 >>>>>>>> >>>>>>>> >>>>>>>> options=3D4219b<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,VLAN_HWCSUM,= TSO4,WOL_MAGIC,VLAN_HWTSO> >>>>>>>> >>>>>>>> >>>>>>>> ether 00:30:48:dd:fe:3e >>>>>>>> >>>>>>>>> inet x.x.x.x netmask 0xfffffe00 broadcast x.x.x.x >>>>>>>>> nd6 options=3D29<PERFORMNUD,IFDISABLED,AUTO_LINKLOCAL> >>>>>>>>> media: Ethernet autoselect (1000baseT <full-duplex>) >>>>>>>>> status: active >>>>>>>>> Today I tried to set mtu to 9000 but in tcpdump I see that >>>>>>>>> during >>>>>>>>> scp >>>>>>>>> >>>>>>>>> it is still 1500: >>>>>>>> >>>>>>>> x.x.x.x.222 > x.x.x.x.37612: Flags [.], cksum 0xb6ee >>>>>>>>> (incorrect -> >>>>>>>>> >>>>>>>>> 0xda6f), seq 35749, ack 113701596, win 7986, options [nop,nop,TS >>>>>>>> val >>>>>>>> 3103966325 ecr 853712893], length 0 >>>>>>>> >>>>>>>> 09:27:33.912354 IP (tos 0x8, ttl 64, id 1028, offset 0, flags >>>>>>>>> [DF], >>>>>>>>> >>>>>>>>> proto TCP (6), length 1500) >>>>>>>> >>>>>>>> 09:27:33.912358 IP (tos 0x8, ttl 64, id 1029, offset 0, flags >>>>>>>>> [DF], >>>>>>>>> >>>>>>>>> proto TCP (6), length 1500) >>>>>>>> >>>>>>>> Any ideas? Thanks guys! >>>>>>>>> _______________________________________________ >>>>>>>>> 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" >>>>>>>> >>>>>>>> _______________________________________________ >>>>>>> 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" >>>>>>> >>>>>>> >>>>>>> _______________________________________________ >>>>>>> >>>>>> 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" >>>>>> >>>>>> _______________________________________________ >>>>> 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" >>>>> >>>>> _______________________________________________ >>> 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" >>> >>> _______________________________________________ >> 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" >> > > _______________________________________________ > 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?CAB2_NwA9i-wMXGH2%2BcP9SWxDMNomFRjoVP25hsGWaTDGjBxFTw>