[175842] in North American Network Operators' Group
Re: Tail-F
daemon@ATHENA.MIT.EDU (Peter teStrake)
Tue Nov 4 13:24:00 2014
X-Original-To: nanog@nanog.org
From: Peter teStrake <Peter.teStrake@tradingscreen.com>
To: Colton Conor <colton.conor@gmail.com>
Date: Tue, 4 Nov 2014 17:55:28 +0000
In-Reply-To: <CAMDdSzPSHS_Wuvgdbwhv=zrhZX4Yzgc4A+fcXdZOxtwEYzy93A@mail.gmail.com>
Cc: NANOG <nanog@nanog.org>
Errors-To: nanog-bounces@nanog.org
Conor,
Tail-f will give you a global view of your network configurations, and will=
keep the local database in sync with the devices. This gives you the abi=
lity to search and update config across devices.
If you want to see the live status, then you can compile an snmp mib and at=
tach that to the device, so your team can see both config and status via a =
single CLI.
Really though, it's about service provisioning and orchestration across mul=
tiple vendors, but there can be a fair amount of work there depending on wh=
at you are trying to achieve.
There is a also a REST interface that makes it pretty simple to access anyt=
hing in the database and present it in a web page.
Cheers
Pete
Sent from my iPhone
> On 2 Nov 2014, at 23:58, Colton Conor <colton.conor@gmail.com> wrote:
>=20
> I am aware that you can see if a port is up or down through SNMP. I guess
> that was a bad example. I want to see the entire output of a show interfa=
ce
> command. For example, we have multiple types of access networks (GPON, DS=
L,
> Cisco ethernet switches). Some of the show interface commands are fairly
> basic, but others like on a DSL port show much more information like sync
> rate, signal loss, etc. The only way to get this information on some
> platforms is to run the show interface command for CLI I believe, or logi=
n
> to the access platform's GUI interface. Both of these task aren't so easy
> when you are dealing with multiple access platforms.
>=20
>=20
>> On Sun, Nov 2, 2014 at 3:15 PM, Phil Bedard <bedard.phil@gmail.com> wrot=
e:
>>=20
>> Tail-F's ConfD can operate as a front-end CLI and do the things he wants
>> it to do in an operational sense but I would agree it may not be the
>> easiest to use tool for simply monitoring and grabbing interface
>> state/statistics. It's fairly flexible and can do a lot of abstracted
>> things through its ConfD element but there is some backend work to make =
it
>> happen. Not as much as doing it from scratch but still a bit of work.
>> It can abstract different device CLIs so they all look the same and use =
the
>> same commands and you can extend the CLI to do custom things as well if =
you
>> want.
>>=20
>> The whole system is fairly powerful and very extensibe. There are
>> monitoring elements built into it and It could be a full blown monitorin=
g
>> system, really just depends on the scale you need and how much work you
>> want to put into it.
>>=20
>> Phil
>>=20
>>=20
>>=20
>>> On Sun, Nov 2, 2014 at 1:19 PM, Jimmy Hess <mysidia@gmail.com> wrote:
>>>=20
>>> On Sun, Nov 2, 2014 at 11:04 AM, Colton Conor <colton.conor@gmail.com>
>>> wrote:
>>>> Is anyone using Tail-f software or know anything similar? We are looki=
ng
>>>> for a solution that is vendor agnostic. Can do simple command like sho=
w
>>>=20
>>> I've only read of this, but my understanding is the Tail-F product is
>>> for configuration management and supporting provisioning automations
>>> anyways, monitoring configs sure. As far as I know they cannot
>>> monitor or show network operational status, so your use case may not
>>> overlap with their capabilities, and perhaps, what you are likely
>>> suggesting is something that unfortunately doesn't exist yet: a tool
>>> for both configuring and observing a detailed operational state of the
>>> network devices in a vendor-agnostic way.
>>>=20
>>> However, for simple bandwidth statistics and port Up/Down; for most
>>> devices, this information is available through SNMP based management
>>> tools.
>>>=20
>>> Basic Up/Down and statistics could generally be gathered by any good
>>> SNMP-based NMS / network monitoring product, there are thousands of
>>> these, or OSS such as Cacti, Zenoss, and proprietary ones such as HP
>>> OpenView, SolarWinds, InterMapper, Whatsup; also, just about every
>>> major network device vendor has their element management system.
>>>=20
>>> Various NMS can also be configured to run some selected code or offer
>>> up a GUI command for running a snmpwalk against the ifOperStatus or
>>> ifIn/Out Octets.
>>>=20
>>>=20
>>>=20
>>>=20
>>>> interface so even non-network techs and CSR's can get basic is the por=
t
>>> up
>>>> or down type stats without having to directly login to the network.
>>> --
>>> -JH
>>=20
>>=20
