Networker console performance issue

Hi Pritesh, Because you are in a production environment, I am going to move this thread to our Support Forums Community. They will be able to address your questions there. -Stephanie

Hi, I've moved your thread to the Networker support forum where it should receive prompt attention. Thanks!

Hi,

There is a lot of good recommendations in the NetWorker 7.6 Performance Optimisation Guide which you should read.  You'll find that the first one is to definitely move your NMC server to another machine.  This will almost-certainly fix the problem, but if you still have problems, you will need to analyse where you are running out of resources. According to the performance guide:

"
Components that determine the performance of NMC are:
- TCP network connectivity to backup server: All communication between NMC
and NW server is over TCP and such high-speed low-latency network
connectivity is essential.
- Memory: Database tasks in larger environments are memory intensive, make sure
that NMC server is equipped with sufficient memory.
- CPU: If NMC server is used by multiple users, make sure that it has sufficient
CPU power to ensure that each user is given enough CPU time slices.

"

-Bobby

Pritesh

Can you give us some information on the specification of your Networker server; are all 1000 clients on the same backup server?  Do you know what sort of size your media database is (nsr/mm directory)

David,

The networker server version is 7.6 and back hand is DD880. Each day we are backing up around 10 TB of data. All 1000 clients are backed up through three different storage node and its all load balanced. 

Do you see any other functions of the software running slow?  Do both users experience the same slow performance?  What about if just one person is using the GUI?

Almost in every scenario  with a large environment.....the NW console is recommended to be installed on a server other than the NW server.

You will definitely see perfomance issues if installed on the same server and its a busy environment. I have seen this time and time again at various places.

David,

yes, when two or more user using GUI, its slowing down too much. we need to wait 5-7 minutes for changing any configuration in clients. Sometime its throwing time out errors as well. With one user its working good but not very good.

Have you gone onto the backup server and monitored performance there?

As Bosadb states if you have a large environment (which you have) it is generally advisable to install the NMC on its own server which you state that you haven't done, that is really your long term solution here.  You have also not gave any details of the specification of your backup server which will also have an impact.  I've seen similar issues in sites with very large media databases so it is still worthwhile looking at the size of your /nsr/mm folder, if this is over 1GB you should think about splitting the server or reducing retention times.

Hi Pritesh

The first step you need to do accordling  with the NMC requirement (  Networker Server with more than 100 client ) is to move your NMC  on another hosts. 

Oh, you resurrected this thread from death   I believe the key point here is that two or more users when logged with slow down NMC overall (and server too) which is true.  And that has nothing to do if server is running NMC or not. Actually, the issue here is such that impact will be even bigger if it is not running on the same host.  This is because slowdown is caused constant polling of information against jobdb and other daemons.  You can fix is somewhat by running this on separate mnt points for mm, index, tmp, res and most importantly for jobdb database, but things get a bit too butchered then.  In NW8 this whole setup was changed so there are some performance gains there.  As for the setup itself, I agree that 1000+ is overkill and against best practice (actually, against any sane recommendation). You would need to split the server into two and I agree that eventually you should deploy some vm and run NMC on that.