Re: LACP LAGG device problems
On Sat, 7/20/13, isp wrote: Subject: LACP LAGG device problems To: freebsd-net@freebsd.org Date: Saturday, July 20, 2013, 10:04 AM Hi! Can anybody tell me, is there any plans to improve LAGG(802.3ad) device driver in FreeBSD? It will be greate to have a possibility to set LACP mode (active/passive) and system priority. Also there is no way to set hashing algorithm and master interface (port). And we can't see any information about our neighbor. The same function in Linux is named Bonding and it is much more better. I realy can donate some money to those who can make this improvements. Best regards. > ___ Why are you using LAGG when 10g cards are like $350? It's not a peering protocol nor it is PTP; can you see your "peer" info on an ethernet? Bonding is a late 90s concept designed to connect 2 slow links to get higher speeds, back in the day when 100Mb/s was ambitious. The point of LAGG is that it's transparent; you can load balance traffic to multiple hosts or create a redundant link without having to have equipment running some special applications, or any special logic above the LAGG device. Describing how you are using LAGG (and why) might be better than just asking for "improvements". BC ___ freebsd-net@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
Re: LACP LAGG device problems
Hah!
I'm pushing 20GE out using lagg right now (and fixing the er, amusing
behaviour of doing so.) I'm aiming to hit 40 once I get hardware that
doesn't get upset pushing that many bits. The netops people at ${JOB}
also point out that even today switches occasionally get confused and
"crash" a switchport. Ew.
So yes, there are people using lagg, both for failover and throughput reasons.
I'm working on debugging/statistics right now as part of general "why
are things behaving crappy" debugging. I'll see about improving some
of the peer reporting at the same time.
-adrian
On 21 July 2013 06:03, Barney Cordoba wrote:
>
>
> On Sat, 7/20/13, isp wrote:
>
> Subject: LACP LAGG device problems
> To: freebsd-net@freebsd.org
> Date: Saturday, July 20, 2013, 10:04 AM
>
>
>
>
> Hi! Can anybody tell me, is there any plans to improve
> LAGG(802.3ad)
> device driver in FreeBSD?
> It will be greate to have a possibility to set LACP mode
> (active/passive)
> and system priority.
> Also there is no way to set hashing algorithm and master
> interface
> (port).
> And we can't see any information about our neighbor.
> The same function in Linux is named Bonding and it is much
> more better.
> I realy can donate some money to those who can make this
> improvements.
> Best regards.
>
> >
> ___
>
> Why are you using LAGG when 10g cards are like $350? It's not
> a peering protocol nor it is PTP; can you see your "peer" info on
> an ethernet?
>
> Bonding is a late 90s concept designed to connect 2 slow links to
> get higher speeds, back in the day when 100Mb/s was ambitious.
> The point of LAGG is that it's transparent; you can load balance
> traffic to multiple hosts or create a redundant link without having
> to have equipment running some special applications, or any special
> logic above the LAGG device.
>
> Describing how you are using LAGG (and why) might be better
> than just asking for "improvements".
>
> BC
>
> ___
> freebsd-net@freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-net
> To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
Re: LACP LAGG device problems
On Sun, Jul 21, 2013 at 6:49 AM, Adrian Chadd wrote:
> Hah!
>
> I'm pushing 20GE out using lagg right now (and fixing the er, amusing
> behaviour of doing so.) I'm aiming to hit 40 once I get hardware that
> doesn't get upset pushing that many bits. The netops people at ${JOB}
> also point out that even today switches occasionally get confused and
> "crash" a switchport. Ew.
>
> So yes, there are people using lagg, both for failover and throughput
> reasons.
>
> I'm working on debugging/statistics right now as part of general "why
> are things behaving crappy" debugging. I'll see about improving some
> of the peer reporting at the same time.
>
>
>
> -adrian
>
> On 21 July 2013 06:03, Barney Cordoba wrote:
> >
> >
> > On Sat, 7/20/13, isp wrote:
> >
> > Subject: LACP LAGG device problems
> > To: freebsd-net@freebsd.org
> > Date: Saturday, July 20, 2013, 10:04 AM
> >
> >
> >
> >
> > Hi! Can anybody tell me, is there any plans to improve
> > LAGG(802.3ad)
> > device driver in FreeBSD?
> > It will be greate to have a possibility to set LACP mode
> > (active/passive)
> > and system priority.
> > Also there is no way to set hashing algorithm and master
> > interface
> > (port).
> > And we can't see any information about our neighbor.
> > The same function in Linux is named Bonding and it is much
> > more better.
> > I realy can donate some money to those who can make this
> > improvements.
> > Best regards.
> >
> > >
> > ___
> >
> > Why are you using LAGG when 10g cards are like $350? It's not
> > a peering protocol nor it is PTP; can you see your "peer" info on
> > an ethernet?
> >
> > Bonding is a late 90s concept designed to connect 2 slow links to
> > get higher speeds, back in the day when 100Mb/s was ambitious.
> > The point of LAGG is that it's transparent; you can load balance
> > traffic to multiple hosts or create a redundant link without having
> > to have equipment running some special applications, or any special
> > logic above the LAGG device.
> >
> > Describing how you are using LAGG (and why) might be better
> > than just asking for "improvements".
> >
> > BC
>
I am aware of at least one case where 100G WAN links are being LAGGed
today. Only two ATM, but 4x100G is on the horizon. I suspect 4x100G or even
more is already in place in the data center, but I have no actual
knowledge. and 100G i still quite a bit more than $320 per port. And that
is ignoring the cost of 100G routing, switching, and optical gear.
LAGG is not going away any time soon. I'm sure we will see Nx400G as soon
as 400G Ethernet is available.
--
R. Kevin Oberman, Network Engineer
E-mail: rkober...@gmail.com
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
Re[2]: LACP LAGG device problems
I'm using LACP becasue: 1. 10G cards are much more expensive than 350$. They cost nearly 700$-800$. 2. I don't have switches with 10G interfaces. They are also very expensive. --- Message --- From: "Barney Cordoba" Date: 21 July 2013, 16:03:23 On Sat, 7/20/13, isp < ml...@ukr.net > wrote: Subject: LACP LAGG device problems To: freebsd-net@freebsd.org Date: Saturday, July 20, 2013, 10:04 AM Hi! Can anybody tell me, is there any plans to improve LAGG(802.3ad) device driver in FreeBSD? It will be greate to have a possibility to set LACP mode (active/passive) and system priority. Also there is no way to set hashing algorithm and master interface (port). And we can't see any information about our neighbor. The same function in Linux is named Bonding and it is much more better. I realy can donate some money to those who can make this improvements. Best regards. > ___ Why are you using LAGG when 10g cards are like $350? It's not a peering protocol nor it is PTP; can you see your "peer" info on an ethernet? Bonding is a late 90s concept designed to connect 2 slow links to get higher speeds, back in the day when 100Mb/s was ambitious. The point of LAGG is that it's transparent; you can load balance traffic to multiple hosts or create a redundant link without having to have equipment running some special applications, or any special logic above the LAGG device. Describing how you are using LAGG (and why) might be better than just asking for "improvements". BC ___ freebsd-net@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
Re: LACP LAGG device problems
I wasn't referring to science projects. Nor did I say it wasn't useful.
Only that 10g is cheap now and quite a bit better. LAGG isn't perfect.
- Original Message -
From: Adrian Chadd
To: Barney Cordoba
Cc: freebsd-net@freebsd.org; isp
Sent: Sunday, July 21, 2013 9:49 AM
Subject: Re: LACP LAGG device problems
Hah!
I'm pushing 20GE out using lagg right now (and fixing the er, amusing
behaviour of doing so.) I'm aiming to hit 40 once I get hardware that
doesn't get upset pushing that many bits. The netops people at ${JOB}
also point out that even today switches occasionally get confused and
"crash" a switchport. Ew.
So yes, there are people using lagg, both for failover and throughput reasons.
I'm working on debugging/statistics right now as part of general "why
are things behaving crappy" debugging. I'll see about improving some
of the peer reporting at the same time.
-adrian
On 21 July 2013 06:03, Barney Cordoba wrote:
>
>
> On Sat, 7/20/13, isp wrote:
>
> Subject: LACP LAGG device problems
> To: freebsd-net@freebsd.org
> Date: Saturday, July 20, 2013, 10:04 AM
>
>
>
>
> Hi! Can anybody tell me, is there any plans to improve
> LAGG(802.3ad)
> device driver in FreeBSD?
> It will be greate to have a possibility to set LACP mode
> (active/passive)
> and system priority.
> Also there is no way to set hashing algorithm and master
> interface
> (port).
> And we can't see any information about our neighbor.
> The same function in Linux is named Bonding and it is much
> more better.
> I realy can donate some money to those who can make this
> improvements.
> Best regards.
>
> >
> ___
>
> Why are you using LAGG when 10g cards are like $350? It's not
> a peering protocol nor it is PTP; can you see your "peer" info on
> an ethernet?
>
> Bonding is a late 90s concept designed to connect 2 slow links to
> get higher speeds, back in the day when 100Mb/s was ambitious.
> The point of LAGG is that it's transparent; you can load balance
> traffic to multiple hosts or create a redundant link without having
> to have equipment running some special applications, or any special
> logic above the LAGG device.
>
> Describing how you are using LAGG (and why) might be better
> than just asking for "improvements".
>
> BC
>
> ___
> freebsd-net@freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-net
> To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
Re: LACP LAGG device problems
Barney,
I now work at netflix. We push >10gig per box. I'm working on making
that much, much more than 10gig. It's not a "science project."
sheesh. :-)
-adrian
On 21 July 2013 09:23, Barney Cordoba wrote:
> I wasn't referring to science projects. Nor did I say it wasn't useful.
> Only that 10g is cheap now and quite a bit better. LAGG isn't perfect.
>
>
> - Original Message -
> From: Adrian Chadd
> To: Barney Cordoba
> Cc: freebsd-net@freebsd.org; isp
> Sent: Sunday, July 21, 2013 9:49 AM
> Subject: Re: LACP LAGG device problems
>
> Hah!
>
> I'm pushing 20GE out using lagg right now (and fixing the er, amusing
> behaviour of doing so.) I'm aiming to hit 40 once I get hardware that
> doesn't get upset pushing that many bits. The netops people at ${JOB}
> also point out that even today switches occasionally get confused and
> "crash" a switchport. Ew.
>
> So yes, there are people using lagg, both for failover and throughput reasons.
>
> I'm working on debugging/statistics right now as part of general "why
> are things behaving crappy" debugging. I'll see about improving some
> of the peer reporting at the same time.
>
>
>
> -adrian
>
> On 21 July 2013 06:03, Barney Cordoba wrote:
>>
>>
>> On Sat, 7/20/13, isp wrote:
>>
>> Subject: LACP LAGG device problems
>> To: freebsd-net@freebsd.org
>> Date: Saturday, July 20, 2013, 10:04 AM
>>
>>
>>
>>
>> Hi! Can anybody tell me, is there any plans to improve
>> LAGG(802.3ad)
>> device driver in FreeBSD?
>> It will be greate to have a possibility to set LACP mode
>> (active/passive)
>> and system priority.
>> Also there is no way to set hashing algorithm and master
>> interface
>> (port).
>> And we can't see any information about our neighbor.
>> The same function in Linux is named Bonding and it is much
>> more better.
>> I realy can donate some money to those who can make this
>> improvements.
>> Best regards.
>>
>> >
>> ___
>>
>> Why are you using LAGG when 10g cards are like $350? It's not
>> a peering protocol nor it is PTP; can you see your "peer" info on
>> an ethernet?
>>
>> Bonding is a late 90s concept designed to connect 2 slow links to
>> get higher speeds, back in the day when 100Mb/s was ambitious.
>> The point of LAGG is that it's transparent; you can load balance
>> traffic to multiple hosts or create a redundant link without having
>> to have equipment running some special applications, or any special
>> logic above the LAGG device.
>>
>> Describing how you are using LAGG (and why) might be better
>> than just asking for "improvements".
>>
>> BC
>>
>> ___
>> freebsd-net@freebsd.org mailing list
>> http://lists.freebsd.org/mailman/listinfo/freebsd-net
>> To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
>
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
Re: LACP LAGG device problems
Adrian, you're killing my spam filter! But yes, our use of FreeBSD at Netflix
is hardly a science project. Http://openconnect.netflix.com
Scott
On Jul 21, 2013, at 1:31 PM, Adrian Chadd wrote:
> Barney,
>
> I now work at netflix. We push >10gig per box. I'm working on making
> that much, much more than 10gig. It's not a "science project."
>
> sheesh. :-)
>
>
>
> -adrian
>
> On 21 July 2013 09:23, Barney Cordoba wrote:
>> I wasn't referring to science projects. Nor did I say it wasn't useful.
>> Only that 10g is cheap now and quite a bit better. LAGG isn't perfect.
>>
>>
>> - Original Message -
>> From: Adrian Chadd
>> To: Barney Cordoba
>> Cc: freebsd-net@freebsd.org; isp
>> Sent: Sunday, July 21, 2013 9:49 AM
>> Subject: Re: LACP LAGG device problems
>>
>> Hah!
>>
>> I'm pushing 20GE out using lagg right now (and fixing the er, amusing
>> behaviour of doing so.) I'm aiming to hit 40 once I get hardware that
>> doesn't get upset pushing that many bits. The netops people at ${JOB}
>> also point out that even today switches occasionally get confused and
>> "crash" a switchport. Ew.
>>
>> So yes, there are people using lagg, both for failover and throughput
>> reasons.
>>
>> I'm working on debugging/statistics right now as part of general "why
>> are things behaving crappy" debugging. I'll see about improving some
>> of the peer reporting at the same time.
>>
>>
>>
>> -adrian
>>
>> On 21 July 2013 06:03, Barney Cordoba wrote:
>>>
>>>
>>> On Sat, 7/20/13, isp wrote:
>>>
>>> Subject: LACP LAGG device problems
>>> To: freebsd-net@freebsd.org
>>> Date: Saturday, July 20, 2013, 10:04 AM
>>>
>>>
>>>
>>>
>>> Hi! Can anybody tell me, is there any plans to improve
>>> LAGG(802.3ad)
>>> device driver in FreeBSD?
>>> It will be greate to have a possibility to set LACP mode
>>> (active/passive)
>>> and system priority.
>>> Also there is no way to set hashing algorithm and master
>>> interface
>>> (port).
>>> And we can't see any information about our neighbor.
>>> The same function in Linux is named Bonding and it is much
>>> more better.
>>> I realy can donate some money to those who can make this
>>> improvements.
>>> Best regards.
>>>
>>> ___
>>>
>>> Why are you using LAGG when 10g cards are like $350? It's not
>>> a peering protocol nor it is PTP; can you see your "peer" info on
>>> an ethernet?
>>>
>>> Bonding is a late 90s concept designed to connect 2 slow links to
>>> get higher speeds, back in the day when 100Mb/s was ambitious.
>>> The point of LAGG is that it's transparent; you can load balance
>>> traffic to multiple hosts or create a redundant link without having
>>> to have equipment running some special applications, or any special
>>> logic above the LAGG device.
>>>
>>> Describing how you are using LAGG (and why) might be better
>>> than just asking for "improvements".
>>>
>>> BC
>>>
>>> ___
>>> freebsd-net@freebsd.org mailing list
>>> http://lists.freebsd.org/mailman/listinfo/freebsd-net
>>> To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
> ___
> freebsd-net@freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-net
> To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
Re: LACP LAGG device problems
> Adrian, you're killing my spam filter! But yes, our use of FreeBSD at > Netflix is hardly a science project. Http://openconnect.netflix.com With my ISP hat on, I expect us to continue using LAGG with 10G member links for many years - simply because 100G is so expensive. One can hope that CFP2 and CFP4 will improve matters somewhat (though it won't help much for the DWDM transport side). Steinar Haug, Nethelp consulting, sth...@nethelp.no ___ freebsd-net@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
Re: Duplicate Address Detection misfire?
I've further narrowed this down. According to the output: em0 DAD detected duplicate IPv6 address fe80:2::250:56ff:fe2e:45fd: NS in/out=2/1, NA in=0 ... FreeBSD *thinks* it has transmitted one and received 2 solicitations. The packet dump shows two solicitations (which would, if it were not bogus, indicate that another machine was booting at the exact same time trying to use the same link-local address). The question becomes: is vmware duplicating the packet, or is FreeBSD? IE: driver problem with em0 and vmware? I'm not completely sure how to debug this. On Thu, Jul 18, 2013 at 9:21 PM, Zaphod Beeblebrox wrote: > > > > On Sun, Jun 30, 2013 at 10:39 PM, Kevin Day wrote: > >> >> On Jun 30, 2013, at 6:48 PM, Zaphod Beeblebrox wrote: >> >> > I have a FreeBSD 9.1-RELEASE vmware guest running. It is using the >> > "bridged" type of networking with VMWare. It gets it's IPv4 address >> from >> > DHCP (successfully) and then fails to initialize IPv6. The relevant >> > rc.conf is: >> > >> > ipv6_activate_all_interfaces="YES" >> > ifconfig_em0_ipv6="inet6 accept_rtadv" >> > ip6addrctl_verbose="YES" >> > >> > The console output says: >> > >> > em0: DAD detected duplicate IPv6 address fe80:2::20c:29ff:fe0a:3989: NS >> > in/out=2/1, NA in=0 >> > em0: DAD complete for fe80:2::20c:29ff:fe0a:3989 - duplicate found >> > em0: manual intervention required >> > em0: possible hardware address duplication deteted, disable IPv6 >> > >> > And subsequently, em0's nd6 has "IFDISABLED" in it. >> > >> > With wireshark, I see two ICMPv6 neighbor solicitations that are >> identical >> > --- is this the problem? >> > >> > How do I fix this? >> >> Did you copy this VM and have both copies running at once? If so, it >> assigned the same MAC address to each VM. >> >> VMware is suppose to detect this and ask if you "copied" or "moved" the >> VM, and if you say "copied" it will randomly assign a new MAC to the VM. If >> this didn't happen or if you said "moved" when you actually copied it, just >> go in and delete/re-create the network interface in the VM's settings to >> create a new MAC for it. >> >> If that's not the issue, we'd probably need more details about your >> configuration. >> > > To further diagnose, there is only one VM running. To ensure that there > were no duplicates, I reassigned the MAC address in the VMWare > configuration dialogue. Additionally, I tried stopping rtadvd on my router > (no effect) and I tried putting the guest on a "host-only network" > (basically isolated it) --- this clears the problem --- both the link-local > and the static address are assigned. > > Frustrated, I dumped the windows interface that is bridged to the VMWare > guest. When it boots, I see the following: > > 246119:24:16.376027000Vmware_2e:46:fdBroadcastARP42 > Gratuitous ARP for 66.96.20.42 (Request) > 246219:24:16.388241000::ff02::1:ff00:42ICMPv678 > Neighbor Solicitation for 2001:1928:1::42 > 246319:24:16.389065000::ff02::1:ff00:42ICMPv678 > Neighbor Solicitation for 2001:1928:1::42 > 246419:24:16.44413::ff02::16ICMPv6130Multicast > Listener Report Message v2 > 246519:24:16.444605000::ff02::16ICMPv6130Multicast > Listener Report Message v2 > 246619:24:16.594663000::ff02::1:ff2e:46fdICMPv678 > Neighbor Solicitation for fe80::250:56ff:fe2e:46fd > 246719:24:16.595179000::ff02::1:ff2e:46fdICMPv678 > Neighbor Solicitation for fe80::250:56ff:fe2e:46fd > 275319:24:22.274728000Vmware_2e:46:fdBroadcastARP42 > Who has 66.96.20.33? Tell 66.96.20.42 > 275419:24:22.274902000Intel_bc:6f:87Vmware_2e:46:fdARP60 > 66.96.20.33 is at 00:0e:0c:bc:6f:87 > > ... and then it goes on to chatter ipv4-wise as expected. Note that there > are two of each packet. Is that normal? The ethernet source of all these > packets is my vmware guest (save the who-has reply that I copied in). > > ___ freebsd-net@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
