Losing network drive

guys,

we have recently move from into a new office location, and with it i
have migrated our core switch from the old cat 5500 to a 4506. becuase
of the distance from the datacenter to the user's area, we have build
a cross connect rack in between using 2950-48 switches, which are
connected back to the 4506 via fiber. problem is after the move,
users has been complaining that their network drive on the file and
print server has been droping, meaning they see a red "X" on the
network drive mapping on the exproler window. but, if you were to
consistantly click on it a couple of times, you will be able to get to
it. we have never had this problem with the 5500. but, before the
move, we were in a smaller area, hence no access switches was needed
then. the F&P server and users were all connected to the same 5500.
the distance from the server room to the user's area is around 90
meters. i am not seeing any crc or drop packets on the gbic interface
on both the 4506 as well as the 2950. since some of our legacy apps
actually utilize the map drive, it would hang, everytime the users
lose the network drive.

Thank you in advance.

LS

 

Is the switch hardcoded for duplex to the user workstations?
It sounds like it may be a duplex issue on the workstation ports, not the
fiber.

--
Steven BerkHolz
Send to Domain TESCOGroup dot com, username SB

Note: you may also want to know that you should never send mail to:




-abuse.org

 

the ports on the 2950s are all set to auto, and by default it picks up
100 full. but let me try hardcoding it to full and see what happens.
thanks.

LS

 

Use pathping on the pCs to find out if you're dropping packets along the
way.


--

hsb

"Somehow I imagined this experience would be more rewarding" Calvin
*************** USE ROT13 TO SEE MY EMAIL ADDRESS ****************
********************************************************************
Due to the volume of email that I receive, I may not not be able to
reply to emails sent to my account. Please post a followup instead.
********************************************************************

 

Did you try 'spanning-tree portfast' on the interfaces (set port host if
you are using CatOS on the 4506)?

Lothar

Lothar Hofmann HRZ / computer center Mail:
Universitaet Siegen
Hoelderlinstr. 3 Phone: +49 271 740 4760
D - 57068 Siegen Fax: +49 271 740 2523

 

See also http://www.cisco.com/warp/public/473/12.html (advice from
Chris Thomas from another thread)

Lothar

Lothar Hofmann HRZ / computer center Mail:
Universitaet Siegen
Hoelderlinstr. 3 Phone: +49 271 740 4760
D - 57068 Siegen Fax: +49 271 740 2523

 

I don't think the problem is with your switches. Take a look at this
Microsoft article
http://tinyurl.com/fmjc

 

PATHPING - fabulous.

I have been doing the same thing for years with
trace(rt)/(route) and ping. The manual process is clearly
cumbersome and prone to error.

I had even considered writing my own but I do so little
programming now that it would have been quite an effort.

Even just the asynchronous (well fast anyway) trace route makes
it worthwhile.

Best thing Microsoft since ...... Word/Excel. I do hear that the
new software development environment is pretty special too.

Thanks.

 

Autodisconnect is not enable on the server. so, what i have did so
far is turn on spantree port fast as well as hardcode full duplex to
the server port. its been running fine so far today..... (touch
wood)... hopefully it will stay that way.

Any way, thanks a million to all that help out. hope this group will
continue to stay informative and usefull to all network admins out
there for years to come.

LS