cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
765
Views
0
Helpful
4
Replies

EM and Corporate directory slow to respond

Two cluster (7835) 6.1 implementation.

Only a handful of users / phones on system at moment (20). Users telling me it takes sometimes 40 seconds to display corporate directory and EM login.

Enterprise parameters set to IP address of publisher not name.

Restart of the call manager service helps to fix the problem to around 10 seconds, but does not fix it permanently.

Customer not happy; where should I look?

Thanks

4 Replies 4

leighharrison
Level 7
Level 7

Hi there,

Both of these options call information from the Call Manager servers. If it's slow to respond, then it'll either be the network or the server that's causing the slowdown.

Have a look at the URL that's called for the corporate directory (for example). That should be something along the lines of:

http://10.17.0.200/CCMCIP/xmldirectory.asp

(it will tell you what it is under service parameters)

If you pop that into IE, it should pop up with a screen with some bits of code on it and it should take the same amount of time as the phone takes to call it.

Have a look to see that the server's process isn't going mad and maybe use a sniffer to see how long the packets are taking to come back to see if you can tie it down to the server or the network.

Hope that helps,

LH

** Please rate all posts **

Hi,

These are the URL's

URL Directories : http://172.16.221.6:8080/ccmcip/xmldirectory.jsp

URL Services : http://172.16.221.6:8080/ccmcip/getservicesmenu.jsp

However it is not just the directories and services. It seems to be the entire web server. Constantantly getting loading whilst trying to do admin, sometimes taking 5 minutes to respond.

Took over a minute to download the RTMT plugin (32mB) and kept stop/starting.

Access to subscriber web pages is very very fast.

Checked network ports, call manager set to auto/auto and plugged into 6509 port also on auto /auto. Had customer change port to speed 1000 and reboot publisher.

Running 6.1.2.1000-13, latest release, but I note that there are some ES out.

RTMT said 97% processor free, but only 24Mb memory free (not much); wondering if there is a memory leak?

Hello again,

24MB is not a great deal to have free. Have a look at the processes and see what's eating it all up. 7835's can handle up to 32GB ish of RAM, but come shipped with usually 2 or 4. Either way, if you've got 2 GB of ram and the server has eaten up 1.98 GB with just a few phones on, then there's a problem there.

If you're nailing a network port speed and duplex, then you have to do it the same on both ends, otherwise it'll not work properly. Best practice with gigabit interfaces is to leave them as auto/auto.

My finger would point strongly at memory usage, that would also explain why a reboot clears it up a little.

Have a look at the processes that are running on the server and see what's taking up all all of the memory. If it's an obvious service that you can operate without, try stopping it and see if that clears up the problem and gives you back the memory. If it's an obscure system service, then you'll have to give TAC a call (unless you can re-write call manager system code!).

Let me know how you get on

LH

I had customer reboot the publisher and things seem to have settled down.

8 seconds to log on now, thats start to finish.

Processor is sitting at less than 10%

Virtual Memory around 36%.

Physical memory is sitting at 29Mb free. Is that correct?

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: