I have a customer that outsources their Exchange environment. They want to use the CUPS calendar integration with their provider. I actually got the certificate issues resolved with Entrust, and I can see appointments in the IPPM. However, The CUPC does not dynamically update the meeting status as I have seen with other installs. The only way to get it to show the correct status is to disable the setting in the status menu under preferences in CUPC then re-enable it. I have a feeling that my issue stems from being behind a firewall/NAT but have not had the opportunity to investigate the firewall for any such activity.
Anybody have any ideas/suggestions on how to proceed? TAC hasn't had a whole lot to offer up to this point.
Thanks for any input!
some quick questions:
1) when your contact change their status (available/away), does it reflect on your CUPC realtime?
2) do you observe the same behavior on LAN? Or does it only happen on VPN?
This book has very detailed explanation on this topic:
If you want a quick solution, call 1-800-553-2447 between 1pm-3pm CST, press 1, requeue the case to Michael Luo. He should be able to identify the problem within 10 minutes (if not shorter).
Thank you for the reply, I actually ordered the book earlier today.
1) Yes, the client responds as it should to status changes.
2) I have no LAN access or VPN, this is going over the internet to an Exchange provider. The owa server is a public hostname.
To troubleshoot the problem, we need the "Presence Engine" logs (with trace level set to "detailed").
If the problem was very complicated, we might need IIS log from Exchange OWA. I guess that would be difficult for your, since the Exchange was hosted.
Anyway, get me the PE log with the steps below:
1) Go to CUPS serviceability > trace > configuration > CUP services > Cisco UP presence engine. Set debug trace level to "detailed". Optionally, set maximum file size to 2M.
2) exit CUPC (if it's running)
3) restart presence engine
4) wait for 2 minutes. Log into CUPC.
5) Use RTMT collect presence engine logs from the time it was restarted to present.
6) zip logs. upload here. let me know the Exchange account for the end user.
Per the PE log, the calendar part was working fine.
01/22/2009 22:45:58.017 EPE|system.pe.pa.owa.backend 409894 DEBUG transitionList: date: 2009-01-23T04:45:57Z status: FREE
The message above means CUPS successfully retrieved calendar status.
So I guess the problem was on the communication between CUPS and CUPC.
Sniffer capture would be useful. Assuming you already know how to run sniffer on client side, on server side, you may use command line:
"utils network capture file cups count 100000 size all host all 192.168.1.100"
Substitute 192.168.1.100 with the IP address of CUPC.
If you need further assistance to do that, call TAC.
Thank you for direction on this. When I do the captures, what am I looking for? Do you think it is some sort of SIP issue between CUPC and CUPS? If that were the case, wouldn't the status changes fail as well?
sniffer will tell you if the "NOTIFY SIP" message got to client side or not. It it got to client side but CUPC didn't display it, then it's CUPC issue.
If it didn't get to client side, then it's network issue.
By the way, the traces you sent indicate the calendar status was FREE. It's not a good test for this problem. Because CUPC will show "Available" anyway when your calendar is FREE.
Again, you'd better ask TAC to do the test for you. If you got to Michael Luo. He should be able to identify the problem pretty quick.