Re: [AFMUG] Trouble Identifying Throughput Issue

[Christopher Gray]

I'm not sure this is leading to anything, but I tried turning off
aggregation.

R3 -> R4 UBNT link throughput test dropped to about 30 Mbps (less than
half) but my R1 -> R4 test almost doubled to about 20 Mbps. [I'm using
firmware v5.6.9, but I think this issue has been present for a while.] I
experimented with various frame / size settings, but did not see any
further improved results.

I tried enabling flow control on the MikroTik ports used for the UBNT link,
but that did not help (and no pause frames were recorded).


It is probably time to take a break and mess with it tomorrow. Thank you
for the help today as usual.

On Wed, Oct 26, 2016 at 7:15 PM, Christopher Gray  wrote:

> R1 is connected to the Internet via an MRV auto negotiated to 1G FD (This
> is a 1 G Internet connection)
>
> R2 is connected to the Internet through a copper connection to a Juniper
> switch on a 30 Mbps layer 2 fiber transport to a switch where the Internet
> service is connected. This R2 -- Juniper connection is set manually because
> they would not play nice if set to auto.
>
> I have a similar setup on the same fiber system in 3 other locations. None
> of which have this behavior (but also, none of which have UBNT M5 hardware,
> so that could be a potential issue).
>
>
> On Wed, Oct 26, 2016 at 5:31 PM, That One Guy /sarcasm <
> thatoneguyst...@gmail.com> wrote:
>
>> define your transport, are you talking a layer 2 circuit, vpn tunnel,
>> fiber, etc?
>>
>> On Wed, Oct 26, 2016 at 4:24 PM, George Skorup  wrote:
>>
>>> The manual 100FD interface... what is that talking to? The Auto 1G on
>>> R1? If that's the case, I'd bet that's your problem. Keep in mind that you
>>> cannot run auto on one side and fixed FDX on the other side. This results
>>> in a duplex mismatch. The interface in auto will fall back to HDX. If you
>>> did auto one side and HDX on the other side, they'd both be HDX, so it
>>> would work fine. But obviously half duplex sux.
>>>
>>> On 10/26/2016 1:54 PM, Christopher Gray wrote:
>>>
>>> R1 is the only router with 1 Gbps ports. Everything is auto except 1
>>> connection that requires manual settings.
>>>
>>> *R1* -- (Auto 1 G FD) ...Internet... (Manual 100 FD) -- *R2 *-- (Auto
>>> 100 FD)  -- *R3* -- (Auto 100 FD) ...M5... (Auto 100 FD) -- *R4*
>>>
>>> MTU is set to 1500 on every port (and the UBNT link).
>>>
>>> Flow control is off, and none of the interfaces show any pause frames
>>> received.
>>>
>>>
>>> This is a live link, but it is only running ~ 1 Mbps otherwise.
>>>
>>>
>>>
>>> On Wed, Oct 26, 2016 at 2:12 PM, Steve  wrote:
>>>
 Few questions come to mind.

 Are all set to auto negotiate or are they fixed at 100Mbit?
 What are the MTU's of each connection?
 Flow control turned on?


 - Original Message -
 From: "Christopher Gray" 
 To: "af" 
 Sent: Wednesday, October 26, 2016 1:57:56 PM
 Subject: [AFMUG] Trouble Identifying Throughput Issue

 I have a section of my network that is lacking something, and I can't
 figure out where the problem is. I'm looking for any thoughts /
 suggestions.

 4x MikroTik routers

 Link speeds:
 R1 -- (30 Mbps IP / Transport) -- R2 -- (100 Mbps Eth) -- R3 -- (75 Mbps
 UBNT M5) -- R4

 The limiting factor for traffic should be the Transport, and I expect
 to be
 able to get 30 Mbps across the system (one-way).

 Testing from R1 to R3 runs 30 Mbps.

 Testing from R2 to R4 runs 75 Mbps.

 Testing from R1 to R4 only runs 10 Mbps (instead of 30).

 Tests were one-way btest with 20 TCP streams.

 Any ideas for something that would cause this?

>>>
>>>
>>>
>>
>>
>> --
>> If you only see yourself as part of the team but you don't see your team
>> as part of yourself you have already failed as part of the team.
>>
>
>

Re: [AFMUG] Google Fiber is no more

[Rory Conaway]

In other cities, they cherry picked.

Rory

From: Af [mailto:af-boun...@afmug.com] On Behalf Of Sterling Jacobson
Sent: Wednesday, October 26, 2016 7:00 PM
To: af@afmug.com
Subject: Re: [AFMUG] Google Fiber is no more

From the director of one of the Google Fiber builds (in Provo) that is not the 
case.

He said they overspent on contractors MAJORLY.
And that was just to expand the existing network to all homes in that area.

He argued with his bosses about he extravagant added fees on construction but 
they just said to pay them, no questions asked.

I had some of those figures from him at that conversation and some costs were 
over 80x what it should have been.

My best guess is that all the fiber build in certain areas increased the 
contract cost of build into the stratosphere.

And now they are reigning it in and going wireless to attempt to defray the 
costs.

At least with Provo they were not allowed to cherry pick, it was build everyone.
And it seems like they picked up a large portion of the communities, but I 
didn’t get overall take rate.

From: Af [mailto:af-boun...@afmug.com] On Behalf Of Rory Conaway
Sent: Wednesday, October 26, 2016 12:56 AM
To: af@afmug.com
Subject: Re: [AFMUG] Google Fiber is no more

Absolutely they cherry picked.  Then they went into MDU’s for pennies and lost 
their shirts.

Rory

From: Af [mailto:af-boun...@afmug.com] On Behalf Of Josh Reynolds
Sent: Tuesday, October 25, 2016 9:34 PM
To: af@afmug.com
Subject: Re: [AFMUG] Google Fiber is no more


I'd love to see their overall take rates. I have heard numbers of 75-85% in 
more affluent areas. They cherry picked neighborhoods for sure though.

On Oct 25, 2016 10:15 PM, "Rory Conaway" 
> wrote:
Big surprise there.  They built it and no one came.

Rory

From: Af [mailto:af-boun...@afmug.com] On Behalf 
Of Tushar Patel
Sent: Tuesday, October 25, 2016 7:14 PM
To: af@afmug.com
Subject: Re: [AFMUG] Google Fiber is no more

Their contractor are still hiring installer in Austin.

Need to probably understand why those cities not others?

Tushar


On Oct 25, 2016, at 9:06 PM, Josh Reynolds 
> wrote:

New ones. They're still deploying existing networks. They just opened up a few 
new areas in Kansas City recently.

On Oct 25, 2016 9:03 PM, "Jaime Solorza" 
> wrote:

Moving folks to wireless Aye Dios

On Oct 25, 2016 7:56 PM, "Gino Villarini" 
> wrote:
https://gizmodo.com/google-fiber-halts-operations-in-ten-cities-1788214992?rev=1477443092657_campaign=socialflow_gizmodo_facebook_source=gizmodo_facebook_medium=socialflow

[AFMUG] Fwd: [WISPA] IPV6 deploymernt

[Faisal Imtiaz]

An excellent detailed solution (from one of the other forums). 

Faisal Imtiaz 
Snappy Internet & Telecom 
7266 SW 48 Street 
Miami, FL 33155 
Tel: 305 663 5518 x 232 

Help-desk: (305)663-5518 Option 2 or Email: supp...@snappytelecom.net 

> From: "Tim Way" 
> To: "WISPA General List" 
> Sent: Tuesday, October 25, 2016 9:01:51 PM
> Subject: Re: [WISPA] IPV6 deploymernt

> Art,
> So I know of two solid methods that could solve your problem. Neither are 
> super
> awesome and both would involve NAT.

> 1. IPv6 only to the client with NAT64 and DNS64 to handle IPv4 only 
> connectivity
> 2. IPv4 CGN Shared Address Space, RFC 6598 100.64.0.0/10 , and IPv6 Global
> Unicast running in Dual Stack

> Either one would work. I apologize in advance for the long post that follows.

> I've only done the configurations on Cisco routers with the radios just 
> passing
> traffic at layer 2. I'd have to check the feature set of your routers routing
> wise but it shouldn't be hard. It also could be built in a lab with static
> routing largely. I think Mikrotik supports NAT64 but again for a lab
> environment any recent Cisco device could be used with IP Services licensing.

> Your address plan for your global unicast IPv6 space comes into play. This is
> how I would lab it up including moving routing to the tower with the CPE in
> bridge mode:

> Your fictional IPv6 prefix: :::/32

> Your NAT64 Prefix: ::cc00::/96

> Customer DHCPv6-PD Allocation Prefix: ::aa00::/40
> Your fictional customer #1: The Johnson Family, ::aa00:0100::/56
> Your fictional customer #2: The Billings' Family, ::aa00:0200::/56

> Fictional Tower 1
> ISP Mgmt VLAN of CPE: 11, ::bb00:0011::/64
> ISP Customer VLAN of CPE: 12, ::bb00:0012::/64
> ISP Router at the tower on VLAN 11: ::bb00:0011::1/64
> ISP Router at the tower on VLAN 12: ::bb00:0012::1/64

> The Johnson Family Setup:
> ISP CPE VLAN 11 IP: ::bb00:0011::f/64
> Customer's Netgear WAN Interface: ::bb00:0012::f/64
> Customer's Netgear LAN Interface: ::aa00:010a::1/64
> Customer's Netgear Guest WiFi: ::aa00:010b::1/64

> The Billings' Family Setup:
> ISP CPE VLAN 11 IP: ::bb00:0011::e/64
> Customer's Netgear WAN Interface: ::bb00:0012::e/64
> Customer's Netgear LAN Interface: ::aa00:020a::1/64
> Customer's Netgear Guest WiFi: ::aa00:020b::1/64

> 1. You'd bridge VLAN 12 through the CPE to customer's WAN interface as the
> native VLAN and put the IP on VLAN 11.
> 2. If you use static routing and manual address assignment to eliminate
> variables in the lab you'll want to add static routes on the tower router for
> the ::/56 prefixes that would be allocated to each customer. Normally these
> routes will be injected into the routing table at the DHCPv6 router and could
> be distributed from there.
> 3. The last piece of the puzzle will be adding in the NAT64 and DNS64 devices.
> BIND can do DNS64 and you could use a Cisco router to do the NAT64. You'd want
> the "Customer's Netgear" to use the DNS64 server as it's upstream DNS server 
> to
> ensure that it receives  records for sites that only have A records. This
> is the fragile component of the DNS64 and NAT64 deployment because it requires
> the customers computer or router uses your resolver. You will want to ensure
> the router performing NAT64 is advertising the prefix it is using for NAT64
> into your IGP or that your default routed traffic lands on that NAT64 to 
> ensure
> it is routed correctly.

> This should get you a functional IPv6 only customer network that only returns
>  records for all DNS requests. It's a little late so I apologize for any
> mistakes in the addressing. Also I will think about doing this with routing at
> the CPE as well overnight and add that response. I'd be very intrigued to see
> this in a lab environment with the fictional customers all setup to see how
> NAT64 and DNS64 actually works in reality instead of just implementing CGN
> which I see as the less visible or resilient change for the customer. That 
> said
> I see the pure IPv6 deployment with NAT64 and DNS64 as the better long term
> solution if you could reliably ensure your customers use your DHCP server or
> ensure that your tech support says to reset that right away. It also would
> break a customer using OpenDNS to restrict web-sites from their kid's for
> example.

> Thanks,

> Tim

> On Tue, Oct 25, 2016 at 4:42 PM, Art Stephens < asteph...@ptera.com > wrote:

>> Tim,
>> So we are an IPV4 ISP not able to get any more IPV4 address space. We have 
>> IPV6
>> working in office, and on server network.
>> I have working windows and linux IPV6 only configured machines but obviously
>> they can only access IPV6 capable web sites and such.

>> But we will need to start assigning IPV6 WAN address to customer routers and
>> UBNT radios in radio router mode when we get a CRM that 

Re: [AFMUG] New Network from Ground Up - How Would You Address It?

[Sterling Jacobson]

We are doing dual stack with IPv6 and IPv4 available on DHCP for each customer.

I have over 600 IPv4 assigned and about 80 IPv6 assigned currently, so you can 
see how well that's going...

I would love to just use IPv6, but there doesn't seem to be a good solution for 
that currently.

Which is sad because IPv6 has been out there for over a decade.


-Original Message-
From: Af [mailto:af-boun...@afmug.com] On Behalf Of Paul Stewart
Sent: Wednesday, October 26, 2016 4:31 PM
To: af@afmug.com
Subject: Re: [AFMUG] New Network from Ground Up - How Would You Address It?

Dual stack and Ipv4 public addresses…. 

> On Oct 26, 2016, at 4:05 PM, Sam Morris  wrote:
> 
> If you were starting a new network from scratch, how would you do your IP 
> addressing?
> 
> Option 1) ipv6 - Doesn't appear that everything on the Internet supports pure 
> v6, which would require...
> 
> Option 2) ipv6 with NAT64 or dual stack (or whatever would be a patch to make 
> Option 1 work)... or
> 
> Option 3) ipv4 with private IP addresses and a single public v4 address at 
> the edges (and use CGN for the calea stuff - CGN which evidently comes with 
> its own set of problems...)
> 
> Or is there a better option that I'm not thinking about?
> 
> Deciding among these seems like picking which presidential candidate to vote 
> for - They all stink, and trying to decide which one stinks the least...
> 
> Thanks,
> Sam

Re: [AFMUG] Google Fiber is no more

[Sterling Jacobson]

From the director of one of the Google Fiber builds (in Provo) that is not the 
case.

He said they overspent on contractors MAJORLY.
And that was just to expand the existing network to all homes in that area.

He argued with his bosses about he extravagant added fees on construction but 
they just said to pay them, no questions asked.

I had some of those figures from him at that conversation and some costs were 
over 80x what it should have been.

My best guess is that all the fiber build in certain areas increased the 
contract cost of build into the stratosphere.

And now they are reigning it in and going wireless to attempt to defray the 
costs.

At least with Provo they were not allowed to cherry pick, it was build everyone.
And it seems like they picked up a large portion of the communities, but I 
didn’t get overall take rate.

From: Af [mailto:af-boun...@afmug.com] On Behalf Of Rory Conaway
Sent: Wednesday, October 26, 2016 12:56 AM
To: af@afmug.com
Subject: Re: [AFMUG] Google Fiber is no more

Absolutely they cherry picked.  Then they went into MDU’s for pennies and lost 
their shirts.

Rory

From: Af [mailto:af-boun...@afmug.com] On Behalf Of Josh Reynolds
Sent: Tuesday, October 25, 2016 9:34 PM
To: af@afmug.com
Subject: Re: [AFMUG] Google Fiber is no more


I'd love to see their overall take rates. I have heard numbers of 75-85% in 
more affluent areas. They cherry picked neighborhoods for sure though.

On Oct 25, 2016 10:15 PM, "Rory Conaway" 
> wrote:
Big surprise there.  They built it and no one came.

Rory

From: Af [mailto:af-boun...@afmug.com] On Behalf 
Of Tushar Patel
Sent: Tuesday, October 25, 2016 7:14 PM
To: af@afmug.com
Subject: Re: [AFMUG] Google Fiber is no more

Their contractor are still hiring installer in Austin.

Need to probably understand why those cities not others?

Tushar


On Oct 25, 2016, at 9:06 PM, Josh Reynolds 
> wrote:

New ones. They're still deploying existing networks. They just opened up a few 
new areas in Kansas City recently.

On Oct 25, 2016 9:03 PM, "Jaime Solorza" 
> wrote:

Moving folks to wireless Aye Dios

On Oct 25, 2016 7:56 PM, "Gino Villarini" 
> wrote:
https://gizmodo.com/google-fiber-halts-operations-in-ten-cities-1788214992?rev=1477443092657_campaign=socialflow_gizmodo_facebook_source=gizmodo_facebook_medium=socialflow

Re: [AFMUG] Trolling ISP support forums

[Kurt Fankhauser]

I actually laughed when I read that. Or better yet you could pose to be a
tech for a large competing cable company on the other ISP's support forum
and really get them riled up!

On Tue, Oct 25, 2016 at 10:03 PM, Mathew Howard 
wrote:

> I have to admit, that does sound enjoyable...
>
> On Oct 25, 2016 8:38 PM, "That One Guy /sarcasm" <
> thatoneguyst...@gmail.com> wrote:
>
>> I found a new pass time. Create an account at on of the major ISPs that
>> uses forum based support and you get to tell asshole customers off without
>> losing customers yourself, its a great release
>>
>> --
>> If you only see yourself as part of the team but you don't see your team
>> as part of yourself you have already failed as part of the team.
>>
>

Re: [AFMUG] UBNT S16 Application and AP grounding question

[chuck]

1)No, I would not put surge protectors between them.

2)  Peeling off the drain wire routes any ground current on the shield directly 
to the tower.  This may keep ground currents out of the attached device.  But 
so does a surge suppressor that has shielded jacks.  

From: Paul McCall 
Sent: Wednesday, October 26, 2016 6:02 PM
To: af@afmug.com 
Subject: Re: [AFMUG] UBNT S16 Application and AP grounding question

Thanks Chuck.  In the scenario I describe, sub 20ft. Cat5 runs from the S16 to 
the APs, you would still put Ethernet protectors inline between them?  

 

What benefit is there in the suggestion to NOT connect the shield at the S16 
end, but rather “peel it off” and attach directly to the tower ground?

 

From: Af [mailto:af-boun...@afmug.com] On Behalf Of ch...@wbmfg.com
Sent: Wednesday, October 26, 2016 6:16 PM
To: af@afmug.com
Subject: Re: [AFMUG] UBNT S16 Application and AP grounding question

 

Lots of different reasons for grounding and shielding.  

Electrostatic shielding, magnetic shielding, faraday shielding etc.

Lightning routing, ground loop prevention etc etc.

 

Some of these things are mutually exclusive.  

 

I have many customers that use my surge suppressors at the top and bottom of 
the tower, with shielded cable, to good effect.  My products have shielded 
jacks so they automatically interface with the cable shield and drain wire if 
you are using the shielded plugs.  

 

Some radios expected to be grounded, some expect to float.  

And when lightning comes, sometimes it comes from the power lines, sometimes 
from the ground, sometimes it is just induced currents from nearby strikes.  
Then of course there is the direct hit.  Nothing survives the direct strike.

 

With radios, antennas, power lines, network cables and tower steel all involved 
in different configurations on each tower, hard to do a true R56 common point 
grounding design.  Much easier if it is just coax and antennas on the tower.

 

Personally I would ground top and bottom.  And I would use my surge 
suppressors... for both DC and data.  

 

From: Paul McCall 

Sent: Wednesday, October 26, 2016 3:35 PM

To: af@afmug.com 

Subject: [AFMUG] UBNT S16 Application and AP grounding question

 

Guys, 

 

Looking for opinions.  I am brave at least.   I have reached out to Cambium for 
opinions and are working with their engineers and just now sent an email to 
UBNT as well.   

 

We are using the S16 with both UBNT and Cambium devices (various APs / BHs) 
always with the S16 placed at the top of the tower with Fiber / DC up the 
tower, with a combination of Ditek surge protectors (at the bottom) and 
Polyphaser surge proectors (at the top) of the DC run.

 

We are grounding the S16 to the tower, and whenever the radios have ground lug, 
are grounding those to a common “ground bus bar” at the top of the tower as 
well.  Almost every Cat 5e cable that connects from the S16 to the AP/BHs is 
less than 20feet.  

 

My questions are in the area of the Cat 5e cable and grounding, along with 
surge protection.

 

We are using shielded Cat 5e cables to each radio and shielded connectors at 
both ends.  However, a grounding consultant says to not connect the shield at 
the end near the S16, rather, peel enough back to ground that to the tower.  
Does that make sense?   Also, if using the drain wire on cable such as 
ToughCable Carrier, does that make sense?

 

Second question,

 

If we have two different levels of APs / BHs (for vertical separation 
front/back ratio reasons) , and say each one is 15 feet above/below the S16, 
should the ground of the radios in relation to this be as close to each level 
(separate bus bar) as possible or all to one point?  Pictures attached.  
Concerned with the concept of single point grounding vs the concept as 
grounding as close to the equipment as possible.

Lastly,

 

In my scenario of the S16 with relatively short cable runs, does inline 
Ethernet surge protection make sense?  Does it add value to the protection or 
not really?  Its not a $$ consideration vs two more points of failure 
(protector itself and the extra Cat 5Cable in the middle.

 

With all the $$ that PDMNet has put into not just equipment and fiber 
retrofits, and extensive sub 5 ohm grounding systems on all towers and 
electrical panels, additional electrical panel protectors, circuit protectors, 
etc., I want to close the loop on this last set of details up top.

 

Thanks!

 

Paul McCall, President

PDMNet, Inc. / Florida Broadband, Inc.

658 Old Dixie Highway

Vero Beach, FL 32962

772-564-6800  

pa...@pdmnet.net

www.pdmnet.com

www.floridabroadband.com

 

 

Re: [AFMUG] UBNT S16 Application and AP grounding question

[Paul McCall]

Thanks Chuck.  In the scenario I describe, sub 20ft. Cat5 runs from the S16 to 
the APs, you would still put Ethernet protectors inline between them?

What benefit is there in the suggestion to NOT connect the shield at the S16 
end, but rather “peel it off” and attach directly to the tower ground?

From: Af [mailto:af-boun...@afmug.com] On Behalf Of ch...@wbmfg.com
Sent: Wednesday, October 26, 2016 6:16 PM
To: af@afmug.com
Subject: Re: [AFMUG] UBNT S16 Application and AP grounding question

Lots of different reasons for grounding and shielding.
Electrostatic shielding, magnetic shielding, faraday shielding etc.
Lightning routing, ground loop prevention etc etc.

Some of these things are mutually exclusive.

I have many customers that use my surge suppressors at the top and bottom of 
the tower, with shielded cable, to good effect.  My products have shielded 
jacks so they automatically interface with the cable shield and drain wire if 
you are using the shielded plugs.

Some radios expected to be grounded, some expect to float.
And when lightning comes, sometimes it comes from the power lines, sometimes 
from the ground, sometimes it is just induced currents from nearby strikes.  
Then of course there is the direct hit.  Nothing survives the direct strike.

With radios, antennas, power lines, network cables and tower steel all involved 
in different configurations on each tower, hard to do a true R56 common point 
grounding design.  Much easier if it is just coax and antennas on the tower.

Personally I would ground top and bottom.  And I would use my surge 
suppressors... for both DC and data.

From: Paul McCall
Sent: Wednesday, October 26, 2016 3:35 PM
To: af@afmug.com
Subject: [AFMUG] UBNT S16 Application and AP grounding question

Guys,

Looking for opinions.  I am brave at least.   I have reached out to Cambium for 
opinions and are working with their engineers and just now sent an email to 
UBNT as well.

We are using the S16 with both UBNT and Cambium devices (various APs / BHs) 
always with the S16 placed at the top of the tower with Fiber / DC up the 
tower, with a combination of Ditek surge protectors (at the bottom) and 
Polyphaser surge proectors (at the top) of the DC run.

We are grounding the S16 to the tower, and whenever the radios have ground lug, 
are grounding those to a common “ground bus bar” at the top of the tower as 
well.  Almost every Cat 5e cable that connects from the S16 to the AP/BHs is 
less than 20feet.

My questions are in the area of the Cat 5e cable and grounding, along with 
surge protection.

We are using shielded Cat 5e cables to each radio and shielded connectors at 
both ends.  However, a grounding consultant says to not connect the shield at 
the end near the S16, rather, peel enough back to ground that to the tower.  
Does that make sense?   Also, if using the drain wire on cable such as 
ToughCable Carrier, does that make sense?

Second question,

If we have two different levels of APs / BHs (for vertical separation 
front/back ratio reasons) , and say each one is 15 feet above/below the S16, 
should the ground of the radios in relation to this be as close to each level 
(separate bus bar) as possible or all to one point?  Pictures attached.  
Concerned with the concept of single point grounding vs the concept as 
grounding as close to the equipment as possible.
Lastly,

In my scenario of the S16 with relatively short cable runs, does inline 
Ethernet surge protection make sense?  Does it add value to the protection or 
not really?  Its not a $$ consideration vs two more points of failure 
(protector itself and the extra Cat 5Cable in the middle.

With all the $$ that PDMNet has put into not just equipment and fiber 
retrofits, and extensive sub 5 ohm grounding systems on all towers and 
electrical panels, additional electrical panel protectors, circuit protectors, 
etc., I want to close the loop on this last set of details up top.

Thanks!

Paul McCall, President
PDMNet, Inc. / Florida Broadband, Inc.
658 Old Dixie Highway
Vero Beach, FL 32962
772-564-6800
pa...@pdmnet.net
www.pdmnet.com
www.floridabroadband.com

Re: [AFMUG] New Network from Ground Up - How Would You Address It?

[Paul Stewart]

Dual stack and Ipv4 public addresses…. 

> On Oct 26, 2016, at 4:05 PM, Sam Morris  wrote:
> 
> If you were starting a new network from scratch, how would you do your IP 
> addressing?
> 
> Option 1) ipv6 - Doesn't appear that everything on the Internet supports pure 
> v6, which would require...
> 
> Option 2) ipv6 with NAT64 or dual stack (or whatever would be a patch to make 
> Option 1 work)... or
> 
> Option 3) ipv4 with private IP addresses and a single public v4 address at 
> the edges (and use CGN for the calea stuff - CGN which evidently comes with 
> its own set of problems...)
> 
> Or is there a better option that I'm not thinking about?
> 
> Deciding among these seems like picking which presidential candidate to vote 
> for - They all stink, and trying to decide which one stinks the least...
> 
> Thanks,
> Sam

Re: [AFMUG] UBNT S16 Application and AP grounding question

[chuck]

Lots of different reasons for grounding and shielding.  
Electrostatic shielding, magnetic shielding, faraday shielding etc.
Lightning routing, ground loop prevention etc etc.

Some of these things are mutually exclusive.  

I have many customers that use my surge suppressors at the top and bottom of 
the tower, with shielded cable, to good effect.  My products have shielded 
jacks so they automatically interface with the cable shield and drain wire if 
you are using the shielded plugs.  

Some radios expected to be grounded, some expect to float.  
And when lightning comes, sometimes it comes from the power lines, sometimes 
from the ground, sometimes it is just induced currents from nearby strikes.  
Then of course there is the direct hit.  Nothing survives the direct strike.

With radios, antennas, power lines, network cables and tower steel all involved 
in different configurations on each tower, hard to do a true R56 common point 
grounding design.  Much easier if it is just coax and antennas on the tower.

Personally I would ground top and bottom.  And I would use my surge 
suppressors... for both DC and data.  

From: Paul McCall 
Sent: Wednesday, October 26, 2016 3:35 PM
To: af@afmug.com 
Subject: [AFMUG] UBNT S16 Application and AP grounding question

Guys, 

 

Looking for opinions.  I am brave at least.   I have reached out to Cambium for 
opinions and are working with their engineers and just now sent an email to 
UBNT as well.   

 

We are using the S16 with both UBNT and Cambium devices (various APs / BHs) 
always with the S16 placed at the top of the tower with Fiber / DC up the 
tower, with a combination of Ditek surge protectors (at the bottom) and 
Polyphaser surge proectors (at the top) of the DC run.

 

We are grounding the S16 to the tower, and whenever the radios have ground lug, 
are grounding those to a common “ground bus bar” at the top of the tower as 
well.  Almost every Cat 5e cable that connects from the S16 to the AP/BHs is 
less than 20feet.  

 

My questions are in the area of the Cat 5e cable and grounding, along with 
surge protection.

 

We are using shielded Cat 5e cables to each radio and shielded connectors at 
both ends.  However, a grounding consultant says to not connect the shield at 
the end near the S16, rather, peel enough back to ground that to the tower.  
Does that make sense?   Also, if using the drain wire on cable such as 
ToughCable Carrier, does that make sense?

 

Second question,

 

If we have two different levels of APs / BHs (for vertical separation 
front/back ratio reasons) , and say each one is 15 feet above/below the S16, 
should the ground of the radios in relation to this be as close to each level 
(separate bus bar) as possible or all to one point?  Pictures attached.  
Concerned with the concept of single point grounding vs the concept as 
grounding as close to the equipment as possible.

Lastly,

 

In my scenario of the S16 with relatively short cable runs, does inline 
Ethernet surge protection make sense?  Does it add value to the protection or 
not really?  Its not a $$ consideration vs two more points of failure 
(protector itself and the extra Cat 5Cable in the middle.

 

With all the $$ that PDMNet has put into not just equipment and fiber 
retrofits, and extensive sub 5 ohm grounding systems on all towers and 
electrical panels, additional electrical panel protectors, circuit protectors, 
etc., I want to close the loop on this last set of details up top.

 

Thanks!

 

Paul McCall, President

PDMNet, Inc. / Florida Broadband, Inc.

658 Old Dixie Highway

Vero Beach, FL 32962

772-564-6800  

pa...@pdmnet.net

www.pdmnet.com

www.floridabroadband.com

 

 

Re: [AFMUG] New Network from Ground Up - How Would You Address It?

[Mathew Howard]

Assuming you can't get enough ipv4 addresses to do option 4 for whatever
reason (which is the best option, really), I would probably say the best
way to go is CGN plus IPv6 dual stack.

On Wed, Oct 26, 2016 at 3:12 PM, Eric Kuhnke  wrote:

> None of the above, unless a ridiculous number of ipv4 addresses are
> required, I'd find the money to acquire a sufficient number of ipv4 /24s
> (preferably in pieces like /22) through the official ARIN transfer process
> and consider the per-customer IP cost paid as an NRC for the transfer as
> part of the network build cost.
>
> Option 4) one ipv4 per customer and ipv6 dual stack
>
>
>
> On Wed, Oct 26, 2016 at 1:05 PM, Sam Morris  wrote:
>
>> If you were starting a new network from scratch, how would you do your IP
>> addressing?
>>
>> Option 1) ipv6 - Doesn't appear that everything on the Internet supports
>> pure v6, which would require...
>>
>> Option 2) ipv6 with NAT64 or dual stack (or whatever would be a patch to
>> make Option 1 work)... or
>>
>> Option 3) ipv4 with private IP addresses and a single public v4 address
>> at the edges (and use CGN for the calea stuff - CGN which evidently comes
>> with its own set of problems...)
>>
>> Or is there a better option that I'm not thinking about?
>>
>> Deciding among these seems like picking which presidential candidate to
>> vote for - They all stink, and trying to decide which one stinks the
>> least...
>>
>> Thanks,
>> Sam
>>
>
>

Re: [AFMUG] Mikrotik DNS Cache

[Josh Baird]

Same here.  The MT DNS cache/proxy is garbage.

On Wed, Oct 26, 2016 at 1:42 PM, Jesse DuPont  wrote:

> I, too, seem to have weird issues where the Mikrotik will slowly stop
> doing recursive lookups until the cache is flushed. Not sure if it's
> related to using IPv6 DNS addresses or what. I've moved back to giving out
> my on-net DNS servers instead of using the Mirkotik cache/proxy.
>
> *Jesse DuPont*
>
> Network Architect
> email: jesse.dup...@celeritycorp.net
> Celerity Networks LLC
>
> Celerity Broadband LLC
> Like us! facebook.com/celeritynetworksllc
>
> Like us! facebook.com/celeritybroadband
> On 10/26/16 7:55 AM, Josh Reynolds wrote:
>
> Shouldn't be any issues, no.
>
> On Oct 26, 2016 8:54 AM, "Matt"  wrote:
>
>> Is anyone using the Mikrotik DNS cache as there primary DNS resolver
>> for there clients?  Say use a CCR and your largest upstreams DNS
>> server as parent.  Should there be any issues with that?
>>
>
>

Re: [AFMUG] Trouble Identifying Throughput Issue

[That One Guy /sarcasm]

define your transport, are you talking a layer 2 circuit, vpn tunnel,
fiber, etc?

On Wed, Oct 26, 2016 at 4:24 PM, George Skorup  wrote:

> The manual 100FD interface... what is that talking to? The Auto 1G on R1?
> If that's the case, I'd bet that's your problem. Keep in mind that you
> cannot run auto on one side and fixed FDX on the other side. This results
> in a duplex mismatch. The interface in auto will fall back to HDX. If you
> did auto one side and HDX on the other side, they'd both be HDX, so it
> would work fine. But obviously half duplex sux.
>
> On 10/26/2016 1:54 PM, Christopher Gray wrote:
>
> R1 is the only router with 1 Gbps ports. Everything is auto except 1
> connection that requires manual settings.
>
> *R1* -- (Auto 1 G FD) ...Internet... (Manual 100 FD) -- *R2 *-- (Auto 100
> FD)  -- *R3* -- (Auto 100 FD) ...M5... (Auto 100 FD) -- *R4*
>
> MTU is set to 1500 on every port (and the UBNT link).
>
> Flow control is off, and none of the interfaces show any pause frames
> received.
>
>
> This is a live link, but it is only running ~ 1 Mbps otherwise.
>
>
>
> On Wed, Oct 26, 2016 at 2:12 PM, Steve  wrote:
>
>> Few questions come to mind.
>>
>> Are all set to auto negotiate or are they fixed at 100Mbit?
>> What are the MTU's of each connection?
>> Flow control turned on?
>>
>>
>> - Original Message -
>> From: "Christopher Gray" 
>> To: "af" 
>> Sent: Wednesday, October 26, 2016 1:57:56 PM
>> Subject: [AFMUG] Trouble Identifying Throughput Issue
>>
>> I have a section of my network that is lacking something, and I can't
>> figure out where the problem is. I'm looking for any thoughts /
>> suggestions.
>>
>> 4x MikroTik routers
>>
>> Link speeds:
>> R1 -- (30 Mbps IP / Transport) -- R2 -- (100 Mbps Eth) -- R3 -- (75 Mbps
>> UBNT M5) -- R4
>>
>> The limiting factor for traffic should be the Transport, and I expect to
>> be
>> able to get 30 Mbps across the system (one-way).
>>
>> Testing from R1 to R3 runs 30 Mbps.
>>
>> Testing from R2 to R4 runs 75 Mbps.
>>
>> Testing from R1 to R4 only runs 10 Mbps (instead of 30).
>>
>> Tests were one-way btest with 20 TCP streams.
>>
>> Any ideas for something that would cause this?
>>
>
>
>


-- 
If you only see yourself as part of the team but you don't see your team as
part of yourself you have already failed as part of the team.

Re: [AFMUG] Trouble Identifying Throughput Issue

[George Skorup]

The manual 100FD interface... what is that talking to? The Auto 1G on 
R1? If that's the case, I'd bet that's your problem. Keep in mind that 
you cannot run auto on one side and fixed FDX on the other side. This 
results in a duplex mismatch. The interface in auto will fall back to 
HDX. If you did auto one side and HDX on the other side, they'd both be 
HDX, so it would work fine. But obviously half duplex sux.


On 10/26/2016 1:54 PM, Christopher Gray wrote:
R1 is the only router with 1 Gbps ports. Everything is auto except 1 
connection that requires manual settings.


*R1* -- (Auto 1 G FD) ...Internet... (Manual 100 FD) -- *R2 *-- (Auto 
100 FD)  -- *R3* -- (Auto 100 FD) ...M5... (Auto 100 FD) -- *R4*


MTU is set to 1500 on every port (and the UBNT link).

Flow control is off, and none of the interfaces show any pause frames 
received.



This is a live link, but it is only running ~ 1 Mbps otherwise.



On Wed, Oct 26, 2016 at 2:12 PM, Steve > wrote:


Few questions come to mind.

Are all set to auto negotiate or are they fixed at 100Mbit?
What are the MTU's of each connection?
Flow control turned on?


- Original Message -
From: "Christopher Gray" >
To: "af" >
Sent: Wednesday, October 26, 2016 1:57:56 PM
Subject: [AFMUG] Trouble Identifying Throughput Issue

I have a section of my network that is lacking something, and I can't
figure out where the problem is. I'm looking for any thoughts /
suggestions.

4x MikroTik routers

Link speeds:
R1 -- (30 Mbps IP / Transport) -- R2 -- (100 Mbps Eth) -- R3 --
(75 Mbps
UBNT M5) -- R4

The limiting factor for traffic should be the Transport, and I
expect to be
able to get 30 Mbps across the system (one-way).

Testing from R1 to R3 runs 30 Mbps.

Testing from R2 to R4 runs 75 Mbps.

Testing from R1 to R4 only runs 10 Mbps (instead of 30).

Tests were one-way btest with 20 TCP streams.

Any ideas for something that would cause this?

Re: [AFMUG] Take one down, put one up.

[chuck]

Not bad.  

From: Ben Royer 
Sent: Wednesday, October 26, 2016 10:18 AM
To: af@afmug.com 
Subject: Re: [AFMUG] Take one down, put one up.

Hey Guys,

Sorry for the delay, took me a minute to gather the totals.  It’s going to 
throw you guys off probably as I’ve got some great deals on some stuff, 
especially crane, but here goes.

Rohn SSV – 160’ 9 through 2 – $12,500 approx.
Excavation, backfill, jet, and rock – $1,325
Concrete – $1,365
Crane – $420 (Lower old, lift new)
_
Total – $ 15,610, or $16,000 approx.

Thank you,
Ben Royer, Operations Manager
Royell Communications, Inc.
217-965-3699 www.royell.net

From: Kurt Fankhauser 
Sent: Monday, October 24, 2016 4:22 PM
To: af@afmug.com 
Subject: Re: [AFMUG] Take one down, put one up.

What was the cost of this 160' SSV tower just the tower alone? I put up an 
American Tower Co 160' that was about $20k for tower, plus 8k concrete, plus 
2.5k for crane and we picked it up in 1 piece.

On Mon, Oct 24, 2016 at 3:00 PM, Ben Royer  wrote:

  Thanks, fun getting to put this stuff together.

  Thank you,
  Ben Royer, Operations Manager
  Royell Communications, Inc.
  217-965-3699 www.royell.net

  From: Jaime Solorza 
  Sent: Monday, October 24, 2016 11:27 AM
  To: Animal Farm 
  Subject: Re: [AFMUG] Take one down, put one up.

  Awesome video


  On Oct 24, 2016 9:59 AM, "Ben Royer"  wrote:

Replaced a 90’ SSV with a 160’ SSV, and upgraded the site from PMP100 
900Mhz. to 450i 5Ghz. in the process.  Also threw some 820s backhauls in there 
too.  The take down of old tower and installation of new tower only took a 
couple of hours once the crane got there, as we had to move the old 900mhz. 
from one tower to the next.

http://www.benroyer.net/thayer.mp4

Thank you,
Ben Royer, Operations Manager
Royell Communications, Inc.
217-965-3699 www.royell.net

[AFMUG] Mikrotik CRS locking up

[TJ Trout]

I have a mikrotik CRS which is uplinked with 10G fiber, and then the copper
1G ports go through a netonix acting as a midspan, occasionally the CRS is
locking up where I can still get into the web interface through the uplink,
but it's not passing passing any data over the copper ports...

Trying to figure out if I need to swap it out or update the firmware?
enable / disable STP, FC, etc?

Re: [AFMUG] Trouble Identifying Throughput Issue

[Christopher Gray]

Neither was... but I rebooted them anyhow.


On Wed, Oct 26, 2016 at 3:29 PM, That One Guy /sarcasm <
thatoneguyst...@gmail.com> wrote:

> log into the ubnt gear and verify neither are sitting at the unexpected
> reboot screen
>
> On Wed, Oct 26, 2016 at 1:54 PM, Christopher Gray <
> cg...@graytechsoftware.com> wrote:
>
>> R1 is the only router with 1 Gbps ports. Everything is auto except 1
>> connection that requires manual settings.
>>
>> *R1* -- (Auto 1 G FD) ...Internet... (Manual 100 FD) -- *R2 *-- (Auto
>> 100 FD)  -- *R3* -- (Auto 100 FD) ...M5... (Auto 100 FD) -- *R4*
>>
>> MTU is set to 1500 on every port (and the UBNT link).
>>
>> Flow control is off, and none of the interfaces show any pause frames
>> received.
>>
>>
>> This is a live link, but it is only running ~ 1 Mbps otherwise.
>>
>>
>>
>> On Wed, Oct 26, 2016 at 2:12 PM, Steve  wrote:
>>
>>> Few questions come to mind.
>>>
>>> Are all set to auto negotiate or are they fixed at 100Mbit?
>>> What are the MTU's of each connection?
>>> Flow control turned on?
>>>
>>>
>>> - Original Message -
>>> From: "Christopher Gray" 
>>> To: "af" 
>>> Sent: Wednesday, October 26, 2016 1:57:56 PM
>>> Subject: [AFMUG] Trouble Identifying Throughput Issue
>>>
>>> I have a section of my network that is lacking something, and I can't
>>> figure out where the problem is. I'm looking for any thoughts /
>>> suggestions.
>>>
>>> 4x MikroTik routers
>>>
>>> Link speeds:
>>> R1 -- (30 Mbps IP / Transport) -- R2 -- (100 Mbps Eth) -- R3 -- (75 Mbps
>>> UBNT M5) -- R4
>>>
>>> The limiting factor for traffic should be the Transport, and I expect to
>>> be
>>> able to get 30 Mbps across the system (one-way).
>>>
>>> Testing from R1 to R3 runs 30 Mbps.
>>>
>>> Testing from R2 to R4 runs 75 Mbps.
>>>
>>> Testing from R1 to R4 only runs 10 Mbps (instead of 30).
>>>
>>> Tests were one-way btest with 20 TCP streams.
>>>
>>> Any ideas for something that would cause this?
>>>
>>
>>
>
>
> --
> If you only see yourself as part of the team but you don't see your team
> as part of yourself you have already failed as part of the team.
>

[AFMUG] New Network from Ground Up - How Would You Address It?

[Sam Morris]

If you were starting a new network from scratch, how would you do your 
IP addressing?


Option 1) ipv6 - Doesn't appear that everything on the Internet supports 
pure v6, which would require...


Option 2) ipv6 with NAT64 or dual stack (or whatever would be a patch to 
make Option 1 work)... or


Option 3) ipv4 with private IP addresses and a single public v4 address 
at the edges (and use CGN for the calea stuff - CGN which evidently 
comes with its own set of problems...)


Or is there a better option that I'm not thinking about?

Deciding among these seems like picking which presidential candidate to 
vote for - They all stink, and trying to decide which one stinks the 
least...


Thanks,
Sam

Re: [AFMUG] Trouble Identifying Throughput Issue

[Sam Morris]

You don't have any ToughSwitches in there anyplace do you?

On 10/26/2016 2:30 PM, Christopher Gray wrote:

While running from R1 to R4 (maxing at only 10 Mbps), I can also run
from R1 to R3 an additional 15+.

Also, while running from R1 to R4 (again, maxing at 10 Mbps), I can also
run from R3 to R4 an additional 60+.


Aggregation on the M5 link is defaulted to 32 frames / 50,000 bytes.
Could this be causing such issues on a good link?






On Wed, Oct 26, 2016 at 2:54 PM, Christopher Gray
> wrote:

R1 is the only router with 1 Gbps ports. Everything is auto except 1
connection that requires manual settings.

*R1* -- (Auto 1 G FD) ...Internet... (Manual 100 FD) -- *R2 *--
(Auto 100 FD)  -- *R3* -- (Auto 100 FD) ...M5... (Auto 100 FD) -- *R4*

MTU is set to 1500 on every port (and the UBNT link).

Flow control is off, and none of the interfaces show any pause
frames received.


This is a live link, but it is only running ~ 1 Mbps otherwise.



On Wed, Oct 26, 2016 at 2:12 PM, Steve > wrote:

Few questions come to mind.

Are all set to auto negotiate or are they fixed at 100Mbit?
What are the MTU's of each connection?
Flow control turned on?


- Original Message -
From: "Christopher Gray" >
To: "af" >
Sent: Wednesday, October 26, 2016 1:57:56 PM
Subject: [AFMUG] Trouble Identifying Throughput Issue

I have a section of my network that is lacking something, and I
can't
figure out where the problem is. I'm looking for any thoughts /
suggestions.

4x MikroTik routers

Link speeds:
R1 -- (30 Mbps IP / Transport) -- R2 -- (100 Mbps Eth) -- R3 --
(75 Mbps
UBNT M5) -- R4

The limiting factor for traffic should be the Transport, and I
expect to be
able to get 30 Mbps across the system (one-way).

Testing from R1 to R3 runs 30 Mbps.

Testing from R2 to R4 runs 75 Mbps.

Testing from R1 to R4 only runs 10 Mbps (instead of 30).

Tests were one-way btest with 20 TCP streams.

Any ideas for something that would cause this?

Re: [AFMUG] Trouble Identifying Throughput Issue

[Christopher Gray]

While running from R1 to R4 (maxing at only 10 Mbps), I can also run from
R1 to R3 an additional 15+.

Also, while running from R1 to R4 (again, maxing at 10 Mbps), I can also
run from R3 to R4 an additional 60+.


Aggregation on the M5 link is defaulted to 32 frames / 50,000 bytes. Could
this be causing such issues on a good link?






On Wed, Oct 26, 2016 at 2:54 PM, Christopher Gray <
cg...@graytechsoftware.com> wrote:

> R1 is the only router with 1 Gbps ports. Everything is auto except 1
> connection that requires manual settings.
>
> *R1* -- (Auto 1 G FD) ...Internet... (Manual 100 FD) -- *R2 *-- (Auto 100
> FD)  -- *R3* -- (Auto 100 FD) ...M5... (Auto 100 FD) -- *R4*
>
> MTU is set to 1500 on every port (and the UBNT link).
>
> Flow control is off, and none of the interfaces show any pause frames
> received.
>
>
> This is a live link, but it is only running ~ 1 Mbps otherwise.
>
>
>
> On Wed, Oct 26, 2016 at 2:12 PM, Steve  wrote:
>
>> Few questions come to mind.
>>
>> Are all set to auto negotiate or are they fixed at 100Mbit?
>> What are the MTU's of each connection?
>> Flow control turned on?
>>
>>
>> - Original Message -
>> From: "Christopher Gray" 
>> To: "af" 
>> Sent: Wednesday, October 26, 2016 1:57:56 PM
>> Subject: [AFMUG] Trouble Identifying Throughput Issue
>>
>> I have a section of my network that is lacking something, and I can't
>> figure out where the problem is. I'm looking for any thoughts /
>> suggestions.
>>
>> 4x MikroTik routers
>>
>> Link speeds:
>> R1 -- (30 Mbps IP / Transport) -- R2 -- (100 Mbps Eth) -- R3 -- (75 Mbps
>> UBNT M5) -- R4
>>
>> The limiting factor for traffic should be the Transport, and I expect to
>> be
>> able to get 30 Mbps across the system (one-way).
>>
>> Testing from R1 to R3 runs 30 Mbps.
>>
>> Testing from R2 to R4 runs 75 Mbps.
>>
>> Testing from R1 to R4 only runs 10 Mbps (instead of 30).
>>
>> Tests were one-way btest with 20 TCP streams.
>>
>> Any ideas for something that would cause this?
>>
>
>

Re: [AFMUG] Trouble Identifying Throughput Issue

[That One Guy /sarcasm]

log into the ubnt gear and verify neither are sitting at the unexpected
reboot screen

On Wed, Oct 26, 2016 at 1:54 PM, Christopher Gray <
cg...@graytechsoftware.com> wrote:

> R1 is the only router with 1 Gbps ports. Everything is auto except 1
> connection that requires manual settings.
>
> *R1* -- (Auto 1 G FD) ...Internet... (Manual 100 FD) -- *R2 *-- (Auto 100
> FD)  -- *R3* -- (Auto 100 FD) ...M5... (Auto 100 FD) -- *R4*
>
> MTU is set to 1500 on every port (and the UBNT link).
>
> Flow control is off, and none of the interfaces show any pause frames
> received.
>
>
> This is a live link, but it is only running ~ 1 Mbps otherwise.
>
>
>
> On Wed, Oct 26, 2016 at 2:12 PM, Steve  wrote:
>
>> Few questions come to mind.
>>
>> Are all set to auto negotiate or are they fixed at 100Mbit?
>> What are the MTU's of each connection?
>> Flow control turned on?
>>
>>
>> - Original Message -
>> From: "Christopher Gray" 
>> To: "af" 
>> Sent: Wednesday, October 26, 2016 1:57:56 PM
>> Subject: [AFMUG] Trouble Identifying Throughput Issue
>>
>> I have a section of my network that is lacking something, and I can't
>> figure out where the problem is. I'm looking for any thoughts /
>> suggestions.
>>
>> 4x MikroTik routers
>>
>> Link speeds:
>> R1 -- (30 Mbps IP / Transport) -- R2 -- (100 Mbps Eth) -- R3 -- (75 Mbps
>> UBNT M5) -- R4
>>
>> The limiting factor for traffic should be the Transport, and I expect to
>> be
>> able to get 30 Mbps across the system (one-way).
>>
>> Testing from R1 to R3 runs 30 Mbps.
>>
>> Testing from R2 to R4 runs 75 Mbps.
>>
>> Testing from R1 to R4 only runs 10 Mbps (instead of 30).
>>
>> Tests were one-way btest with 20 TCP streams.
>>
>> Any ideas for something that would cause this?
>>
>
>


-- 
If you only see yourself as part of the team but you don't see your team as
part of yourself you have already failed as part of the team.

Re: [AFMUG] Trouble Identifying Throughput Issue

[Christopher Gray]

R1 is the only router with 1 Gbps ports. Everything is auto except 1
connection that requires manual settings.

*R1* -- (Auto 1 G FD) ...Internet... (Manual 100 FD) -- *R2 *-- (Auto 100
FD)  -- *R3* -- (Auto 100 FD) ...M5... (Auto 100 FD) -- *R4*

MTU is set to 1500 on every port (and the UBNT link).

Flow control is off, and none of the interfaces show any pause frames
received.


This is a live link, but it is only running ~ 1 Mbps otherwise.



On Wed, Oct 26, 2016 at 2:12 PM, Steve  wrote:

> Few questions come to mind.
>
> Are all set to auto negotiate or are they fixed at 100Mbit?
> What are the MTU's of each connection?
> Flow control turned on?
>
>
> - Original Message -
> From: "Christopher Gray" 
> To: "af" 
> Sent: Wednesday, October 26, 2016 1:57:56 PM
> Subject: [AFMUG] Trouble Identifying Throughput Issue
>
> I have a section of my network that is lacking something, and I can't
> figure out where the problem is. I'm looking for any thoughts /
> suggestions.
>
> 4x MikroTik routers
>
> Link speeds:
> R1 -- (30 Mbps IP / Transport) -- R2 -- (100 Mbps Eth) -- R3 -- (75 Mbps
> UBNT M5) -- R4
>
> The limiting factor for traffic should be the Transport, and I expect to be
> able to get 30 Mbps across the system (one-way).
>
> Testing from R1 to R3 runs 30 Mbps.
>
> Testing from R2 to R4 runs 75 Mbps.
>
> Testing from R1 to R4 only runs 10 Mbps (instead of 30).
>
> Tests were one-way btest with 20 TCP streams.
>
> Any ideas for something that would cause this?
>

[AFMUG] Google Fiber is stopping new deployments and why its Great News!

[Gino Villarini]

My BLOG here:
http://aeronetprceo.blogspot.com/2016/10/google-fiber-is-stopping-why-its-great.html

Comments welcomed!



Gino Villarini


President
Metro Office Park #18 Suite 304 Guaynabo, Puerto Rico 00968

[cid:aeronet-logo_310cfc3e-6691-4f69-bd49-b37b834b9238.png]

Re: [AFMUG] Trouble Identifying Throughput Issue

[That One Guy /sarcasm]

is this an idle end to end circuit your testing?, monitor all 6 interfaces
while testing to make sure youre not fighting yourself

On Wed, Oct 26, 2016 at 1:12 PM, Steve  wrote:

> Few questions come to mind.
>
> Are all set to auto negotiate or are they fixed at 100Mbit?
> What are the MTU's of each connection?
> Flow control turned on?
>
>
> - Original Message -
> From: "Christopher Gray" 
> To: "af" 
> Sent: Wednesday, October 26, 2016 1:57:56 PM
> Subject: [AFMUG] Trouble Identifying Throughput Issue
>
> I have a section of my network that is lacking something, and I can't
> figure out where the problem is. I'm looking for any thoughts /
> suggestions.
>
> 4x MikroTik routers
>
> Link speeds:
> R1 -- (30 Mbps IP / Transport) -- R2 -- (100 Mbps Eth) -- R3 -- (75 Mbps
> UBNT M5) -- R4
>
> The limiting factor for traffic should be the Transport, and I expect to be
> able to get 30 Mbps across the system (one-way).
>
> Testing from R1 to R3 runs 30 Mbps.
>
> Testing from R2 to R4 runs 75 Mbps.
>
> Testing from R1 to R4 only runs 10 Mbps (instead of 30).
>
> Tests were one-way btest with 20 TCP streams.
>
> Any ideas for something that would cause this?
>



-- 
If you only see yourself as part of the team but you don't see your team as
part of yourself you have already failed as part of the team.

Re: [AFMUG] new powerinjector plus

[Ryan Mano]

thanks

From: Af [mailto:af-boun...@afmug.com] On Behalf Of George Skorup
Sent: Wednesday, October 26, 2016 2:22 PM
To: af@afmug.com
Subject: Re: [AFMUG] new powerinjector plus

Put your +24 on PowerA and return/negative to Common. Plug in SyncPIpe/Box. 
Plug in radios.

All of the internal jumpers are set for PowerA and -4,5 +7,8 by default.

If you were doing 450i, you still wouldn't have to change anything. Just put 
+48 on PowerA and plug everything in.

Pretty much the only time you need to get to the jumpers is if you're doing a 
mix of 24 and 48 volt radios and/or 4 vs 2 pair power on any radio (as far as 
the GigE version anyway).
On 10/26/2016 12:55 PM, Ryan Mano wrote:
Can someone tell me how to wire the 4 pin green connector
�
It use to be 2 prong but its 4 now its labeled PowerA PowerB Comon and 
shield�am just powering up pmp100�s
�
thanks

Re: [AFMUG] new powerinjector plus

[George Skorup]

Put your +24 on PowerA and return/negative to Common. Plug in 
SyncPIpe/Box. Plug in radios.


All of the internal jumpers are set for PowerA and -4,5 +7,8 by default.

If you were doing 450i, you still wouldn't have to change anything. Just 
put +48 on PowerA and plug everything in.


Pretty much the only time you need to get to the jumpers is if you're 
doing a mix of 24 and 48 volt radios and/or 4 vs 2 pair power on any 
radio (as far as the GigE version anyway).


On 10/26/2016 12:55 PM, Ryan Mano wrote:


Can someone tell me how to wire the 4 pin green connector

It use to be 2 prong but its 4 now its labeled PowerA PowerB Comon and 
shield�am just powering up pmp100�s


thanks

Re: [AFMUG] Trouble Identifying Throughput Issue

[Steve]

Few questions come to mind.  

Are all set to auto negotiate or are they fixed at 100Mbit?  
What are the MTU's of each connection? 
Flow control turned on?


- Original Message -
From: "Christopher Gray" 
To: "af" 
Sent: Wednesday, October 26, 2016 1:57:56 PM
Subject: [AFMUG] Trouble Identifying Throughput Issue

I have a section of my network that is lacking something, and I can't
figure out where the problem is. I'm looking for any thoughts / suggestions.

4x MikroTik routers

Link speeds:
R1 -- (30 Mbps IP / Transport) -- R2 -- (100 Mbps Eth) -- R3 -- (75 Mbps
UBNT M5) -- R4

The limiting factor for traffic should be the Transport, and I expect to be
able to get 30 Mbps across the system (one-way).

Testing from R1 to R3 runs 30 Mbps.

Testing from R2 to R4 runs 75 Mbps.

Testing from R1 to R4 only runs 10 Mbps (instead of 30).

Tests were one-way btest with 20 TCP streams.

Any ideas for something that would cause this?

[AFMUG] Trouble Identifying Throughput Issue

[Christopher Gray]

I have a section of my network that is lacking something, and I can't
figure out where the problem is. I'm looking for any thoughts / suggestions.

4x MikroTik routers

Link speeds:
R1 -- (30 Mbps IP / Transport) -- R2 -- (100 Mbps Eth) -- R3 -- (75 Mbps
UBNT M5) -- R4

The limiting factor for traffic should be the Transport, and I expect to be
able to get 30 Mbps across the system (one-way).

Testing from R1 to R3 runs 30 Mbps.

Testing from R2 to R4 runs 75 Mbps.

Testing from R1 to R4 only runs 10 Mbps (instead of 30).

Tests were one-way btest with 20 TCP streams.

Any ideas for something that would cause this?

[AFMUG] new powerinjector plus

[Ryan Mano]

Can someone tell me how to wire the 4 pin green connector

It use to be 2 prong but its 4 now its labeled PowerA PowerB Comon and 
shield...am just powering up pmp100's

thanks

Re: [AFMUG] Mikrotik DNS Cache

[Josh Luthman]

Not limited to v6 as I have the same issues but no v6

Josh Luthman
Office: 937-552-2340
Direct: 937-552-2343
1100 Wayne St
Suite 1337
Troy, OH 45373

On Oct 26, 2016 1:42 PM, "Jesse DuPont" 
wrote:

> I, too, seem to have weird issues where the Mikrotik will slowly stop
> doing recursive lookups until the cache is flushed. Not sure if it's
> related to using IPv6 DNS addresses or what. I've moved back to giving out
> my on-net DNS servers instead of using the Mirkotik cache/proxy.
>
> *Jesse DuPont*
>
> Network Architect
> email: jesse.dup...@celeritycorp.net
> Celerity Networks LLC
>
> Celerity Broadband LLC
> Like us! facebook.com/celeritynetworksllc
>
> Like us! facebook.com/celeritybroadband
> On 10/26/16 7:55 AM, Josh Reynolds wrote:
>
> Shouldn't be any issues, no.
>
> On Oct 26, 2016 8:54 AM, "Matt"  wrote:
>
>> Is anyone using the Mikrotik DNS cache as there primary DNS resolver
>> for there clients?  Say use a CCR and your largest upstreams DNS
>> server as parent.  Should there be any issues with that?
>>
>
>

Re: [AFMUG] Mikrotik DNS Cache

[Jesse DuPont]

  
  
I, too, seem to have weird issues where the Mikrotik will slowly
stop doing recursive lookups until the cache is flushed. Not sure if
it's related to using IPv6 DNS addresses or what. I've moved back to
giving out my on-net DNS servers instead of using the Mirkotik
cache/proxy.


  
  
  
  
  
  
  
  
  
  
  
  
  
Jesse DuPont

  Network
  Architect
  email: jesse.dup...@celeritycorp.net
  Celerity Networks LLC
  Celerity
  Broadband LLC
Like us! facebook.com/celeritynetworksllc
  Like us! facebook.com/celeritybroadband
  

  

On 10/26/16 7:55 AM, Josh Reynolds
  wrote:


  Shouldn't be any issues, no.
  
On Oct 26, 2016 8:54 AM, "Matt" 
  wrote:
  Is anyone
using the Mikrotik DNS cache as there primary DNS resolver
for there clients?  Say use a CCR and your largest upstreams
DNS
server as parent.  Should there be any issues with that?
  

  


  

Re: [AFMUG] Take one down, put one up.

[Cameron Crum]

I would have thought the crane would be much more. I thought those guys had
a minimum charge just to get the thing out.

On Wed, Oct 26, 2016 at 11:29 AM, Paul Stewart  wrote:

> Wow.. that’s really great price for sure!  Any engineering/design costs
> you can share?
>
> On Oct 26, 2016, at 12:18 PM, Ben Royer  wrote:
>
> Hey Guys,
>
> Sorry for the delay, took me a minute to gather the totals.  It’s going to
> throw you guys off probably as I’ve got some great deals on some stuff,
> especially crane, but here goes.
>
> Rohn SSV – 160’ 9 through 2 – $12,500 approx.
> Excavation, backfill, jet, and rock – $1,325
> Concrete – $1,365
> Crane – $420 (Lower old, lift new)
> _
> Total – $ 15,610, or $16,000 approx.
>
> Thank you,
> Ben Royer, Operations Manager
> Royell Communications, Inc.
> 217-965-3699 www.royell.net
>
> *From:* Kurt Fankhauser 
> *Sent:* Monday, October 24, 2016 4:22 PM
> *To:* af@afmug.com
> *Subject:* Re: [AFMUG] Take one down, put one up.
>
> What was the cost of this 160' SSV tower just the tower alone? I put up an
> American Tower Co 160' that was about $20k for tower, plus 8k concrete,
> plus 2.5k for crane and we picked it up in 1 piece.
>
> On Mon, Oct 24, 2016 at 3:00 PM, Ben Royer  wrote:
>
>> Thanks, fun getting to put this stuff together.
>>
>> Thank you,
>> Ben Royer, Operations Manager
>> Royell Communications, Inc.
>> 217-965-3699 www.royell.net
>>
>> *From:* Jaime Solorza 
>> *Sent:* Monday, October 24, 2016 11:27 AM
>> *To:* Animal Farm 
>> *Subject:* Re: [AFMUG] Take one down, put one up.
>>
>> Awesome video
>>
>> On Oct 24, 2016 9:59 AM, "Ben Royer"  wrote:
>>
>>> Replaced a 90’ SSV with a 160’ SSV, and upgraded the site from PMP100
>>> 900Mhz. to 450i 5Ghz. in the process.  Also threw some 820s backhauls in
>>> there too.  The take down of old tower and installation of new tower only
>>> took a couple of hours once the crane got there, as we had to move the old
>>> 900mhz. from one tower to the next.
>>>
>>> http://www.benroyer.net/thayer.mp4
>>>
>>> Thank you,
>>> Ben Royer, Operations Manager
>>> Royell Communications, Inc.
>>> 217-965-3699 www.royell.net
>>>
>>
>
>
>

Re: [AFMUG] Kodi question

[Paul Stewart]

Usually a black screen means there is no “signal” .. the Internet stream should 
still be flowing through though …. 

This is a bit of a guess as the client side could default to showing a black 
screen when no stream is occurring ….



> On Oct 26, 2016, at 12:57 PM, Ken Hohhof  wrote:
> 
> Question for someone familiar with Android based Kodi boxes.  (I know nothing 
> about them.)
>  
> If customer is watching a video and gets a black screen, can that possibly be 
> Internet related?  Seems to me you should at least get a menu, but customer 
> is being told by tech support it means “your Internet is slow”.
>  
> I am assuming it is an Android based Kodi box, since the hostname shows up in 
> DHCP as Android-blahblahblah.  Customer says no, this is from Seventh Day 
> Adventists to watch their TV channels, they have their own SDATV box.  Yeah, 
> right.  This is all I need, a customer who is one step away from being Amish 
> in the sense of being technology averse, and someone sends her a Kodi box to 
> watch holistic Christian TV.  She doesn’t even have a working computer, 
> although she does have an iPhone.  When I said it looked like she had an 
> Android device, you’d think I had accused her of having a Satanic shrine in 
> her house.  I wonder if the SDATV box came with a remote, and if it is WiFi 
> based.

[AFMUG] Kodi question

[Ken Hohhof]

Question for someone familiar with Android based Kodi boxes.  (I know
nothing about them.)

 

If customer is watching a video and gets a black screen, can that possibly
be Internet related?  Seems to me you should at least get a menu, but
customer is being told by tech support it means "your Internet is slow".

 

I am assuming it is an Android based Kodi box, since the hostname shows up
in DHCP as Android-blahblahblah.  Customer says no, this is from Seventh Day
Adventists to watch their TV channels, they have their own SDATV box.  Yeah,
right.  This is all I need, a customer who is one step away from being Amish
in the sense of being technology averse, and someone sends her a Kodi box to
watch holistic Christian TV.  She doesn't even have a working computer,
although she does have an iPhone.  When I said it looked like she had an
Android device, you'd think I had accused her of having a Satanic shrine in
her house.  I wonder if the SDATV box came with a remote, and if it is WiFi
based.

Re: [AFMUG] Take one down, put one up.

[Paul Stewart]

Wow.. that’s really great price for sure!  Any engineering/design costs you can 
share?

> On Oct 26, 2016, at 12:18 PM, Ben Royer  wrote:
> 
> Hey Guys,
>  
> Sorry for the delay, took me a minute to gather the totals.  It’s going to 
> throw you guys off probably as I’ve got some great deals on some stuff, 
> especially crane, but here goes.
>  
> Rohn SSV – 160’ 9 through 2 – $12,500 approx.
> Excavation, backfill, jet, and rock – $1,325
> Concrete – $1,365
> Crane – $420 (Lower old, lift new)
> _
> Total – $ 15,610, or $16,000 approx.
>  
> Thank you,
> Ben Royer, Operations Manager
> Royell Communications, Inc.
> 217-965-3699 www.royell.net
>  
> From: Kurt Fankhauser 
> Sent: Monday, October 24, 2016 4:22 PM
> To: af@afmug.com 
> Subject: Re: [AFMUG] Take one down, put one up.
>  
> What was the cost of this 160' SSV tower just the tower alone? I put up an 
> American Tower Co 160' that was about $20k for tower, plus 8k concrete, plus 
> 2.5k for crane and we picked it up in 1 piece.
>  
> On Mon, Oct 24, 2016 at 3:00 PM, Ben Royer  > wrote:
>> Thanks, fun getting to put this stuff together.
>>  
>> Thank you,
>> Ben Royer, Operations Manager
>> Royell Communications, Inc.
>> 217-965-3699  www.royell.net 
>>  
>> From: Jaime Solorza 
>> Sent: Monday, October 24, 2016 11:27 AM
>> To: Animal Farm 
>> Subject: Re: [AFMUG] Take one down, put one up.
>>  
>> Awesome video
>>  
>> On Oct 24, 2016 9:59 AM, "Ben Royer" > > wrote:
>>> Replaced a 90’ SSV with a 160’ SSV, and upgraded the site from PMP100 
>>> 900Mhz. to 450i 5Ghz. in the process.  Also threw some 820s backhauls in 
>>> there too.  The take down of old tower and installation of new tower only 
>>> took a couple of hours once the crane got there, as we had to move the old 
>>> 900mhz. from one tower to the next.
>>>  
>>> http://www.benroyer.net/thayer.mp4 
>>>  
>>> Thank you,
>>> Ben Royer, Operations Manager
>>> Royell Communications, Inc.
>>> 217-965-3699  www.royell.net 
>  

Re: [AFMUG] Take one down, put one up.

[That One Guy /sarcasm]

lol we have a customer telling us he will put up a 300 foot tower to get
our service. I mean we are pretty good and all, but i think he knows not
what he speaks. and he only needs 60 foot

On Wed, Oct 26, 2016 at 11:18 AM, Ben Royer  wrote:

> Hey Guys,
>
> Sorry for the delay, took me a minute to gather the totals.  It’s going to
> throw you guys off probably as I’ve got some great deals on some stuff,
> especially crane, but here goes.
>
> Rohn SSV – 160’ 9 through 2 – $12,500 approx.
> Excavation, backfill, jet, and rock – $1,325
> Concrete – $1,365
> Crane – $420 (Lower old, lift new)
> _
> Total – $ 15,610, or $16,000 approx.
>
> Thank you,
> Ben Royer, Operations Manager
> Royell Communications, Inc.
> 217-965-3699 www.royell.net
>
> *From:* Kurt Fankhauser 
> *Sent:* Monday, October 24, 2016 4:22 PM
> *To:* af@afmug.com
> *Subject:* Re: [AFMUG] Take one down, put one up.
>
> What was the cost of this 160' SSV tower just the tower alone? I put up an
> American Tower Co 160' that was about $20k for tower, plus 8k concrete,
> plus 2.5k for crane and we picked it up in 1 piece.
>
> On Mon, Oct 24, 2016 at 3:00 PM, Ben Royer  wrote:
>
>> Thanks, fun getting to put this stuff together.
>>
>> Thank you,
>> Ben Royer, Operations Manager
>> Royell Communications, Inc.
>> 217-965-3699 www.royell.net
>>
>> *From:* Jaime Solorza 
>> *Sent:* Monday, October 24, 2016 11:27 AM
>> *To:* Animal Farm 
>> *Subject:* Re: [AFMUG] Take one down, put one up.
>>
>>
>> Awesome video
>>
>> On Oct 24, 2016 9:59 AM, "Ben Royer"  wrote:
>>
>>> Replaced a 90’ SSV with a 160’ SSV, and upgraded the site from PMP100
>>> 900Mhz. to 450i 5Ghz. in the process.  Also threw some 820s backhauls in
>>> there too.  The take down of old tower and installation of new tower only
>>> took a couple of hours once the crane got there, as we had to move the old
>>> 900mhz. from one tower to the next.
>>>
>>> http://www.benroyer.net/thayer.mp4
>>>
>>> Thank you,
>>> Ben Royer, Operations Manager
>>> Royell Communications, Inc.
>>> 217-965-3699 www.royell.net
>>>
>>
>



-- 
If you only see yourself as part of the team but you don't see your team as
part of yourself you have already failed as part of the team.

Re: [AFMUG] Take one down, put one up.

[Ben Royer]

Hey Guys,

Sorry for the delay, took me a minute to gather the totals.  It’s going to 
throw you guys off probably as I’ve got some great deals on some stuff, 
especially crane, but here goes.

Rohn SSV – 160’ 9 through 2 – $12,500 approx.
Excavation, backfill, jet, and rock – $1,325
Concrete – $1,365
Crane – $420 (Lower old, lift new)
_
Total – $ 15,610, or $16,000 approx.

Thank you,
Ben Royer, Operations Manager
Royell Communications, Inc.
217-965-3699 www.royell.net

From: Kurt Fankhauser 
Sent: Monday, October 24, 2016 4:22 PM
To: af@afmug.com 
Subject: Re: [AFMUG] Take one down, put one up.

What was the cost of this 160' SSV tower just the tower alone? I put up an 
American Tower Co 160' that was about $20k for tower, plus 8k concrete, plus 
2.5k for crane and we picked it up in 1 piece.

On Mon, Oct 24, 2016 at 3:00 PM, Ben Royer  wrote:

  Thanks, fun getting to put this stuff together.

  Thank you,
  Ben Royer, Operations Manager
  Royell Communications, Inc.
  217-965-3699 www.royell.net

  From: Jaime Solorza 
  Sent: Monday, October 24, 2016 11:27 AM
  To: Animal Farm 
  Subject: Re: [AFMUG] Take one down, put one up.

  Awesome video


  On Oct 24, 2016 9:59 AM, "Ben Royer"  wrote:

Replaced a 90’ SSV with a 160’ SSV, and upgraded the site from PMP100 
900Mhz. to 450i 5Ghz. in the process.  Also threw some 820s backhauls in there 
too.  The take down of old tower and installation of new tower only took a 
couple of hours once the crane got there, as we had to move the old 900mhz. 
from one tower to the next.

http://www.benroyer.net/thayer.mp4

Thank you,
Ben Royer, Operations Manager
Royell Communications, Inc.
217-965-3699 www.royell.net

[AFMUG] ISP Radio Today

[Dennis Burgess]

www.ispradio.com - Going live at 11am CST, Alex 
Phillips and Trina Coffee discuss WISPAPALOOZA 2016!!!


Dennis Burgess - Network Solution Engineer - Consultant
MikroTik Certified 
Trainer/Consultant
 - MTCNA, MTCRE, MTCWE, MTCTCE, MTCINE

For Wireless Hardware/Routers visit www.linktechs.net
Radio Frequiency Coverages: www.towercoverage.com
Office: 314-735-0270
E-Mail: dmburg...@linktechs.net

Re: [AFMUG] Mikrotik DNS Cache

[Ken Hohhof]

BIND or Unbound are free other than your time to install them, and pretty much 
any x86 box you can scrounge up will have enough horsepower, so again free 
other than the power to run it.  You should probably have an authoritative DNS 
server in addition to resolvers.  And it never hurts to have a couple *nix 
boxes on your network for miscellaneous testing and troubleshooting.

 

If you can’t afford a couple standalone DNS servers or don’t have a suitable 
NOC environment to locate them, I’d consider maybe something like OpenDNS.

 

I would also throw out that many DDoS attacks involve DNS, so I think I’d want 
my router to be acting as a router and firewall, and some separate server can 
be the target of some amplification or IoT-based DNS attack.

 

 

From: Af [mailto:af-boun...@afmug.com] On Behalf Of Josh Reynolds
Sent: Wednesday, October 26, 2016 10:02 AM
To: af@afmug.com
Subject: Re: [AFMUG] Mikrotik DNS Cache

 

I'm running two ad/malware/ransomware blocking, recursive, caching dns servers 
right now: one in Chicago and one in Dallas. My local one caches results from 
those.

... And this is for my house :P

 

On Oct 26, 2016 9:45 AM, "Dennis Burgess"  > wrote:

Does it work, yes it is the same as a high performance DNS server, no.  Is a 
dedicated DNS resolvers expensive, no.  Getting starting say under 100-150 
users, ok, for a while, once you go over that, really need to move to dedicated 
resolvers.


Dennis Burgess – Network Solution Engineer – Consultant
MikroTik Certified Trainer/Consultant – MTCNA, MTCRE, MTCWE, MTCTCE, MTCINE

For Wireless Hardware/Routers visit www.linktechs.net 
 
Radio Frequiency Coverages: www.towercoverage.com 
 
Office: 314-735-0270  
E-Mail: dmburg...@linktechs.net  

-Original Message-
From: Af [mailto:af-boun...@afmug.com  ] On Behalf 
Of Matt
Sent: Wednesday, October 26, 2016 8:54 AM
To: af@afmug.com  
Subject: [AFMUG] Mikrotik DNS Cache

Is anyone using the Mikrotik DNS cache as there primary DNS resolver for there 
clients?  Say use a CCR and your largest upstreams DNS server as parent.  
Should there be any issues with that?

Re: [AFMUG] Mikrotik DNS Cache

[Steve]

Webmin makes it terribly easy for starters.  You can run it on a very very 
small footprint.  Even as Raspberry PI. As long as you change your ports and 
keep it locked down via firewall,  make ACL's for access (upstream as well) you 
should be safe.  

Webmin has been known to be insecure.  

- Original Message -
From: "That One Guy /sarcasm" 
To: "af" 
Sent: Wednesday, October 26, 2016 11:06:30 AM
Subject: Re: [AFMUG] Mikrotik DNS Cache

we just got up two bind dns servers, turned out to not be all that
complicated, uses very little resources for our size network. Ill get shot
for saying it, but I used webmin, made it really simple, just make sure you
ACL the webmin port to your managment IPs and ACL your recursion to your
customer IPs. We did the ACLs on a mikrotik thats external to the server.
We had considered what you asked, but as part of the routing infrastructure
it didnt seem appropriate, leave the routers to route. Even dropping one in
as a dedicated box didnt seem appropriate as it is not its defined purpose.
And when SHTF for whatever, and we would have needed to reach out, getting
help on BIND issues would be a whole lot more likely than getting fast and
efficient help on a wrong purposed router.



On Wed, Oct 26, 2016 at 9:58 AM, Dennis Burgess 
wrote:

> Yep, its not perfect, sooner you can get on your own DNS servers the
> better. Should be a standard practice for ISPs.
>
>
>
> We have done many of them for clients, its simple and easy to do, there
> are plenty of not super expensive options out there, some are free other
> than the server.
>
>
>
> *Dennis Burgess** –** Network Solution Engineer – Consultant *
>
> MikroTik Certified Trainer/Consultant
>  –
> MTCNA, MTCRE, MTCWE, MTCTCE, MTCINE
>
>
>
> For Wireless Hardware/Routers visit www.linktechs.net
>
> Radio Frequiency Coverages: www.towercoverage.com
>
> Office: 314-735-0270
>
> E-Mail: dmburg...@linktechs.net
>
>
>
> *From:* Af [mailto:af-boun...@afmug.com] *On Behalf Of *Josh Luthman
> *Sent:* Wednesday, October 26, 2016 9:50 AM
> *To:* af@afmug.com
> *Subject:* Re: [AFMUG] Mikrotik DNS Cache
>
>
>
> I had issues with just myself...
>
> Josh Luthman
> Office: 937-552-2340
> Direct: 937-552-2343
> 1100 Wayne St
> Suite 1337
> Troy, OH 45373
>
>
>
> On Oct 26, 2016 10:45 AM, "Dennis Burgess" 
> wrote:
>
> Does it work, yes it is the same as a high performance DNS server, no.  Is
> a dedicated DNS resolvers expensive, no.  Getting starting say under
> 100-150 users, ok, for a while, once you go over that, really need to move
> to dedicated resolvers.
>
>
> Dennis Burgess – Network Solution Engineer – Consultant
> MikroTik Certified Trainer/Consultant – MTCNA, MTCRE, MTCWE, MTCTCE, MTCINE
>
> For Wireless Hardware/Routers visit www.linktechs.net
> Radio Frequiency Coverages: www.towercoverage.com
> Office: 314-735-0270
> E-Mail: dmburg...@linktechs.net
>
> -Original Message-
> From: Af [mailto:af-boun...@afmug.com] On Behalf Of Matt
> Sent: Wednesday, October 26, 2016 8:54 AM
> To: af@afmug.com
> Subject: [AFMUG] Mikrotik DNS Cache
>
> Is anyone using the Mikrotik DNS cache as there primary DNS resolver for
> there clients?  Say use a CCR and your largest upstreams DNS server as
> parent.  Should there be any issues with that?
>
>


-- 
If you only see yourself as part of the team but you don't see your team as
part of yourself you have already failed as part of the team.

Re: [AFMUG] Mikrotik DNS Cache

[That One Guy /sarcasm]

we just got up two bind dns servers, turned out to not be all that
complicated, uses very little resources for our size network. Ill get shot
for saying it, but I used webmin, made it really simple, just make sure you
ACL the webmin port to your managment IPs and ACL your recursion to your
customer IPs. We did the ACLs on a mikrotik thats external to the server.
We had considered what you asked, but as part of the routing infrastructure
it didnt seem appropriate, leave the routers to route. Even dropping one in
as a dedicated box didnt seem appropriate as it is not its defined purpose.
And when SHTF for whatever, and we would have needed to reach out, getting
help on BIND issues would be a whole lot more likely than getting fast and
efficient help on a wrong purposed router.



On Wed, Oct 26, 2016 at 9:58 AM, Dennis Burgess 
wrote:

> Yep, its not perfect, sooner you can get on your own DNS servers the
> better. Should be a standard practice for ISPs.
>
>
>
> We have done many of them for clients, its simple and easy to do, there
> are plenty of not super expensive options out there, some are free other
> than the server.
>
>
>
> *Dennis Burgess** –** Network Solution Engineer – Consultant *
>
> MikroTik Certified Trainer/Consultant
>  –
> MTCNA, MTCRE, MTCWE, MTCTCE, MTCINE
>
>
>
> For Wireless Hardware/Routers visit www.linktechs.net
>
> Radio Frequiency Coverages: www.towercoverage.com
>
> Office: 314-735-0270
>
> E-Mail: dmburg...@linktechs.net
>
>
>
> *From:* Af [mailto:af-boun...@afmug.com] *On Behalf Of *Josh Luthman
> *Sent:* Wednesday, October 26, 2016 9:50 AM
> *To:* af@afmug.com
> *Subject:* Re: [AFMUG] Mikrotik DNS Cache
>
>
>
> I had issues with just myself...
>
> Josh Luthman
> Office: 937-552-2340
> Direct: 937-552-2343
> 1100 Wayne St
> Suite 1337
> Troy, OH 45373
>
>
>
> On Oct 26, 2016 10:45 AM, "Dennis Burgess" 
> wrote:
>
> Does it work, yes it is the same as a high performance DNS server, no.  Is
> a dedicated DNS resolvers expensive, no.  Getting starting say under
> 100-150 users, ok, for a while, once you go over that, really need to move
> to dedicated resolvers.
>
>
> Dennis Burgess – Network Solution Engineer – Consultant
> MikroTik Certified Trainer/Consultant – MTCNA, MTCRE, MTCWE, MTCTCE, MTCINE
>
> For Wireless Hardware/Routers visit www.linktechs.net
> Radio Frequiency Coverages: www.towercoverage.com
> Office: 314-735-0270
> E-Mail: dmburg...@linktechs.net
>
> -Original Message-
> From: Af [mailto:af-boun...@afmug.com] On Behalf Of Matt
> Sent: Wednesday, October 26, 2016 8:54 AM
> To: af@afmug.com
> Subject: [AFMUG] Mikrotik DNS Cache
>
> Is anyone using the Mikrotik DNS cache as there primary DNS resolver for
> there clients?  Say use a CCR and your largest upstreams DNS server as
> parent.  Should there be any issues with that?
>
>


-- 
If you only see yourself as part of the team but you don't see your team as
part of yourself you have already failed as part of the team.

Re: [AFMUG] Mikrotik DNS Cache

[Steve]

I wouldn't do it. But it would work fine.  Use your routers as routers, your 
switches as switches and your servers as servers.  

I like the versatility that Mikrotik has as well but its a best practice to 
keep each use specifically for its own purpose.  The more eggs in one basket 
shall we say ... 


- Original Message -
From: "Matt" 
To: "af" 
Sent: Wednesday, October 26, 2016 9:54:28 AM
Subject: [AFMUG] Mikrotik DNS Cache

Is anyone using the Mikrotik DNS cache as there primary DNS resolver
for there clients?  Say use a CCR and your largest upstreams DNS
server as parent.  Should there be any issues with that?

Re: [AFMUG] Mikrotik DNS Cache

[Josh Reynolds]

I'm running two ad/malware/ransomware blocking, recursive, caching dns
servers right now: one in Chicago and one in Dallas. My local one caches
results from those.

... And this is for my house :P

On Oct 26, 2016 9:45 AM, "Dennis Burgess"  wrote:

> Does it work, yes it is the same as a high performance DNS server, no.  Is
> a dedicated DNS resolvers expensive, no.  Getting starting say under
> 100-150 users, ok, for a while, once you go over that, really need to move
> to dedicated resolvers.
>
>
> Dennis Burgess – Network Solution Engineer – Consultant
> MikroTik Certified Trainer/Consultant – MTCNA, MTCRE, MTCWE, MTCTCE, MTCINE
>
> For Wireless Hardware/Routers visit www.linktechs.net
> Radio Frequiency Coverages: www.towercoverage.com
> Office: 314-735-0270
> E-Mail: dmburg...@linktechs.net
>
> -Original Message-
> From: Af [mailto:af-boun...@afmug.com] On Behalf Of Matt
> Sent: Wednesday, October 26, 2016 8:54 AM
> To: af@afmug.com
> Subject: [AFMUG] Mikrotik DNS Cache
>
> Is anyone using the Mikrotik DNS cache as there primary DNS resolver for
> there clients?  Say use a CCR and your largest upstreams DNS server as
> parent.  Should there be any issues with that?
>

[AFMUG] ePMP DNS on Management / Private IP's

[Steve]

We are trying to add ePMP's to CNMaestro by hostname (in case I need to change 
our server's IP in the future).  The A record for our cnmaestro server is a 
private IP for our internal network. We don't want it public for now.  

The odd part is from the limited cli we can ping outside domains by hostname 
but not any domain that we have A records pointing to private records. 10.x.x.x 
172.16.x.x etc.  The DNS servers that are set up in the network configuration 
for the NAT ip are indeed correct and our servers that host the zone we are 
using.  But it will not resolve a hostname to a private IP. 

Anyone else encounter this?

Re: [AFMUG] Mikrotik DNS Cache

[Dennis Burgess]

Yep, its not perfect, sooner you can get on your own DNS servers the better. 
Should be a standard practice for ISPs.

We have done many of them for clients, its simple and easy to do, there are 
plenty of not super expensive options out there, some are free other than the 
server.

Dennis Burgess – Network Solution Engineer – Consultant
MikroTik Certified 
Trainer/Consultant
 – MTCNA, MTCRE, MTCWE, MTCTCE, MTCINE

For Wireless Hardware/Routers visit www.linktechs.net
Radio Frequiency Coverages: www.towercoverage.com
Office: 314-735-0270
E-Mail: dmburg...@linktechs.net

From: Af [mailto:af-boun...@afmug.com] On Behalf Of Josh Luthman
Sent: Wednesday, October 26, 2016 9:50 AM
To: af@afmug.com
Subject: Re: [AFMUG] Mikrotik DNS Cache


I had issues with just myself...

Josh Luthman
Office: 937-552-2340
Direct: 937-552-2343
1100 Wayne St
Suite 1337
Troy, OH 45373

On Oct 26, 2016 10:45 AM, "Dennis Burgess" 
> wrote:
Does it work, yes it is the same as a high performance DNS server, no.  Is a 
dedicated DNS resolvers expensive, no.  Getting starting say under 100-150 
users, ok, for a while, once you go over that, really need to move to dedicated 
resolvers.


Dennis Burgess – Network Solution Engineer – Consultant
MikroTik Certified Trainer/Consultant – MTCNA, MTCRE, MTCWE, MTCTCE, MTCINE

For Wireless Hardware/Routers visit www.linktechs.net
Radio Frequiency Coverages: www.towercoverage.com
Office: 314-735-0270
E-Mail: dmburg...@linktechs.net

-Original Message-
From: Af [mailto:af-boun...@afmug.com] On Behalf 
Of Matt
Sent: Wednesday, October 26, 2016 8:54 AM
To: af@afmug.com
Subject: [AFMUG] Mikrotik DNS Cache

Is anyone using the Mikrotik DNS cache as there primary DNS resolver for there 
clients?  Say use a CCR and your largest upstreams DNS server as parent.  
Should there be any issues with that?

Re: [AFMUG] Mikrotik DNS Cache

[Josh Luthman]

I had issues with just myself...

Josh Luthman
Office: 937-552-2340
Direct: 937-552-2343
1100 Wayne St
Suite 1337
Troy, OH 45373

On Oct 26, 2016 10:45 AM, "Dennis Burgess"  wrote:

> Does it work, yes it is the same as a high performance DNS server, no.  Is
> a dedicated DNS resolvers expensive, no.  Getting starting say under
> 100-150 users, ok, for a while, once you go over that, really need to move
> to dedicated resolvers.
>
>
> Dennis Burgess – Network Solution Engineer – Consultant
> MikroTik Certified Trainer/Consultant – MTCNA, MTCRE, MTCWE, MTCTCE, MTCINE
>
> For Wireless Hardware/Routers visit www.linktechs.net
> Radio Frequiency Coverages: www.towercoverage.com
> Office: 314-735-0270
> E-Mail: dmburg...@linktechs.net
>
> -Original Message-
> From: Af [mailto:af-boun...@afmug.com] On Behalf Of Matt
> Sent: Wednesday, October 26, 2016 8:54 AM
> To: af@afmug.com
> Subject: [AFMUG] Mikrotik DNS Cache
>
> Is anyone using the Mikrotik DNS cache as there primary DNS resolver for
> there clients?  Say use a CCR and your largest upstreams DNS server as
> parent.  Should there be any issues with that?
>

Re: [AFMUG] Mikrotik DNS Cache

[Paul Stewart]

Everyone’s network and traffic is different but using a router for DNS resolver 
is typically a bad idea …. if there is an influx of dirty DNS traffic and/or 
queries is there risk that you’ll take down the box?

I know nothing about Microtik but just the principal of using  your “core 
router” for DNS lookups for customers sends chills down my spine… just saying ;)


> On Oct 26, 2016, at 9:55 AM, Josh Reynolds  wrote:
> 
> Shouldn't be any issues, no.
> 
> 
> On Oct 26, 2016 8:54 AM, "Matt"  > wrote:
> Is anyone using the Mikrotik DNS cache as there primary DNS resolver
> for there clients?  Say use a CCR and your largest upstreams DNS
> server as parent.  Should there be any issues with that?

Re: [AFMUG] Mikrotik DNS Cache

[Dennis Burgess]

Does it work, yes it is the same as a high performance DNS server, no.  Is a 
dedicated DNS resolvers expensive, no.  Getting starting say under 100-150 
users, ok, for a while, once you go over that, really need to move to dedicated 
resolvers.  


Dennis Burgess – Network Solution Engineer – Consultant 
MikroTik Certified Trainer/Consultant – MTCNA, MTCRE, MTCWE, MTCTCE, MTCINE

For Wireless Hardware/Routers visit www.linktechs.net
Radio Frequiency Coverages: www.towercoverage.com 
Office: 314-735-0270
E-Mail: dmburg...@linktechs.net 

-Original Message-
From: Af [mailto:af-boun...@afmug.com] On Behalf Of Matt
Sent: Wednesday, October 26, 2016 8:54 AM
To: af@afmug.com
Subject: [AFMUG] Mikrotik DNS Cache

Is anyone using the Mikrotik DNS cache as there primary DNS resolver for there 
clients?  Say use a CCR and your largest upstreams DNS server as parent.  
Should there be any issues with that?

Re: [AFMUG] Mikrotik DNS Cache

[Josh Luthman]

I've had random issues with it.  Do as Mike suggested IMO

Josh Luthman
Office: 937-552-2340
Direct: 937-552-2343
1100 Wayne St
Suite 1337
Troy, OH 45373

On Oct 26, 2016 9:55 AM, "Josh Reynolds"  wrote:

> Shouldn't be any issues, no.
>
> On Oct 26, 2016 8:54 AM, "Matt"  wrote:
>
>> Is anyone using the Mikrotik DNS cache as there primary DNS resolver
>> for there clients?  Say use a CCR and your largest upstreams DNS
>> server as parent.  Should there be any issues with that?
>>
>

Re: [AFMUG] Mikrotik DNS Cache

[Mike Hammett]

Always use a real, on-net DNS resolver. 




- 
Mike Hammett 
Intelligent Computing Solutions 

Midwest Internet Exchange 

The Brothers WISP 




- Original Message -

From: "Matt"  
To: af@afmug.com 
Sent: Wednesday, October 26, 2016 8:54:28 AM 
Subject: [AFMUG] Mikrotik DNS Cache 

Is anyone using the Mikrotik DNS cache as there primary DNS resolver 
for there clients? Say use a CCR and your largest upstreams DNS 
server as parent. Should there be any issues with that? 

Re: [AFMUG] Mikrotik DNS Cache

[Josh Reynolds]

Shouldn't be any issues, no.

On Oct 26, 2016 8:54 AM, "Matt"  wrote:

> Is anyone using the Mikrotik DNS cache as there primary DNS resolver
> for there clients?  Say use a CCR and your largest upstreams DNS
> server as parent.  Should there be any issues with that?
>

[AFMUG] Mikrotik DNS Cache

[Matt]

Is anyone using the Mikrotik DNS cache as there primary DNS resolver
for there clients?  Say use a CCR and your largest upstreams DNS
server as parent.  Should there be any issues with that?

Re: [AFMUG] Google Fiber is no more

[Mike Hammett]

I don't think that's true. 




- 
Mike Hammett 
Intelligent Computing Solutions 

Midwest Internet Exchange 

The Brothers WISP 




- Original Message -

From: "Rory Conaway"  
To: af@afmug.com 
Sent: Tuesday, October 25, 2016 10:15:32 PM 
Subject: Re: [AFMUG] Google Fiber is no more 



Big surprise there. They built it and no one came. 

Rory 



From: Af [mailto:af-boun...@afmug.com] On Behalf Of Tushar Patel 
Sent: Tuesday, October 25, 2016 7:14 PM 
To: af@afmug.com 
Subject: Re: [AFMUG] Google Fiber is no more 


Their contractor are still hiring installer in Austin. 



Need to probably understand why those cities not others? 

Tushar 




On Oct 25, 2016, at 9:06 PM, Josh Reynolds < j...@kyneticwifi.com > wrote: 



New ones. They're still deploying existing networks. They just opened up a few 
new areas in Kansas City recently. 



On Oct 25, 2016 9:03 PM, "Jaime Solorza" < losguyswirel...@gmail.com > wrote: 


Moving folks to wireless Aye Dios 



On Oct 25, 2016 7:56 PM, "Gino Villarini" < ginovi...@gmail.com > wrote: 


https://gizmodo.com/google-fiber-halts-operations-in-ten-cities-1788214992?rev=1477443092657_campaign=socialflow_gizmodo_facebook_source=gizmodo_facebook_medium=socialflow
 

Re: [AFMUG] stp

[Josh Reynolds]

The merchant silicon hasn't really shifted down yet.

(for SoCs that handle those protocols, either/or)

On Wed, Oct 26, 2016 at 7:51 AM, Mike Hammett  wrote:

> It'd be nice if more vendors adopted TRILL or SPB.
>
> STP and RSTP are garbage.
>
>
>
> -
> Mike Hammett
> Intelligent Computing Solutions 
> 
> 
> 
> 
> Midwest Internet Exchange 
> 
> 
> 
> The Brothers WISP 
> 
>
>
> 
> --
> *From: *"CBB - Jay Fuller" 
> *To: *af@afmug.com
> *Sent: *Tuesday, October 25, 2016 11:03:10 AM
> *Subject: *[AFMUG] stp
>
>
> just discovered on one of our nettronix switches spanning tree protocol
> was enabled.
> we've run this switch probably four months - no real side effects - but i
> don't run stp anywhere
> else.  any reason to leave this on?
>
> thanks
>
>
>
>

Re: [AFMUG] stp

[Gino Villarini]

Both are still lacking features for MW based networks, but are a huge 
improvement over xSTP

From: Af > on behalf of Mike 
Hammett >
Reply-To: "af@afmug.com" 
>
Date: Wednesday, October 26, 2016 at 8:51 AM
To: "af@afmug.com" >
Subject: Re: [AFMUG] stp

It'd be nice if more vendors adopted TRILL or SPB.

STP and RSTP are garbage.



-
Mike Hammett
Intelligent Computing Solutions
[http://www.ics-il.com/images/fbicon.png][http://www.ics-il.com/images/googleicon.png][http://www.ics-il.com/images/linkedinicon.png][http://www.ics-il.com/images/twittericon.png]
Midwest Internet Exchange
[http://www.ics-il.com/images/fbicon.png][http://www.ics-il.com/images/linkedinicon.png][http://www.ics-il.com/images/twittericon.png]
The Brothers WISP
[http://www.ics-il.com/images/fbicon.png][http://www.ics-il.com/images/youtubeicon.png]






Gino Villarini


President
Metro Office Park #18 Suite 304 Guaynabo, Puerto Rico 00968

[cid:aeronet-logo_310cfc3e-6691-4f69-bd49-b37b834b9238.png]


From: "CBB - Jay Fuller" 
>
To: af@afmug.com
Sent: Tuesday, October 25, 2016 11:03:10 AM
Subject: [AFMUG] stp


just discovered on one of our nettronix switches spanning tree protocol was 
enabled.
we've run this switch probably four months - no real side effects - but i don't 
run stp anywhere
else.  any reason to leave this on?

thanks

Re: [AFMUG] OpenSRS Email?

[Jeremy Grip]

We still use it for a handful of customers.  The spam filtering has 
deteriorated over the last year or more.

Jeremey Grip
North Branch Networks, LLC

-Original Message-
From: Af [mailto:af-boun...@afmug.com] On Behalf Of Paul Stewart
Sent: Wednesday, October 26, 2016 6:12 AM
To: af@afmug.com
Subject: Re: [AFMUG] OpenSRS Email?

A customer I dealt with about 3 years ago was testing openSRS email … it may 
have changed since then, but the feedback was less than stellar.  I believe the 
email itself was less of a concern and the complaint was more about the spam 
presentation being less than effective in their opinion …

Totally second hand information - would suggest doing a trial with them maybe?


> On Oct 25, 2016, at 4:37 PM, Christopher Gray  
> wrote:
> 
> I'm looking for feedback on the OpenSRS email product. Does anyone here use 
> it / have any thoughts? Offlist is ok if necessary.
> 
> Thank you - Chris
> 

Re: [AFMUG] AF-5U vs. AF-5x

[Mike Hammett]

I'm more accepting of it on backhauls because there's another path. There isn't 
for PtMP. 

BFD or (preferably) the radio's ability to drop the Ethernet port should have 
taken the link down immediately. 




- 
Mike Hammett 
Intelligent Computing Solutions 

Midwest Internet Exchange 

The Brothers WISP 




- Original Message -

From: "Ken Hohhof"  
To: af@afmug.com 
Sent: Tuesday, October 25, 2016 9:14:08 AM 
Subject: Re: [AFMUG] AF-5U vs. AF-5x 



Exactly. I don’t consider a minute of downtime acceptable on a backhaul, 
especially if you have no control over when or how often it happens. Also it’s 
long enough to trigger OSPF to reroute traffic. 

Customer expectations about the Internet have changed, the phone may already be 
ringing before the AP has finished scanning and the SMs have reregistered on 
the new frequency. Their phone calls drop, their video pauses, they get kicked 
out of games, their VPN session to work drops. 




From: Af [mailto:af-boun...@afmug.com] On Behalf Of Mike Hammett 
Sent: Tuesday, October 25, 2016 9:05 AM 
To: af@afmug.com 
Subject: Re: [AFMUG] AF-5U vs. AF-5x 


There's still the LBT timer. 



- 
Mike Hammett 
Intelligent Computing Solutions 

Midwest Internet Exchange 

The Brothers WISP 




- Original Message -


From: "Kurt Fankhauser" < lists.wavel...@gmail.com > 
To: af@afmug.com 
Sent: Tuesday, October 25, 2016 9:03:35 AM 
Subject: Re: [AFMUG] AF-5U vs. AF-5x 

With the PMP450 you can select up to 2 alternative frequencies for the radio to 
go to if it gets a DFS hit. Very rarely would all 3 frequencies get hits and 
cause the AP to go down. 



On Tue, Oct 25, 2016 at 9:04 AM, Mike Hammett < af...@ics-il.net > wrote: 




DFS is great for PtP because if it goes down, you have another one on another 
path. Not so with PtMP. 



- 
Mike Hammett 
Intelligent Computing Solutions 

Midwest Internet Exchange 

The Brothers WISP 







From: "Ken Hohhof" < af...@kwisp.com > 
To: af@afmug.com 
Sent: Monday, October 24, 2016 9:14:47 AM 
Subject: Re: [AFMUG] AF-5U vs. AF-5x 

I think PTP links are narrow beams and carefully engineered. On the other hand 
some PTP radios allow auto frequency change. 

I would be very cautious about using those frequencies and make sure the path 
isn’t inline with any TDWR sites no matter the distance. For that matter, I 
really don’t like the idea of using DFS frequencies for PTP links, can you 
really afford the downtime for a false radar detect? Do you not have lower 5 
GHz non-DFS frequencies available to use instead? 


From: Af [mailto: af-boun...@afmug.com ] On Behalf Of Kurt Fankhauser 
Sent: Monday, October 24, 2016 8:57 AM 
To: af@afmug.com 
Subject: Re: [AFMUG] AF-5U vs. AF-5x 


I am 60 miles from closest TDWR. I just thought it was odd that the AF5X allows 
transmit on that freq but the PMP450 AP has those frequencies greyed out. I 
figured that Cambium would be certified for more freqs than UBNT. Maybe cambium 
just made that choice to make those freqs unavailable. 





On Mon, Oct 24, 2016 at 8:53 AM, Josh Luthman < j...@imaginenetworksllc.com > 
wrote: 


As long as you're at a safe distance yes. I'm sure you're more than 30 km from 
me for the TDWR here. 
Josh Luthman 
Office: 937-552-2340 
Direct: 937-552-2343 
1100 Wayne St 
Suite 1337 
Troy, OH 45373 



On Oct 24, 2016 8:30 AM, "Kurt Fankhauser" < lists.wavel...@gmail.com > wrote: 



Is AF5X legal to transmit on 5600-5650mhz? Firmware 3.2.1 allows it to be 
selected. Is that a mistake? 



On Mon, Oct 24, 2016 at 8:24 AM, Josh Reynolds < j...@kyneticwifi.com > wrote: 




Yeah, the U is upper band only. 
The AF-5 goes down to 5470 (if the datasheet is still accurate). Very good 
antennas, but limited gain. 1024QAM. GPS sync. Good radio for the money, as 
long as it works for your link distance. 





On Oct 24, 2016 7:14 AM, "Josh Luthman" < j...@imaginenetworksllc.com > wrote: 


I think one or two modulations. 
You can't do the lower frequencies on the U. 
Josh Luthman 
Office: 937-552-2340 
Direct: 937-552-2343 
1100 Wayne St 
Suite 1337 
Troy, OH 45373 



On Oct 24, 2016 8:05 AM, "Paul McCall" < pa...@pdmnet.net > wrote: 




Is there any difference performance wise on these two? 

What is the distance for full modulation on the AF-5U ? 

What is the longest link (at decent modulation) on the AF-5U 

I understand that will vary quite a bit with the normal environmental factors, 
but looking to get close on a comparison 

Paul McCall, President 
PDMNet, Inc. / Florida Broadband, Inc. 
658 Old Dixie Highway 
Vero Beach, FL 32962 
772-564-6800 
pa...@pdmnet.net 
www.pdmnet.com 
www.floridabroadband.com 

Re: [AFMUG] stp

[Mike Hammett]

It'd be nice if more vendors adopted TRILL or SPB. 

STP and RSTP are garbage. 




- 
Mike Hammett 
Intelligent Computing Solutions 

Midwest Internet Exchange 

The Brothers WISP 




- Original Message -

From: "CBB - Jay Fuller"  
To: af@afmug.com 
Sent: Tuesday, October 25, 2016 11:03:10 AM 
Subject: [AFMUG] stp 



just discovered on one of our nettronix switches spanning tree protocol was 
enabled. 
we've run this switch probably four months - no real side effects - but i don't 
run stp anywhere 
else. any reason to leave this on? 

thanks 

Re: [AFMUG] Cambium 650 vs AF5x/NxN

[Mike Hammett]

The platform would need to have stable enough timing so that when Txing it 
never spat into another radio Rxing. I know they sync, but the timing would be 
critical wired together. 




- 
Mike Hammett 
Intelligent Computing Solutions 

Midwest Internet Exchange 

The Brothers WISP 




- Original Message -

From: "Gino Villarini"  
To: "Animal Farm"  
Sent: Tuesday, October 25, 2016 3:42:51 PM 
Subject: Re: [AFMUG] Cambium 650 vs AF5x/NxN 


why not? the nxn adapters are just spliters afaik... 


On Tue, Oct 25, 2016 at 2:25 PM, Sam Morris < w...@csilogan.com > wrote: 


Does Cambium have the ability (using presumably ptp650) to bind radios together 
ala the NxN to achieve > 1 Gbps throughput in the 5 GHz spectrum? 

Re: [AFMUG] OpenSRS Email?

[Paul Stewart]

A customer I dealt with about 3 years ago was testing openSRS email … it may 
have changed since then, but the feedback was less than stellar.  I believe the 
email itself was less of a concern and the complaint was more about the spam 
presentation being less than effective in their opinion …

Totally second hand information - would suggest doing a trial with them maybe?


> On Oct 25, 2016, at 4:37 PM, Christopher Gray  
> wrote:
> 
> I'm looking for feedback on the OpenSRS email product. Does anyone here use 
> it / have any thoughts? Offlist is ok if necessary.
> 
> Thank you - Chris
> 

Re: [AFMUG] stp

[Paul Stewart]

Well said … it’s incredibly rare to see someone set this in their networks ….

(yes, I’m reading emails backwards and catching up)

> On Oct 25, 2016, at 11:22 PM, Jesse DuPont  
> wrote:
> 
> I don't know what all vendors use for their defaults, but one of the simplest 
> things you can do to help stabilize RSTP is to be intentional about which 
> device is the root bridge (i.e. closest switch/bridge to traffic 
> egress/ingress or core or what ever you define as the "root"). Make sure that 
> device has the lowest priority so that all other bridges are working to have 
> the best path to that device. if you want traffic in a bridged network to 
> flow a specific way, you can influence that with RSTP path cost. None of this 
> replaces routing (ducks), but it's better than leaving them all at default 
> priority and letting them elect the root.
> 
> Jesse DuPont
> Network Architect
> email: jesse.dup...@celeritycorp.net 
> Celerity Networks LLC
> Celerity Broadband LLC
> Like us! facebook.com / 
> celeritynetworksllc 
> 
> Like us! facebook.com 
> /celeritybroadband 
> 
> 
> On 10/25/16 9:08 PM, That One Guy /sarcasm wrote:
>> STP is dynamic, when we were solely a Milan switched network it was off by 
>> default, when we moved to HP it was off by default, now its on by default 
>> for everything even HP, I assume if I read release notes on new models I 
>> would have known that.
>> I prefer dynamic things be off and we turn them on
>> loop mitigation i guess i can see being worthwhile to have on by default, 
>> but generally when you create a loop its by mistake, and if you dont know 
>> its on, it creates a dynamic troubleshooting environment
>> I like knowing its generally going to be on, the mikrotik thing was 
>> unexpected, global off would be nice
>> when i was just a tech they implemented rstp on the network for redundant 
>> crummy links and dicked with the timers the links would flap and kick off a 
>> random root run around the network for hours, you cant get greedy with stp
>> 
>> 
>> On Tue, Oct 25, 2016 at 9:23 PM, Faisal Imtiaz > > wrote:
>> very simple..
>> 
>> a) it is on by default in CISCO Equipment
>> b) it catches people by surprise, cause it does not output to logs as to 
>> what is going on.
>> c) one can only see the status of a blocked port if one issues the stp 
>> command.
>> but the biggest one is...
>> 
>> d)  Cisco switches will react to STP messages being broadcast from a totally 
>> different section of a connected network... !
>> 
>> so, simply by the virtue of 'Cisco of Right' and they have the highest 
>> quantity of switches deployed in the field... and this is one of the biggest 
>> gotchas, when the folks experience it. they  (cisco) automatically gets  
>> the brunt of the black eye !
>> 
>> No one starts a story with .. " Let me tell you how I made a mistake" 
>> but the story always starts with .." Ahhh what a POS, let me tell you what I 
>> had to endure due to this  POS !  ".
>> 
>> :)
>> 
>> 
>> Faisal Imtiaz
>> Snappy Internet & Telecom
>> 7266 SW 48 Street
>> Miami, FL 33155
>> Tel: 305 663 5518 x 232 
>> 
>> Help-desk: (305)663-5518  Option 2 or Email: 
>> supp...@snappytelecom.net 
>> 
>> From: "Ken Hohhof" >
>> To: af@afmug.com 
>> Sent: Tuesday, October 25, 2016 6:18:21 PM
>> Subject: Re: [AFMUG] stp
>> Why do people act like STP is some evil proprietary Cisco thing?  If you 
>> don’t want it, turn if off, but don’t make it sound like Cisco pulled it out 
>> of their ass, it’s a perfectly standard Layer 2 protocol.
>>  
>> CDP you can yell at Cisco (although Mikrotik supports it).  Certain VLAN 
>> error messages you can yell at Cisco.
>>  
>>  <> 
>> From: Af [mailto:af-boun...@afmug.com ] On 
>> Behalf Of Carlos Alcantar
>> Sent: Tuesday, October 25, 2016 4:58 PM
>> To: af@afmug.com 
>> Subject: Re: [AFMUG] stp
>>  
>> on the access side where you might have issues with stp is if your customer 
>> is then connecting to a cisco switch these stp messages between equipment 
>> can get funny and start shutting off ports because one side supports it and 
>> the other doesn't.  you'll go crazy locating these issues.
>> 
>>  
>>  
>>  
>> Carlos Alcantar
>> 
>> Race Communications / Race Team Member 
>> 
>> 1325 Howard Ave. #604, Burlingame, CA. 94010
>> 
>> Phone: +1 415 376 3314  / car...@race.com 
>>  / http://www.race.com 
>>  
>> From: Af 

Re: [AFMUG] stp

[Paul Stewart]

Mainly because most people don’t understand STP so they leave it on and when 
something happens they are puzzled …. or they’ve read somewhere that it’s evil 
and turn it off, experience a “melt down” and think you have to have it enabled.

STP is a very important and useful protocol in layer2 networks …. it’s like 
other protocols - you should understand them and how they work 

CDP is same way .. LDP/LLDP another …. 

My personal favourite is folks who don’t set a root bridge in STP, especially 
corporate environments :)


> On Oct 25, 2016, at 6:18 PM, Ken Hohhof  wrote:
> 
> Why do people act like STP is some evil proprietary Cisco thing?  If you 
> don’t want it, turn if off, but don’t make it sound like Cisco pulled it out 
> of their ass, it’s a perfectly standard Layer 2 protocol.
>  
> CDP you can yell at Cisco (although Mikrotik supports it).  Certain VLAN 
> error messages you can yell at Cisco.
>  
>   <>
> From: Af [mailto:af-boun...@afmug.com ] On 
> Behalf Of Carlos Alcantar
> Sent: Tuesday, October 25, 2016 4:58 PM
> To: af@afmug.com 
> Subject: Re: [AFMUG] stp
>  
> on the access side where you might have issues with stp is if your customer 
> is then connecting to a cisco switch these stp messages between equipment can 
> get funny and start shutting off ports because one side supports it and the 
> other doesn't.  you'll go crazy locating these issues.
>  
>  
>  
> Carlos Alcantar
> Race Communications / Race Team Member 
> 1325 Howard Ave. #604, Burlingame, CA. 94010
> Phone: +1 415 376 3314 / car...@race.com  / 
> http://www.race.com 
>  
> From: Af > on behalf of 
> Josh Reynolds >
> Sent: Tuesday, October 25, 2016 2:38:04 PM
> To: af@afmug.com 
> Subject: Re: [AFMUG] stp
>  
> You veered way off into something I wasn't even talking about.
> There's nothing wrong with a loop protect on an access port, but since its 
> not an official standard, there will be variances in loop detection algorithm 
> quality and design between vendors. YMMV.
> That said, there's nothing wrong with STP on access ports either.
>  
> On Oct 25, 2016 4:27 PM, "George Skorup"  > wrote:
>> Care to explain? What's wrong with simple loop-protect on an edge port 
>> facing a dumb customer?
>> 
>> On 10/25/2016 3:07 PM, Josh Reynolds wrote:
>>> *facepalm*
>>>  
>>> On Oct 25, 2016 3:06 PM, "George Skorup" >> > wrote:
 Lots of switch vendors and even MikroTik (in the 6.37 branch, IIRC) 
 support loop protection. If all you care about is stopping a loop, then 
 use that. Use STP if you need its functionality.
 
 On 10/25/2016 1:16 PM, Josh Reynolds wrote:
> If one of the ports has a loop, it will block only that port. Obviously, 
> disable STP on uplinks.
>  
> On Oct 25, 2016 1:08 PM, "Chuck McCown"  > wrote:
>> But if you only have one upstream connection and you fall, it isn’t 
>> going to do anything, right?
>>  
>> From: Josh Reynolds 
>> Sent: Tuesday, October 25, 2016 12:01 PM
>> To: af@afmug.com 
>> Subject: Re: [AFMUG] stp
>>  
>> STP is a safety net. Its not doing much unless you fall.
>>  
>> On Oct 25, 2016 12:44 PM, "Chuck McCown" > > wrote:
>>> How can STP being enabled help anything if you are not using it?
>>>  
>>> From: Josh Luthman 
>>> Sent: Tuesday, October 25, 2016 10:10 AM
>>> To: af@afmug.com 
>>> Subject: Re: [AFMUG] stp
>>>  
>>> I just learned the other day thanks to Steve the Mikrotik software 
>>> bridges are (R?) STP by default.
>>>  
>>>  
>>> Josh Luthman
>>> Office: 937-552-2340 
>>> Direct: 937-552-2343 
>>> 1100 Wayne St
>>> Suite 1337
>>> Troy, OH 45373
>>>  
>>> On Tue, Oct 25, 2016 at 12:07 PM, Bill Prince >> > wrote:
 Actually Netonix (one t, no r). I don't know that I would leave it on, 
 but I don't know how you're using it.
> https://www.netonix.com/wisp-switch.html 
> bp
 
  
 On 10/25/2016 9:04 AM, Josh Reynolds wrote:
> Nettonix.
> You could always leave it on... If your access network is layer2 up 
> to that switch, it could help.
>  
> On Oct 25, 2016 11:03 AM, "CBB - Jay Fuller" 
> > wrote:
>>  
>> just discovered on one of our nettronix switches spanning 

Re: [AFMUG] stp

[Paul Stewart]

It’s pretty typical in the industry to have STP (in one form or another) 
enabled by default for loop prevention which is one of STP’s primary uses … 
blocking or forwarding


> On Oct 25, 2016, at 12:18 PM, That One Guy /sarcasm 
>  wrote:
> 
> Why is there a trend toward this being enabled by default? was there an RFC 
> or something, or is it a just cause they can type of thing
> 
> On Tue, Oct 25, 2016 at 11:10 AM, Josh Luthman  > wrote:
> I just learned the other day thanks to Steve the Mikrotik software bridges 
> are (R?) STP by default.
> 
> 
> Josh Luthman
> Office: 937-552-2340 
> Direct: 937-552-2343 
> 1100 Wayne St
> Suite 1337
> Troy, OH 45373
> 
> On Tue, Oct 25, 2016 at 12:07 PM, Bill Prince  > wrote:
> Actually Netonix (one t, no r). I don't know that I would leave it on, but I 
> don't know how you're using it.
> 
> https://www.netonix.com/wisp-switch.html 
> 
> bp
> 
> 
> On 10/25/2016 9:04 AM, Josh Reynolds wrote:
>> Nettonix.
>> 
>> You could always leave it on... If your access network is layer2 up to that 
>> switch, it could help.
>> 
>> 
>> On Oct 25, 2016 11:03 AM, "CBB - Jay Fuller" > > wrote:
>>  
>> just discovered on one of our nettronix switches spanning tree protocol was 
>> enabled.
>> we've run this switch probably four months - no real side effects - but i 
>> don't run stp anywhere
>> else.  any reason to leave this on?
>>  
>> thanks
>>  
>>  
> 
> 
> 
> 
> 
> -- 
> If you only see yourself as part of the team but you don't see your team as 
> part of yourself you have already failed as part of the team.

Re: [AFMUG] Google Fiber is no more

[Rory Conaway]

Absolutely they cherry picked.  Then they went into MDU’s for pennies and lost 
their shirts.

Rory

From: Af [mailto:af-boun...@afmug.com] On Behalf Of Josh Reynolds
Sent: Tuesday, October 25, 2016 9:34 PM
To: af@afmug.com
Subject: Re: [AFMUG] Google Fiber is no more


I'd love to see their overall take rates. I have heard numbers of 75-85% in 
more affluent areas. They cherry picked neighborhoods for sure though.

On Oct 25, 2016 10:15 PM, "Rory Conaway" 
> wrote:
Big surprise there.  They built it and no one came.

Rory

From: Af [mailto:af-boun...@afmug.com] On Behalf 
Of Tushar Patel
Sent: Tuesday, October 25, 2016 7:14 PM
To: af@afmug.com
Subject: Re: [AFMUG] Google Fiber is no more

Their contractor are still hiring installer in Austin.

Need to probably understand why those cities not others?

Tushar


On Oct 25, 2016, at 9:06 PM, Josh Reynolds 
> wrote:

New ones. They're still deploying existing networks. They just opened up a few 
new areas in Kansas City recently.

On Oct 25, 2016 9:03 PM, "Jaime Solorza" 
> wrote:

Moving folks to wireless Aye Dios

On Oct 25, 2016 7:56 PM, "Gino Villarini" 
> wrote:
https://gizmodo.com/google-fiber-halts-operations-in-ten-cities-1788214992?rev=1477443092657_campaign=socialflow_gizmodo_facebook_source=gizmodo_facebook_medium=socialflow