Nov 01 2016 01:12 AM
Nov 01 2016 01:12 AM
First off I know direct booking is bad and that wherever possible the use of free / busy scheduling should be used.
I have observed a number of our executive assistants who manage the calendar and meetings of senior executives co-ordinate the diaries of up to 12 people. They are currently using Outlook 2010 and move in and out of various calendars sometimes maintaining a side by side view of all 12 calendars and managing direct booking of the executives.
Migrating the users to Office365 we have placed the executives into hybrid cache mode as we also update the client to Outlook 2016. The executive assistants are complaining of slowness in moving between calendars, updating and deleting appoitments. Their own mailboxes are in cache mode but we have disabled caching all the various executives as shared folders, so accessing shared folders is done online.
Reproducing the practice on a test setup "feels" like the additional Calendar sessions are timing out and the renegotiation of the session is what is leading to the slow performance. For example staying within a single mailbox calendar and creating one appointment after another doesn't seem to lag. But this isn't really how they work across the calendars as they dip in and out during the day.
Can anyone point me in the direction of how the HTTP sessions "stay alive" and whether or not this can be configured to keep it alive for longer? Another other suggestins of how to improve the overall end client experience in the scenario I have described?
Nov 01 2016 06:48 AM
Nov 01 2016 08:08 PM
@Matt Karel thanks for sharing, I'd be interested in what Premier find and suggest. I might do the same if not further feedback here. What are the average response times you are seeing?
Our tenant is in Ireland and the users being most vocal are located in Hong Kong, the average response time is around 600ms.
Nov 24 2016 03:12 AM
@Matt Karel any update on your case with Microsoft? Even if you could share the troubleshooting steps to date. Calendaring remains our number one compliant for migrated users
Dec 01 2016 12:20 PM
Feb 28 2017 10:20 AM
I dont know if our issue s are exactly the same as we are seeing perf issues when opening mail, caledar and other activites within Outoook. At this point we think the issue is somewere within the MAPI/HTTP stack in IIS or Windows but more debugging would be necessarry to see and we dont have the hours to burn. We have found that if we move the user from using MAPI/HTTP to connect to their mailbox to use RPC/HTTP the issues seem to be resolved. I would sugesst that you try the same. to do this with in powershell enter Set-CasMailbox -Identity "user Alias" -MapiHttpEnabled $False
if you find that this helps your issue please if you can open a ticket with Microsoft and ref SR 116102414841137