They should get the troubleshooting tools first, in my opinion. There are 
free protocol analyzers available. The problem may be some network-hog 
application that doesn't belong on the network, such as Napster or 
something. The problem could continue to lurk even if they did get a switch.

Of course, protocol analysis can be very time-consuming, and time is money, 
so perhaps throwing a switch in the network might be a good approach 
too.  Mwave.com is advertising a D-Link 8-port 10/100 switch for only 
$69.99, while supplies last. They also have a 3-Com 4-port switch for only 
$92 and really cheap LinkSys switches. Those are all good name brands. (No, 
I don't work for them! ;-)

Priscilla

At 09:05 PM 4/17/01, Jason J. Roysdon wrote:
>Convince them to get troubleshooting tools when they don't even have
>switches?  *chuckles*  Good luck.
>
>--
>Jason Roysdon, CCNP+Security/CCDP, MCSE, CNA, Network+, A+
>List email: [EMAIL PROTECTED]
>Homepage: http://jason.artoo.net/
>
>
>
>""Chuck Larrieu""  wrote in message
>[EMAIL PROTECTED]">news:[EMAIL PROTECTED]...
> > Hhmmmmm......
> >
> > Are these hubs daisy chained?
> >
> > Does the noticeable slowdown happen al the time, or can you isolate it to
> > particular times of day?
> >
> > Do you have an internet connection?
> >
> > Do you have anyone using any kind of dial up to an external service of
>some
> > kind?
> >
> > Have people set up their own little Windows networking networks, in
>addition
> > to your network - file and print sharing stuff?
> >
> > Are people having to print a lot of things they weren't doing before?
> >
> > Story time:
> >
> > Back at the brokerage firm, there was an occasion where my help desk
>started
> > getting calls about the network being down.  In general, this kind of
> > complaint could be attributed to not being logged on to the network, and
> > usually we would blow off the callers with the instruction to log on.
>Well,
> > upon thinking about the fact that people who were complaining were in
many
> > cases "good" users, and the fact that there were so many calls that
>morning,
> > I traced back one of the end user stations to a particular hub ( we had
>hubs
> > plugged into switches at the time ) and I was shocked to see the
collision
> > light solid red. I was able to use the HP stack manager software to
>discover
> > that a particular port was just saturating the hub with traffic. Tracking
> > down that user, I learned that particular person was connected to a
> > particular internet based service ( some kind of research database ) and
>was
> > downloading and updating a complex database file using a particular
> > proprietary piece of software. The damn thing practically seized the
>entire
> > bandwidth of that hub, and so monopolized the traffic that other folks
>were
> > losing their connections to the Novell servers, I am guessing because of
> > lack of keepalives.
> >
> > Once the problem was identified, I gave this particular user a dedicated
> > switch port, and life was good after that.
> >
> > My point being that even though you have a very few users, all it takes
is
> > one bandwidth piggy, and your shared collision domain network is toast.
> > Might want to convince the boss that investment in a Fluke meter or some
> > kind of management software is a good thing.
> >
> > Chuck
> >
> > -----Original Message-----
> > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] On Behalf Of
>John
> > Brandis
> > Sent: Tuesday, April 17, 2001 5:09 PM
> > To: [EMAIL PROTECTED]
> > Subject: Network Collisions [7:1006]
> >
> > G'day all where ever you may be.
> >
> > I have been watching my network here in my office and I have noticed that
> > over
> > the last week, that the network is slowing down. Due to financial
> > constraints,
> > we are using 10/100 16 port hubs (2) {just thought I would point that
out}
>I
> > have noticed that the collision LED's are on a fair bit these days. I
> > checked
> > to see if the errors where due to cable problems or broken ports on the
>hub,
> > but this was not the case. I made sure all the PC's were using the same
> > protocol and still I have an abnormal amount of collisions. I understand
> > that
> > I will have collisons but for a 11 user network that is centerd around a
> > WIN2k
> > Server/Exchange server I have about a 40% collision rate.
> > Does any one have any idea's (besides the obvious of buying a switch) on
>how
> > I
> > can troubleshoot this or fix the problem.......
> >
> > Thanks gang
> >
> > John Brandis
> > Network Engineer
> > GoWireless Communications
> > 155 George Street Sydney
> > +61 2 9251 5000
> > FAQ, list archives, and subscription info:
> > http://www.groupstudy.com/list/cisco.html
> > Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
> > FAQ, list archives, and subscription info:
>http://www.groupstudy.com/list/cisco.html
> > Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
>FAQ, list archives, and subscription info: 
>http://www.groupstudy.com/list/cisco.html
>Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]


________________________

Priscilla Oppenheimer
http://www.priscilla.com




Message Posted at:
http://www.groupstudy.com/form/read.php?f=7&i=1030&t=1006
--------------------------------------------------
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]

Reply via email to