Date: Fri, 25 Sep 2009 07:14:21 -0700 (PDT) From: Barney Cordoba <barney_cordoba@yahoo.com> To: Peter Steele <psteele@maxiscale.com>, Nikolay Denev <ndenev@gmail.com> Cc: freebsd-net@freebsd.org Subject: Re: Can lagg0 failback be prevented? Message-ID: <461385.86611.qm@web63904.mail.re1.yahoo.com> In-Reply-To: <C0778F03-C450-4C69-8AF0-4A6A8F9E690F@gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
=0A=0A--- On Thu, 9/24/09, Nikolay Denev <ndenev@gmail.com> wrote:=0A=0A> F= rom: Nikolay Denev <ndenev@gmail.com>=0A> Subject: Re: Can lagg0 failback b= e prevented?=0A> To: "Peter Steele" <psteele@maxiscale.com>=0A> Cc: freebsd= -net@freebsd.org=0A> Date: Thursday, September 24, 2009, 2:32 AM=0A> How lo= ng exactly is that "momentary=0A> network outage"?=0A> If it's longer than = a few (1-5) seconds it may be due to=0A> RSTP or STP enabled on the switch.= =0A> You could try disabling STP on these ports and see if it=0A> will=0A> = start forwarding traffic faster.=0A> =0A> --=0A> Regards,=0A> Nikolay Denev= =0A> =0A> =0A> =0A> =0A> On 16 Sep, 2009, at 03:23 , Peter Steele wrote:=0A= > =0A> > We're using the lag driver to provide automatic=0A> failover in ca= se of a network outage. The default=0A> configuration looks like this:=0A> = > =0A> > lagg0:=0A> flags=3D8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST>=0A= > metric 0 mtu 1500=0A> >=A0 =A0 =A0 =A0=0A> options=3D19b<RXCSUM,TXCSUM,VL= AN_MTU,VLAN_HWTAGGING,VLAN_HWCSUM,TSO4>=0A> >=A0 =A0 =A0 =A0 ether 00:a0:d1= :e3:58:26=0A> >=A0 =A0 =A0 =A0 inet 192.168.17.40 netmask=0A> 0xfffff000 br= oadcast 192.168.31.255=0A> >=A0 =A0 =A0 =A0 inet 192.168.22.11 netmask=0A> = 0xffffff00 broadcast 192.168.22.255=0A> >=A0 =A0 =A0 =A0 media: Ethernet au= toselect=0A> >=A0 =A0 =A0 =A0 status: active=0A> >=A0 =A0 =A0 =A0 laggproto= failover=0A> >=A0 =A0 =A0 =A0 laggport: nfe1=0A> flags=3D0<>=0A> >=A0 =A0 = =A0 =A0 laggport: nfe0=0A> flags=3D5<MASTER,ACTIVE>=0A> > =0A> > If nfe0 wa= s to fail, we get an (almost) automatic=0A> failover to nfe1:=0A> > =0A> > = lagg0:=0A> flags=3D8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST>=0A> metric = 0 mtu 1500=0A> >=A0 =A0 =A0 =A0=0A> options=3D19b<RXCSUM,TXCSUM,VLAN_MTU,VL= AN_HWTAGGING,VLAN_HWCSUM,TSO4>=0A> >=A0 =A0 =A0 =A0 ether 00:a0:d1:e3:58:26= =0A> >=A0 =A0 =A0 =A0 inet 192.168.17.40 netmask=0A> 0xfffff000 broadcast 1= 92.168.31.255=0A> >=A0 =A0 =A0 =A0 inet 192.168.22.11 netmask=0A> 0xffffff0= 0 broadcast 192.168.22.255=0A> >=A0 =A0 =A0 =A0 media: Ethernet autoselect= =0A> >=A0 =A0 =A0 =A0 status: active=0A> >=A0 =A0 =A0 =A0 laggproto failove= r=0A> >=A0 =A0 =A0 =A0 laggport: nfe1=0A> flags=3D4<ACTIVE>=0A> >=A0 =A0 = =A0 =A0 laggport: nfe0=0A> flags=3D1<MASTER>=0A> > =0A> > The problem we're= having is when nfe0 comes online=0A> again, a failback occurs making nfe0 = active again. This=0A> causes a momentary network outage that we want to pr= event.=0A> Is there a way to configure the lagg device to stay with the=0A>= currently active interface, even if the MASTER interface=0A> comes back on= line?=0A> > =0A=0AThe "outage" is because the failover isn't detected until= you get a =0Afailure. When a link down is detected, the backup port is ena= bled. =0ASo there is going to be some traffic loss. There may be packets in= =0Athe transmit queue as well.=0A=0AI asked earlier why you were using fail= over if you really don't care=0Awhich link it uses? If you load balance it = will use both links=0Aunless one of the links is down, in which case it wou= ld use one link.=0AYou'd avoid the problem that seems to concern you.=0A=0A= Barney=0A=0A=0A
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?461385.86611.qm>