Can't get LACP to work and mDNS error

[ddaenen1]

Hello, i decided to give FreeNAS a try after many years of consumer NAS'es as i wanted a more scaleable and future-proof solution.

Before i get into the issue details, here my setup:

Supermicro server with X8SIL-F motherboard, Intel Xeon X3440 and 16Gb ECC RDIMM ram and 2x2TB Toshiba HDD's connected to a Netgear GS724Tv3 managed switch, hooked up to a Mikrotik RB3011 router on 1000/40 fibernet. Topology below.

I have 2 main issues that i struggle finding a solution for in the forums or other locations.

1. LACP configuration: every time i try to configure LACP, my connection drops. My approach is pretty straight forward:

- configure LACP in the switch, hook up the UTP cables to both the switch and the server. Configure LAG in FreeNAS following the manual. Once i save the config, the "waiting" window pops up and the connection is dropped and it doesn't come up again until i delete the LAG and reboot FreeNAS. What am i doing wrong? I read a couple of recommendations about first plugging in one cable, set up the LAG and the connect the second one.

Here is my ifconfig at the moment. Switch is currently still configured to LACP but only one cable plugged in:

root@FreeNAS-Server[~]# ifconfig
em0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500
options=209b<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,VLAN_HWCSUM,WOL_MAGIC>
ether 00:25:90:34:74:54
hwaddr 00:25:90:34:74:54
nd6 options=9<PERFORMNUD,IFDISABLED>
media: Ethernet autoselect
status: no carrier
em1: flags=8943<UP,BROADCAST,RUNNING,PROMISC,SIMPLEX,MULTICAST> metric 0 mtu 1500
options=2098<VLAN_MTU,VLAN_HWTAGGING,VLAN_HWCSUM,WOL_MAGIC>
ether 00:25:90:34:74:55
hwaddr 00:25:90:34:74:55
inet 192.168.1.52 netmask 0xffffff00 broadcast 192.168.1.255
nd6 options=9<PERFORMNUD,IFDISABLED>
media: Ethernet autoselect (1000baseT <full-duplex>)
status: active
lo0: flags=8049<UP,LOOPBACK,RUNNING,MULTICAST> metric 0 mtu 16384
options=600003<RXCSUM,TXCSUM,RXCSUM_IPV6,TXCSUM_IPV6>
inet6 ::1 prefixlen 128
inet6 fe80::1%lo0 prefixlen 64 scopeid 0x3
inet 127.0.0.1 netmask 0xff000000
nd6 options=21<PERFORMNUD,AUTO_LINKLOCAL>
groups: lo
bridge0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500
ether 02:4d:33:5b:21:00
nd6 options=1<PERFORMNUD>
groups: bridge
id 00:00:00:00:00:00 priority 32768 hellotime 2 fwddelay 15
maxage 20 holdcnt 6 proto rstp maxaddr 2000 timeout 1200
root id 00:00:00:00:00:00 priority 32768 ifcost 0 port 0
member: vnet0:3 flags=143<LEARNING,DISCOVER,AUTOEDGE,AUTOPTP>
ifmaxaddr 0 port 5 priority 128 path cost 2000
member: em1 flags=143<LEARNING,DISCOVER,AUTOEDGE,AUTOPTP>
ifmaxaddr 0 port 2 priority 128 path cost 20000
vnet0:3: flags=8943<UP,BROADCAST,RUNNING,PROMISC,SIMPLEX,MULTICAST> metric 0 mtu 1500
description: associated with jail: plex as nic: epair0b
options=8<VLAN_MTU>
ether 00:25:90:14:fa:09
hwaddr 02:07:10:00:05:0a
nd6 options=1<PERFORMNUD>
media: Ethernet 10Gbase-T (10Gbase-T <full-duplex>)
status: active
groups: epair


Any help or advice on how i can successfully set this up is much appreciated.

Second issue:

I see this error popping up frequently. I am not sure what it means or what it relates to and am looking for clarification:

Let me know if more information, details are needed.

Many thanks, Dominique

 

[Elliot Dierksen]

LACP configuration: every time i try to configure LACP, my connection drops.

Have you configured LACP in the switch? LACP has to be configured on both sides for it to work.

 

[ddaenen1]

Have you configured LACP in the switch? LACP has to be configured on both sides for it to work.

Yes, LACP on the switch is configured.

Just for the record, this worked flawless with my ReadNAS.

 

[Elliot Dierksen]

Yes, LACP on the switch is configured.

The first section looks like the regular port configuration page. Did you select LACP in Switching LAG  Advanced  LACP Port Configuration ? I did look in the manual about that. http://www.downloads.netgear.com/files/GDC/GS716TV2/GS716T_GS724T-SWA-October2012.pdf
I am mostly a Cisco person, but I have done LACP to Netgear in the past. My recollection was that it was a bit touchy. It is also suspicious to me that one of the ports doesn't show as being physically up. In Cisco speak, they should both be physically up even if there is a disagreement on LACP. Once LACP completes, then it would go to line protocol up.

 

[Elliot Dierksen]

Switch is currently still configured to LACP but only one cable plugged in

Is it possible the switch won't bring up the LACP group if both interfaces aren't up? That would be minimum channels/members in Cisco speak.

 

[ddaenen1]

The first section looks like the regular port configuration page. Did you select LACP in Switching LAG  Advanced  LACP Port Configuration ? I did look in the manual about that.

Yes, that would this one:

Well, i just got it to work. Not sure there was a lot of science in my approach but i plugged in the 2nd UTP in the switch and configured the lag in Freenas. Again, the connection went down and didn't come back up. I rebooted the server via IPMI and the connection was still down. Then i rebooted the router. No change, link still down. After that, selected "1 Configure Network Interfaces" from the FreeNAS console menu, and selected DHCP and in a couple of seconds the Lag0: link went up under a new IP address. My UPS link was still down and so was Plex but after a reboot, everything working like a charm. Maybe FreeNAS and my router were having a bit of a fight.

Next stop: figuring out what this mDNSResponder error is all about.

 

[Jessep]

Taking a step back, what are you hoping to accomplish with LACP/LAG? What is the desired outcome?

 

[ddaenen1]

Taking a step back, what are you hoping to accomplish with LACP/LAG? What is the desired outcome?

Simply to share the load when several plex clients are accessing the server at the same time.

 

[Jessep]

1080p should max out about 12Mbps, you can host quite a few of those on a 1Gb wired link.

 

[ddaenen1]

1080p should max out about 12Mbps, you can host quite a few of those on a 1Gb wired link.

True, and even for the 4K 2160p HVEC movies i have, 1Gbps might suffice. The thing is, i want stuff to work and when it doesn't, it bothers me and i have a strong urge to fix it :)

 

[Yorick]

I have LACP configured as well - more “because it’s there”. With two PCs and one TV, it won’t ever make a performance difference. For all I know they’ll all traverse over the same link. And since I have a single switch and a single PSU in the FreeNAS server, and it’s all hooked up to a single power circuit, I really can’t make noise about “redundancy” either. Still! It’s there, and so it got configured. Worked first attempt.

One thing I learned in my years of networking: Always configure active-active LACP. It’s just easier on the brain, easier to keep after. No wondering who initiates - both do.

FreeNAS 11.3, btw, changes the behavior of the lagg interface. It now gets a random-ish MAC on boot. That means my “DHCP with static assignment” shenanigans had to come to an end, and I had to hard-code an IPv4.

 

[ddaenen1]

So i got everything to function the way i wanted and the NAS seems to work flawless. However, i still see mDNSresponder errors in my log.

When FreeNAS boots, i get this:

And while it is running, the below entries show up:

A bit of research showed me that mDNSresponder is part of Apple's Bonjour service, used to discover devices in the LAN by pinging the local DNS server. Could it be that this is related to a router setting rather than FreeNAS? The fact that i have one internal (192.168.1.1) and 2 additional IP's provided by my ISP in the name server settings?

 

[ddaenen1]

Unfortunately i still haven't got the mDNS issue sorted out. What i do know in the mean time is that it is related to my Macbook as i hadn't seen the entry in the logs anymore for a long time and when i created an additional zpool today and wanted access it via my Macbook, the entries reappeared in my log.

Anyone that can help with this one?

 

[ddaenen1]

It seems i cannot find a solution for this issue. I am unsure if it has any impact on performance, but it is simply very annoying. I have googled this and searched the forums but never got any further than that it was reported as a bug but closed as being believed it was fixed. Obviously it is not.

Surely there must be a way to root cause this. Any help on where i need to start is much appreciated.