Solved

CSU/DSU and demarc question

Posted on 2004-10-25
2,615 Views
Last Modified: 2008-03-03
Ok, I'm reading about Demarcs and CSUs/DSUs.  Apparently, CSUs are considered DSUs nowadays because most service providers leave them on the clients site?

Can someone provide an example of what a demarc would be?   What about a DSU?  I am in my server room at work and see our T1 line plug into something that looks like a cable modem box. Would that be the DSU.
thanks
0
Question by:dissolved
    11 Comments
     
    LVL 79

    Accepted Solution

    by:
    T1's generally terminate in what's called a "smart jack". This is typically a locked box on the wall with what appears to be a modem of some sort, with a RJ-45 plug, and from there, a cable (CAT5 or flat ribbon) goes to the "Network" port on your CSU/DSU.
    The SmartJack is typically the "point of demarcation" or "demarc" that divides the equipment owned and maintained by the telco (hence the lock on the box), and your equipment.
    If you have to extend the T1 connection to another room or another floor where the CSU/DSU resides, this would be referred to as a "demarc extension" and is not maintained by the telco.
    Yes, pretty much DSU and CSU are used interchangably, but really are two separate functions, usually performed within the same box. To convert T1 channels into a Data stream, you need both the Channel Service Unit and the Data Service Unit functions.

    The CSU/DSU can be external, meaning an independent box that sits between the demark and connects to a router with a serial cable connection, or it can be internal to the router and all you can see is a single cable from the demarc to the router..
    This is an example of the current Adtran (market leader in external CSU/DSU) model
    http://www.adtran.com/adtranpx/Rooms/DisplayPages/LayoutInitial?ProductCategory=com.webridge.entity.Entity%5BOID%5BE9E4E419ED90AF46BD2E21B62FF44FAB%5D%5D&Product=com.webridge.entity.Entity%5BOID%5B3A4574BCC9A798499E9452423C7346DE%5D%5D&Container=com.webridge.entity.Entity%5BOID%5BF5C7CEE8D8313E49B4D65B30BDDF4734%5D%5D
    0
     
    LVL 11

    Assisted Solution

    by:PennGwyn
    If the T1 you see comes off of some kind of phone company gear (usually mounted on a plywood wall), then you're standing in the demarc -- it's the room where phone company equipment hands off to customer/provider equipment.

    Many modern routers include CSU/DSU functionality inside an interface that can accept a T1.  Older routers require an external CSU/DSU, and a serial connection from there to a serial port on the router.  So odds are that your "box" is either an external CSU/DSU, or a router with an internal CSU/DSU.  Are there markings or indicators or a brand name on the box?

    0
     

    Author Comment

    by:dissolved
    Thanks guys.  There is a box (looks like a data drop) that is labeled T1. It has a patch cable that plugs into it, the other side plugs into the CSU I'm guessing.
    We use a Cisco 4000 router, it looks old so I'm guessing it is old. Which means it doesnt have internal CSU/DSU functionality?  When CSU/DSUs are internal, how does it connect to demarc? Regular patch cable?

    here is a short diagram I made  http://mvpbaseball.cc/test.jpg

    Thanks
    0
     
    LVL 79

    Expert Comment

    by:lrmoore
    If the Serial cable from the 4000 to the external box is a heavy teal blue cable, then for sure the device is a CSU/DSU.
    I don't know of any T1 DSU modules for the 4000 series.
    If it did, it would be just a patch cable from the router to the T1 line.
    If it just looks like a data drop, labeled T1, then this is most likely a demarc extension. Somewhere else, you will find the telco's smartjack mounted on the wall..with a cable running from it to the back of the data drop..

    0
     

    Author Comment

    by:dissolved
    Got ya. The data drop is an extension of the demarc. I've never seen a physical T1 line. I'm guessing it's either copper or fiber?  How did they make it a data drop?  Wire it to a female 568a or b?
    Thanks!
    0
     
    LVL 79

    Expert Comment

    by:lrmoore
    Either. They typically wire 1,2 4 and 5 only, else a standard cat5 568B connection on the back of the data drop, to patch panel in room where smart jack is (actual demarc), and then a patch cable to SmartJack.. this patch cable is usually a flat silver ribbon 8-wire cable. Sometimes it's a rolled cable at that point.. (depends on if it is connecting to RJ-48X or RJ48C)
    Confused yet?

    0
     

    Author Comment

    by:dissolved
    "Confused yet?"

    Always :)

    Thanks for the help, greatly appreciated.
    0
     
    LVL 13

    Assisted Solution

    by:Dr-IP
    The demark is where your providers equipment ends, and yours starts. Normally on a T1, it’s at the smart jack, but when contracting for data services it can even be the router it’s self, making that your demark. This also means there can be more than one demark before your demark, in other words, the one to you from your data provider; and one for them from the telecom company providing them the local loop. So depending on your data services contract, your demark could be the smart jack, the CSU/DSU, or the routers, it all depends on how the contract was written.

    Because of this you should dig up the contract so you know where your providers line of responsibility stops, and yours starts. As all to often I see companies IT departments make a assumptions like the routers not their responsibility since there service provider provided it, when in reality it was sold to them via a monthly service charge for the first year or two, which is very common. Then later when it breaks they find out what they though was covered, wasn’t, and all hell breaks loose.

    As for external CSU/DSU’s, a lot of time they end up being the demark for data services, especially when using a single T1 for both phone and data services, and the CSU/DSU is being used to strip out channels for the router, and forward the remaining channels to the PBX. This is often called a fractional T1 in the industry, and is quite common.

    By the way, when it comes to external verses external CSU/DSU’s, a lot of time it just boils down to choice, external units frequently have more sophisticated features, and are more flexible, but internal ones make for neater installations, and less clutter. So unless you need a feature only available from an externally unit, which ever way you go really doesn’t matter.
    0
     
    LVL 5

    Assisted Solution

    by:AutoSponge
    On a 64k circuit (DS0) there is no CSU.  The telco equipment (OCU, NI, or total reach card) plugs into the customer DSU (which may be a router or a separate device).

    DSU operates on the DTE side and CSU operates on the DCE side.  Both my reside on the same box (think of the way they loop where a CSU loop is the full circuit and is in front of the DSU loop which will be the logical portion of the circuit).  CSUs and DSUs have their own test patterns.  

    In the states, CSU is almost always considered CPE (customer premise equipment) and you have to provide that yourself.  Overseas, the PTT provides the CSU.

    For a contract, the dmarc isn't the smart jack (THIS IS IMPORTANT TO KNOW).  The dmarc is actually the end of the wire coming off the smart jack.  This may be a cable that's patched into a 66M block or it may be what plugs directly into your CSU, either way if you ask telco to "dispatch and test from the dmarc" they should be testing that wire.  Anything beyond that is considered extended wiring and therefore an extended dmarc (or xdmarc).  Telcos will charge you for testing from an xdmarc unless they installed it.  My advise, get your LEC to install the xdmarc if it's not cost prohibitive as it can save you a dispatch (and time) during an outage.

    When chosing a CSU/DSU, go with a name you trust (i.e, you can get a vendor for it when needed).  Get something that has OOB (out of band--meaning it has it's own modem) capabilities.  Get something that has test capabilities as well as good performance monitors.  A robust T1 CSU should be able to generate QRSS (called quassi) and 2047 (a DDS pattern for the DSU portion).  It should be able to loop CSU or DSU to network and loop to DTE.  Other nice features will allow you to telnet to the device from the router or passthrough to the router without having to redial.

    IT'S NOT JUST CHOICE.  CSU's are a vital part of protecting your investment.  A CSU is built to withstand surges from the network as well as lightning strikes that surge the line (or at the very least, it's built to prevent that power from reaching the router).  These units are much cheaper than your high end routers and therefore if you have a router you spent over $1000 on, getting an external CSU is a no-brainer.  Even if you don't use it for test access, get it to act as a surge protector for the WAN line.
    0
     
    LVL 5

    Expert Comment

    by:AutoSponge
    BTW, yes that device between the dmarc and the router is a CSU.  Something to remember, if you have a point-to-point connection, you'll typically need to provide your own clocking on the ckt.  This means you need a GOOD CSU.  The better the CSU, the higher the stratum of clock.  When you plug into someone's frame network, they provide clocking and it's not an issue--again, it's not just "do you feel like using a second box".
    0
     

    Author Comment

    by:dissolved
    Great advice (as always) guys. Thanks
    0

    Write Comment

    Please enter a first name

    Please enter a last name

    We will never share this with anyone.

    Featured Post

    Anonabox PRO Tor & VPN Router

    PRO is the most advanced way to fortify your privacy and online anonymity by layering the Tor network with VPN services. Use both together or separately, and without needing to download software onto your devices.

    This article is a guide to configure bridging on Cisco Routers.  This is something I never knew was possible until after making a few phone calls to Cisco.  Using bridging saved our company money by not requiring us to purchase a new switch.  Bridgi…
    In the world of WAN, QoS is a pretty important topic for most, if not all, networks. Some WAN technologies have QoS mechanisms built in, but others, such as some L2 WAN's, don't have QoS control in the provider cloud.
    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…
    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…

    931 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

    16 Experts available now in Live!

    Get 1:1 Help Now