Lync Server 2013, some funky issues identified. One Pool, Two Front End Servers.
Users connected to FE1 can see status of User X.
Users connected to FE2 cannot see status of User X. User X appears offline for X days.
User can connect to FE1 with firstname.lastname@example.org
User cannot connect to FE2 with email@example.com – Username and password is rejected.
Dbanalyze of firstname.lastname@example.org on each FE server show different results.
FE2 shows old contacts and conference data.
FE1 shows the live and accurate data.
Although a pool with Two FE servers aren’t recommended, they are supported.
Through further investigation, I connected to each servers RTCLocal Databases, under Resources.dbo I found the user. The ID of the user differed. This looked like orphaned data that was not being cleared.
To resolve, I’d need to remove the user from the SQL Database. I followed this process
Disable from Lync Server
Delete from Databases
OSQL -S fe1\rtclocal -d rtc -E -Q “exec RtcDeleteResource ‘email@example.com'”
OSQL -S fe2\rtclocal -d rtc -E -Q “exec RtcDeleteResource ‘firstname.lastname@example.org'”
What causes this?
Each impacted end user, I found had either
– Left the company and came back
– Had their account removed and readded for one reason or another
This looks to have been a problem in Live Server, http://support.microsoft.com/kb/885342 . Our organization had followed an upgrade path to date of Live > OCS > Lync 2010 > Lync 2013.
This appears to be a problem with users who have followed along since OCS or Live, which communicating with each affected person appears to be the case.
At this stage, nuking the users from the database appears to be the resolution. As to the actual cause, I do not know yet. More investigation required.