Thank G-d, the NT admin came in.  He explained the diagrams to me.  The
people across the street, the people in Culver City, and the people in TJ
go through the 3600 to get to the 6509.  (And in TJ they have a whole bunch
more switches etc which I don't care about right now.)  The people in this
building go through various 3548's to get to the 6509.  There is also a
3548 between some of the servers (not mine) and the 6509.  Both my SQL
Server servers plug directly into the 6509. 

Like this except there should be more than 4 3548's on the right:

                     other servers
                           |
TJ ---                   3548             /- 3548 -- L.A. 2nd floor
       \                   |              /- 3548 -- L.A. 3rd floor
CC ---    3600 --------- 6509 -----------
       /                   |              \- 3548 -- L.A. 4th floor 
BH ---                impt servers        \- 3548 -- L.A. 5th floor



We now have the test running as follows:

1.  between my servers #1 and #2 which both plug directly into the 6509

2.  between another server that goes through the separate 3548 to get to
the 6509 and server #2

3.  between a desktop that goes through a 3548 on the 3rd floor and server #2

4.  between a desktop on the 4th floor that we think goes through a
different 3548 than our area (which is also on the 4th floor) and the 6509 

plus the tests from my desktop to server #1, mydesktop to server #2, and
the other desktop to server #1, all of which would be going through the
same 3548, then through the 6509, to get to whichever server they're
talking to.

So hopefully this should isolate which thing is having the problem.  I hope
it shows up in the next couple of hours so we at least have a shot at a
replacement before tomorrow.  Today, just to be contrary, there haven't
been any dropped packets since 8:44 this morning.

Well, whatever happens, happens.  At least now we're finally attacking it.

I can't thank you enough for all your help.


> From: "Geo." 
>It's hard to tell what could be affecting what from the description but my bet
> is still on the switch being the problem.
> I'm guessing from the entries below about fiber gigabit that there is more
> hardware than just a switch so the network guy is really going to have to
> figure out how to test to the different points to see where the network is
> failing.
> One thing you could ask him to do is see if it's possible (it may not be) to
>put your desktop and the server in question on the same network segment so you
> bypass the switch, then run your test and I bet you find no drops.
> Geo.
> "Ellen K"  wrote in message
> news:b09718.c66801{at}harborwebs.com...
>> I couldn't figure out the diagrams at ServerGuy's desk and neither he nor
>>NewlyMintedNetworkGuy are answering their cellphones.  However, on looking at
>>the results of the monitoring tool posted on the intranet (which only shows a
>>3600 between the outlying folks and the 6509) I see that the entire 5th floor
>> in TJ is down, so I guess they have their hands full with that.
>> 
>> Could the 6509 itself be bad?
>> 
>> On the intranet thing I also see a bunch of warnings and alerts but I don't
>>know what most of it means.  Like, there will be an entry
"Ellen's Gateway is
>> warning" and then the next one is "Ellen's Gateway is up".
>> 
>> The latest two entries are:
>> 9:51
>> Reset: Percent Utilization of LA-6509-Port 3/1 ù short wave fiber gigabit
>> ethernet has returned to normal. Current traffic load of this interface is
>> Received : 210 Kbps Transmitted : 4865 bps
>> 
>> 9:53
>> Alert: Transmit Percent Utilization of LA-6509-Port 3/1 ù short wave fiber
>>gigabit ethernet is 4 %. Current traffic load of this interface is Received :
>> 299 bps Transmitted : 44 Mbps
>> 
>> ???
>> 
>> > From: "Geo." 
>> > "Ellen K."  wrote in message
>> > news:kub6c0lqrnmqbr0cfim0ech6sdcl4c7q2b{at}4ax.com...
>> >> I think there is one thing (can't remember exactly, is there such a
>> >> thing as a 3600?) between my desktop and the big switch. 
I'm leaving to
>> >> work now, will try to see if I can figure out the diagram
at ServerGuy's
>> >> desk after I get in.
>> > Yes, there is a cisco 3600 series router
>> >http://www.cisco.com/warp/public/cc/pd/rt/3600/ but imo that's
probably not
> th
>> > cause unless it has a bad port or something, I think that's
possible but
>> > unlikely. Easy enough to check the uptime and see if it's rebooting or
>> > something. In fact that might be a good check for the switch
as well, have
>> >network guy check uptime on both devices, if one appears to
have rebooted
> abou
>> > the time of the drops then you found the problem.
>>> I'd still suspect the switch, routers are usually very
dependable, it's the
>> >switches that nobody ever pays attention to setup that end up causing
> problems
>> > like this. (that's why I hate switches)
>> > Geo.

--- BBBS/NT v4.01 Flag-5