Daily slowdown on LAN

A

Andy

Hi

I maintain a small corporate network with aprox 30 workstation, all running
Windows XP Sp2 with all current patches. The network runs on Nortel
switches, and the machines have primarily 256 mb of ram, and CPUs from 933
MHz to 1,6 GHz.
Almost every day, at aprox the same time, several of my network users
complain over slow machines. The complaints are all the same:

1) Takes time to access network resources
2) Takes time to save documents in Word and Excel primarily
3) System hangs and is none-responsive

This seems to be a fairly new problem, so I don't really thing this is
hardware related, even though I recognize the fact that we should have had
more ram.

/A.
 
R

Roberto

better if you post your question here - microsoft.public.windows.server.sbs

rgds
Roberto
 
R

Rick \Nutcase\ Rogers

Hi Andy,

If it happens at the same time every day, then it would lead me to believe
that there is a problem with network congestion buildup. It's peaking and
the current topology is falling over. I would re-examine the way it is set
up, particularly if there a number of hubs in use between the switches and
the end users. Sure the ram is a bit light, but this should not cause the
network slowdown. You mentioned that it is new, was anything recently added
to the network or was some part of it changed?

--
Best of Luck,

Rick Rogers, aka "Nutcase" - Microsoft MVP

Windows help - www.rickrogers.org
 
H

HeyBub

Andy said:
Hi

I maintain a small corporate network with aprox 30 workstation, all
running Windows XP Sp2 with all current patches. The network runs on
Nortel switches, and the machines have primarily 256 mb of ram, and
CPUs from 933 MHz to 1,6 GHz.
Almost every day, at aprox the same time, several of my network users
complain over slow machines. The complaints are all the same:

1) Takes time to access network resources
2) Takes time to save documents in Word and Excel primarily
3) System hangs and is none-responsive

This seems to be a fairly new problem, so I don't really thing this is
hardware related, even though I recognize the fact that we should
have had more ram.

A scheduled task (i.e., virus scanning) that snatches all the resources on
one heavily-used machine?
 
R

Rick \Nutcase\ Rogers

Further thoughts, as it seems to occur at roughly the same time every day...

- Is someone coming in at that time? They could be logging in and using the
company's pc's for large amounts of downloading. Are you tracking activity
and blocking well known ports?

- Is there some online multimedia webcast that is popular in that crowd? It
could be a number of them are trying to view it and traffic slows
accordingly because of the sudden increased demand.

--
Best of Luck,

Rick Rogers, aka "Nutcase" - Microsoft MVP

Windows help - www.rickrogers.org
 
G

Guest

I think that you are into a 'classic' problem area of windows networking.
There are many, many, many knowledge base articles offering help with
specific symptoms in specific applications versions.

Windows tries too hard to find and keep a good connection to a network
resource. And sometimes doesn't do it very well.

Good When they track down a specific instance (usually application,
sometimes OS, bug related) they usually fix it with a patch. So make sure
EVERYTHING, especially server applications, are as up to date and as fully
patched as you can stand.

Bad It's endemic, has roots back to 4.0, and if not a (patchable) bug
only ,usually means a mis-configuration in your setup somewhere (client,
server, application, service, domains, protocols.)

Simple example: an application is trying to access a network resource and
you get an unusual 10 second to 15 MINUTE+! delay (sometimes lockup). You
used to use Netware on your network and (some?) clients still have NetWare in
their protocol stack, but the well known NetWare resource is NOT available on
your network. If a client has NetWare on top of the protocol list some
accesses will wait for NetWare to respond with the resource availability
before trying another protocol. If nothing NetWare answers it may wait for a
timeout period before trying the next protocol. I have seen these timeouts
set by default to as much as 900 seconds.

More complicated example: anything to do with Webdav. Which includes
anything with Office 2000 or newer, and LOTS more Microsoft stuff. One
example of aligned problems with this is that some applications try to find
the BEST access method, so they try all of the ones they can find, and then
don't use any until ALL access methods try, respond or timeout! Often a 600
second timeout if you hit a DFS server resource running a web server not
properly configured to answer webdav resource requests.

More insideous example: Cached or saved references to a resource that is no
longer usable or is very sub-optimal. My worst experience; saved Macro
reference in a Word document family that pointed to a no longer existent
Netware share that had been replaced with a Windows Share of the same name
but with content in a different arrangement, partially. The delays &
timeouts! The weird error messages! The inconsistencies!

For the following –
THE MORE EXACT, COMPLETE, AND SPECIFIC THE SYMPTOMS COLLECTED THE BETTER!!!!

Some of this Microsoft network troubleshooters actually help solve. You
have to know enough facts to answer their questions well though.

Microsoft support can also usually run them down for me if all else fails.
It may not be fast! It may though be free if a patch fixes it.

The knowledge base has a lot of articles. Online search engines often give
better results. Google especially is often MUCH better at finding
appropriate Microsoft KB articles then TechNet or MSDN search is.

If anyone knows of a good central resource at Microsoft to use in these
cases I'd love to know about it though! (I have one, or maybe two ;-{ such
ongoing problems right now!)
 
G

Guest

Andy,

I'm having a pretty much identical problem. Our network is about 750
workstations, XP SP2, hardware well within specs. The problem is related to
either three of our file servers or the workstations accessing them (i.e. My
Documents). What we've seen is that if we access the servers either locally
or over rdp the problem goes away almost instantly. Right now I've got rdp
sessions running continually to three servers to keep the document
directories functioning.

The problem isn't affecting any of my other 75 servers. It seemed to start
around 2/25, a few days before you posted this, but we haven't been able to
find a patch that should be causing the problem. Any info you've got would
be greatly appreciated. I'll pass along anything I find.
 
G

Guest

Hi all,

Have you checked to make sure there isn't a user doing a backup to a server?

This has happened to a couple of clients.

Cya,

Brett
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Top