Solved

Comcast vs T1...

Posted on 2011-03-09
8
1,224 Views
Last Modified: 2013-12-23
I have been called in recently to maintain an existing office network with an older rack mount Win 2003 server environment. It is a DC which also is a DHCP server as well. Their T1 service has been with Cbeyond and they have switched to Comcast Business. They have 2 3Com access points in the building for their wireless. I have installed a Netgear wireless"N" access point now. Also in the mix is a Sonos audio/video stream network. They run one SQL database. There are 18 workstations connected to the domain. A total of 55 devices connected via DHCP including RAS. Printers of course too! I'm wanting to simplify this to a "file server" environment. I'm suggesting to dismiss the DC server, DHCP and RAS to allow the Comcast modem to hand out DHCP. Is the Comcast robust enough for this application? It was a much busier place back in 2000 when it was built. I'd even prefer to switch to a Win 2008 server on a multi-core processor PC vs. the rack mount device. It's just been complicated with the Comcast not being DHCP and DNS issues. I appreciate your thoughts...

PCWoes in Denver

My experience level is self taught for the record :)

Their existing server info is:
Dell Computer Corporation PowerEdge SC1425
Bus Clock: 800 megahertz
3.00 gigahertz Intel Xeon (2 installed)
16 kilobyte primary memory cache
2048 kilobyte secondary memory cache
64-bit ready
Hyper-threaded (4 total)
0
Comment
Question by:PCWoes
8 Comments
 
LVL 95

Assisted Solution

by:Lee W, MVP
Lee W, MVP earned 500 total points
Comment Utility
A domain is not complicated.  A workgroup without centralized administration for a group of 18 workstations would be a nightmare.  Domains are simple and the adjustments needed to your internet connection should be equally simple.

1.  DNS is VITAL - CRITICAL - to a properly functioning domain.  The DC(s) should be the ONLY DNS server configured in the network settings - no third party DNS should be listed - Yes, this means if the DC is down, people can't browse the internet, but the way AD works and the fact that primary and secondary DNS does NOT mean primary is ALWAYS used means you cannot have non-AD aware DNS servers referenced.
2. DHCP can be run by anything (though for SIMPLICITY I would run it from the server - it's a lot easier to manage there).  The only requirement is that the DHCP server hand out ONLY the AD DNS server as the DNS server for clients.
0
 

Author Comment

by:PCWoes
Comment Utility
What about the degradation of Comcast broadband split with all the DHCP clients vs. the T1 circuit? Would it be noticeable you think?

My apologizes as I think I was asking even more than 1 question in all reality for this network !!

The modem has LAN ports and I hooked up the "N" access point direct to it. Should it be hooked into the Cisco switch as well where the DC feeds the DHCP network? It has been assigned a static IP on the existing subnet. I had set the forwarding DNS on the DC to the primary and secondary of the WAN provided by Comcast. That went wrong and the Sonos IT remoted in and set the forwarding DNS to the IP of of the DC. The streams are back to speed now. We could access the internet using the IP of the Comcast modem less the quality audio streams.

Settings FYI:

Comcast modem: 192.168.0.1
DC: 192.168.0.200
DHCP: 25-99
Forwarding DNS: 192.168.0.200 (now)
Workstation IP's: Auto IP
Workstation DNS: Primary 192.168.0.200 Secondary: 192.168.0.1
Wireless "N" AP: 192.168.0.250 (direct to modem) DNS: Comcast provided primary and secondary

0
 
LVL 95

Assisted Solution

by:Lee W, MVP
Lee W, MVP earned 500 total points
Comment Utility
Comcast DNS should NEVER be used for anything.  (Well, you could use them as forwarders, but I would sooner use OpenDNS for forwarders and I don't typically use forwarders.  Some people argue that forwarders are faster... the 10-50 ms (.01-.05 seconds) faster should not be noticeable to any human.

I'm not familiar with Comcast cable in general (worked on a client using them once a couple years back) but the concepts are all the same...

> What about the degradation of Comcast broadband split with all the
> DHCP clients vs. the T1 circuit? Would it be noticeable you think?

Comcast cable (unless you have some extremely cheap service that I would never recommend for a business service and don't even know if Comcast would offer) is going to be at least 6x faster on the download than a T1.  Upload could be anywhere from 1/6 as fast to 3x+ as fast depending on your level of service.  But unless you're using VoIP and/or serving large files to the internet (transferring large files), then your performance with cable should be NOTICEABLY faster than with the T1 regardless of how you have it configured.

If the modem has LAN ports then it has LAN ports.  They are just like any other LAN ports.  The only thing is the more devices you have the more risk of failure of one device.  But performance wise, no one should notice anything slower with Comcast (vs. T1) unless you have something misconfigured in a managed switch or something.  Or perhaps you plugged your server into a 100 Mbit port while a workstation is on another switch that's capable of 1000 Mbit (1 Gb).

For example, I would (pending confirmation that it has gigabit ports) plug the server into the Wireless "N" router as N speeds are faster than 100 Mbit networks and often (but not always) an N router should have gigabit ports.  By putting the server on the gigabit port, in theory, you could have 10 clients at 100 Mbit access the server at full speed.

NOTE: various things can affect network performance, including quality of the equipment, the standards the equipment operates at, the software used to interact with that equipment (drivers), the type of data being accessed, where that data is, how many other people are using the network at that moment, and other factors.  So I wouldn't be shocked if you see only 75% of the reported connection speeds... potentially even less.  
0
 
LVL 32

Expert Comment

by:aleghart
Comment Utility
Cable internet will have more bandwidth available than a T1.  Same goes for downstream ADSL, which I installed here to take the load off our T1 lines.

I would not leave DHCP & DNS to any cheap device provided by a cable company, unless it's a very small residential/SOHO installation with no servers/services.

Leave it all with the DC, as you'll have far more control over it.
0
PRTG Network Monitor: Intuitive Network Monitoring

Network Monitoring is essential to ensure that computer systems and network devices are running. Use PRTG to monitor LANs, servers, websites, applications and devices, bandwidth, virtual environments, remote systems, IoT, and many more. PRTG is easy to set up & use.

 

Accepted Solution

by:
PCWoes earned 0 total points
Comment Utility
I'm humbled by the immediate and intelligent replies I'm receiving. I can clearly see my simplistic mind hasn't given credit to the DC environment, I've been so exposed to Workgroup networks I'm obviously very amateurish to the Server/DC capabilities. I concede to the fact I'm wrong in this instance and will admit so to my client as well. I'm absorbing this and it all makes complete sense. I think even the existing equipment is quite capable (i.e. Dell PowerEdge SC1425 ) for their needs. I'll move the server NIC connection over to the "N" AP connected to the modem just to see if there is indeed improvement.

This is a fantastic tool for the likes of myself !! In this case I was over my head with the true value of a DC.

My summary then is as follows:

1. Comcast Business modem: 192.168.0.1...No DHCP connected to Cisco switch
2. Comcast modem LAN 1 to Wireless "N" AP assigned IP: 192.168.0.250...DNS: Auto
3. Server IP: 192.168.0.200...DHCP: 25-99  Forward DNS: 192.168.0.200 connected to Cisco switch for DHCP clients
4. Workstation(s) TCP/IP: Auto  DNS1: 192.168.0.200
0
 
LVL 13

Expert Comment

by:kdearing
Comment Utility
Just one more thing, just to be clear...
All devices should point to the server for DNS (even the server's own network cards).
DNS Forwarders (in the DNS server configuration) should point to either the Comcast Busniess modem (192.168.0.1) or Comcast's DNS servers.
0
 
LVL 95

Expert Comment

by:Lee W, MVP
Comment Utility
Some final thoughts/explanation:

Active Directory uses DNS to locate resources - like the server that will authenticate it.  When the server starts up, it registers itself with its DNS server.  If your DNS server accepts the registration, it will now know where your resources are.  By default, a Windows DNS Server in its own domain will accept the registration information.  Comcast's DNS servers will NOT accept your server's registration.  So if your workstations or server point to comcast for DNS, Comcast responds with a (in simplistic terms) "sorry, don't know where that is" and your clients are stuck.  That's why it's vitally important to use the DNS server on Windows.  And even though the terminology suggests that you can have multiple DNS servers and only one will be used unless it's not available, any number of reasons could result in the "secondary" DNS servers being used at any time, so even listing Comcast's servers as backup isn't good... doing so often leads to sporadic slowdowns logging in and access network resources.

I don't put in workgroups - sharing between the machines, I find, is too unreliable.  So I only put in domains.  With a properly configured DNS, there really should be no significant/likely downside to using a domain and you gain the ability to have a single logon for all users and, if you want to ease your management, to control the systems through the implementation of Group Policies.

If you are not well versed in AD, then I strongly suggest you download some trial versions of server and setup a small test network to play around and learn on.  Using virtualization, trial versions, and a single, reasonably powerful (read: LOTS OF RAM) computer and you can setup a test network of 5-10 computers (or more) with servers to test and understand the implications of everything without putting this (or any other) production network at risk.
0
 

Author Closing Comment

by:PCWoes
Comment Utility
I summarized my network process with the answers provided me !
0

Featured Post

What Is Threat Intelligence?

Threat intelligence is often discussed, but rarely understood. Starting with a precise definition, along with clear business goals, is essential.

Join & Write a Comment

Hello to you all, I hear of many people congratulate AWS (Amazon Web Services) on how easy it is to spin up and create new EC2 (Elastic Compute Cloud) instances, but then fail and struggle to connect to them using simple tools such as SSH (Secure…
AWS has developed and created its highly available global infrastructure allowing users to deploy and manage their estates all across the world through the use of the following geographical components   RegionsAvailability ZonesEdge Locations  Wh…
After creating this article (http://www.experts-exchange.com/articles/23699/Setup-Mikrotik-routers-with-OSPF.html), I decided to make a video (no audio) to show you how to configure the routers and run some trace routes and pings between the 7 sites…
This video gives you a great overview about bandwidth monitoring with SNMP and WMI with our network monitoring solution PRTG Network Monitor (https://www.paessler.com/prtg). If you're looking for how to monitor bandwidth using netflow or packet s…

771 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question

Need Help in Real-Time?

Connect with top rated Experts

10 Experts available now in Live!

Get 1:1 Help Now