Answer the question
In order to leave comments, you need to log in
High load on the network (maybe the switch is faulty)?
Good afternoon friends!
In general, the task is this, there is an institution with 400-450 computers on the network.
And there are several profile programs, for example, "Sail" and a program for recording by a client and sending online reports (users work with this program via RDP).
The other day a new server came and it was installed in a new server room, and there is another server room, it stands separately 40 meters from the new one. The new one has one switch, the old one has 2 switches. Also, the institution is scattered in almost every office on the switch, each switch is not manageable, mainly Dlink and HP.
The problem is this.
1) As soon as people start working with the program for recording clients, the ping to the server increases and is approximately 10-55 and varies constantly. Of course, the program starts to freeze. It is the program itself that hangs, the RDP continues to work. And it doesn’t freeze at all, but it starts to think for a long time and then gives out information.
2) When you start copying files over the network, the connection also sags, and then the server with the "Sail" program starts to fall off.
Both programs are on different servers, in different places. They are connected to each other by a single cable through switches. What could be such a problem? How to identify and how to fix. I'm not going to use my mind anymore.
If you need any more information, please write.
Answer the question
In order to leave comments, you need to log in
400-450 computers + 2 servers.
Such a network is a priori not able to work on unmanaged 5-16 port dgluks and xp, how do you manage routing over it?
Here, only the reorganization of the network, or at least the core on L2-L3.
And better and enlargement of nodes up to 24-48 ports is mandatory with gigabit uplinks
>> Also, the institution is scattered in almost every office on the switch, each switch is not >> controllable, mainly Dlink and HP.
You most likely answered your own question. At the time of the drawdown, start turning off the network segment by segment until the problem is identified.
Good afternoon.
First you need to determine the problem - do you have a problem with the network, or with the server? If there are problems with access ONLY to a specific server - the problem is NOT with the network. I understood correctly - ALL program components are installed locally on the 1st server, do not have any clients, and are used only in RDP sessions? If yes, then this again indicates that the problem is exclusively local, and has NOTHING to do with the network. Look at the logs, study what is happening on the server. A good incentive to learn how to use Performance Monitor.
Didn't find what you were looking for?
Ask your questionAsk a Question
731 491 924 answers to any question