Q: Upgraded to OS X Server 3 and now all my Enrolled devices are gone
-
Oct 23, 2013 11:36 AMby Network Engineer,I am also seeing the same behavior... Using pgAdmin3 I was able to connect to the PostgreSQL instance... It seems there is an entirely new database titled devicemgr_v2m0. This is what the current profile manager is making use of... I noticed that the old schema was in fact migrated in the database device_management. It still contains records for all registered devices and such, however it seems that information never made it over to the devicemgr_v2m0 database.
The new table schemas are different otherwise I'd try backing up and restoring various tables into the new database..
Have any of you noticed that device_groups are being restored? That is the only thing on my MDM that made it over, however device group memberships did not transfer...
-
Oct 28, 2013 11:50 AMby mscott_mdm,You might try the command I suggested at the end of this other, similar discussion:
https://discussions.apple.com/thread/5467701?start=15&tstart=0
Which is this:
sudo psql -U _devicemgr -d device_management -h /Library/Server/ProfileManager/Config/var/PostgreSQL -c "update devices set \"InstalledApplicationList\" = NULL;"
sudo /Applications/Server.app/Contents/ServerRoot/usr/share/devicemgr/config/wipeDB. sh
Note that the 2nd command (wipeDB.sh) will erase ALL current Server 3 Profile Manager data (but only Profile Manager data--data of the other Server 3 services will not be affected) and, if you upgraded from Server 2.2 or earlier, will try to migrate the pre-Server 3 Profile Manager data again.
-
Oct 29, 2013 7:50 AMby trazzel,I lost all the settings in the devicegroups after the upgrade to 3.0.
I also realized that the upgrade destroyed my open directory leaving me no choise to restore my server to 10.8.5 again.
I have 1594 iPads and a handful of MacBook Pros that I manage with PM, and with no open dir, no new devices could enroll.
So I'm quite upset that Apple isn't testing their software good enough before selling it.
It's Ship-n-fix mentality.
/traz
-
Nov 4, 2013 6:58 AMby organicchunkysalsa,Has anyone that is having this issue tried the fix that mscott_mdm has mentioned?
I too am in the same boat and had to re-add all of my profile settings but the server isn't pushing profiles as it should. I am trying to decide now if I should press forward and try to fix the server or if I should roll back.
-
Nov 4, 2013 9:36 AMby mscott_mdm,lelio98,
Your problem appears to be different from the one in this thread. You would mind starting a new discussion and adding the link to it here? If you would also paste into the new thread the last 50 or so lines from the migration_tool.log and devicemgrd.log files (in /Library/Logs/ProfileManager) and I will look at those in the new thread. Thanks.
-
Nov 4, 2013 9:47 AMby organicchunkysalsa,I also tried the fix and it didn't work but I may have also had a different issue. In the meantime I rolled back my server to 10.8. I am going to continue to follow these discussions to see how things pan out but it sounds like a handful of people are running into these issues.
-
Nov 4, 2013 12:27 PMby lelio98,According to the migration log on mine, it isn't the installed applications:
1:: [552] [2013/11/04 08:32:37.280] Starting migration of 'LabSession'...
2:: [552] [2013/11/04 08:32:37.575] Updating PGModel 'LabSession' (table 'lab_sessions')
0:: [552] [2013/11/04 08:32:39.548] [1;7;31mEXCEPTION: [0;31m Postgres <-[PGConnection executeSQL:withParams:] (/SourceCache/RemoteDeviceManagement/RemoteDeviceManagement-848.17/Compiled/Fra mework-Models/Postgres/PGConnection.m:432): "Postgres error 22001 (ERROR: value too long for type character varying(63)
)"> [0m [0m
Is it safe to remove this too?
