Looks like no one’s replied in a while. To start the conversation again, simply ask a new question.

User profile for user: timaeuz
timaeuz Author
User level: Level 1
0 points

Profile Manager Error prevents startup

Immediately after installing OSX Server and choosing or creating an SSL certificate,

Profile Manager gives the error "Error Reading Settings" while updating services.


So as a result,

when OSX Server tries to write out the Profile Manager Settings when I change to a different certificate,

OSX Server generates these two errors:

The error occurred while processing a command of type 'writeSettings' in plug-in 'servermgr_devicemgr'.


Followed by...


Multiple errors occurred on the server while processing commands.

Use the Console application to view the error messages.



So I tried running this command from the terminal:



/Applications/Server.app/Contents/ServerRoot/usr/share/devicemgr/backend/wipeDB. sh



and got the following response in the error log on the console:



2012-09-06 11:22:24.449 AM ProfileManager[8659]: devicemgrd: Terminating on unhandled exception Rake command failed with 256 at /Applications/Server.app/Contents/ServerRoot/usr/share/devicemgr/webserver/devi cemgrd:258:in `Rake'

/Applications/Server.app/Contents/ServerRoot/usr/share/devicemgr/webserver/devic emgrd:222:in `SetupRails'

/Applications/Server.app/Contents/ServerRoot/usr/share/devicemgr/webserver/devic emgrd:150:in `Run'

/Applications/Server.app/Contents/ServerRoot/usr/share/devicemgr/webserver/devic emgrd:833

2012-09-06 11:22:24.450 AM com.apple.launchd[1]: (com.apple.devicemanager[8659]) Exited with code: 1

2012-09-06 11:22:24.530 AM ProfileManager[8678]: Failed to delete '/var/devicemgr/ServiceData/Data/tmp'. No such file or directory - /var/devicemgr/ServiceData/Data/tmp

2012-09-06 11:22:26.662 AM ProfileManager[8678]: Retry #1 of database migration...

2012-09-06 11:22:28.778 AM ProfileManager[8678]: Retry #2 of database migration...

2012-09-06 11:22:30.015 AM assistantd[294]: <Error>: AceConnection - NSStreamEventErrorOccurred <__NSCFInputStream: 0x7fe17100be60>, error = Error Domain=NSPOSIXErrorDomain Code=54 "The operation couldn’t be completed. Connection reset by peer", domain = NSPOSIXErrorDomain, code = 54

2012-09-06 11:22:30.022 AM assistantd[294]: <Error>: Session - Connection error: <ADAceConnection: 0x7fe173319640> Error Domain=NSPOSIXErrorDomain Code=54 "The operation couldn’t be completed. Connection reset by peer"

2012-09-06 11:22:30.031 AM assistantd[294]: <Error>: Daemon - Session Error Error Domain=NSPOSIXErrorDomain Code=54 "The operation couldn’t be completed. Connection reset by peer"

2012-09-06 11:22:30.931 AM ProfileManager[8678]: Retry #3 of database migration...

2012-09-06 11:22:33.052 AM ProfileManager[8678]: Retry #4 of database migration...



I think something is missing in the installation that I may have deleted. I have tried reinstallation of the OSX Server after deleting the /Library/Server directory and the server.app file and rebooting.



I have tried reinstalling OSX Mountain Lion and the latest OSX Server version as well.



No joy, not even a tiny bit.



Question is how or where do I find out how to restore what's missing, or fix what is broken?



Thank-you for your expertise,



Tim

iMac, OS X Server, Mountain Lion 10.8.1

Posted on Sep 6, 2012 10:54 AM

Reply
8 replies
User profile for user: Mark23
Mark23
User level: Level 4
994 points

Sep 9, 2012 10:49 AM in response to timaeuz

No need to reinstall server:


Important notes

  • Resetting Profile Manager to its original state removes all profiles, devices, and device groups. Only do this if you want to start over with Profile Manager without reinstalling OS X Server.
  • Any devices that have profiles installed from this server will no long receive updates, and you won't be able to manage them unless they re-enroll. The profiles should be deleted from the devices.
  • The certificates will not be removed.


To reset Profile Manager

  1. Quit Server.app if it's running
  2. Execute this Terminal command as an admin user, entering the password when prompted.
    For OS X Mountain Lion only, execute:
    /Applications/Server.app/Contents/ServerRoot/usr/share/devicemgr/backend/wipeDB. sh
    For OS X Lion only, execute:

    /usr/share/devicemgr/backend/wipeDB.sh

  3. Open Server.app and click Service, then click Profile Manager. Note that Profile Manager Manager is on and Device Management is Disabled.
  4. Switch Profile Manager off.
  5. In Profile Manager, Device Management click Configure, then Next.
  6. Enter the Organization Name and Admin email address.
  7. Click Next.
  8. Select the Certificate.
  9. Click Next, then Finish.
  10. Start Profile Manager.
  11. Click your server's name in the Hardware section.
  12. Click Settings.
  13. In the SSL Certificate, click Edit...
  14. Click OK.
Reply
User profile for user: ThePulsingEye
ThePulsingEye
User level: Level 1
0 points

May 17, 2013 1:18 PM in response to timaeuz

I was having this problem too, and it was due to a conflicting ruby gem. rake 10.0.4 was installed and server.app was using that because it did not have a version of rake installed in it's own gem path.


here's what I did from cli


## install rake locally

$ gem install rake -v 0.8.7


##then copy to the Server.app directory

$ cp -r /Library/Ruby/Gems/1.8/gems/rake-0.8.7 /Applications/Server.app/Contents/ServerRoot/usr/share/devicemgr/webserver/gems /gems/


$ cp /Library/Ruby/Gems/1.8/specifications/rake-0.8.7.gemspec /Applications/Server.app/Contents/ServerRoot/usr/share/devicemgr/webserver/gems /specifications/rake-0.8.7.gemspec


### if this dosn't sort it out in and of itself you may want to try

$ /Applications/Server.app/Contents/ServerRoot/usr/share/devicemgr/backend/wipeDB .sh

Reply
User profile for user: Halloween_Jack
Halloween_Jack
User level: Level 1
4 points

Aug 21, 2013 8:43 AM in response to Mark23

That IS the command timaeuz ran, and it didn't work for him.


It's also not working for me. I'm looking at the devicemgrd.log file scrolling now, and I'm getting the same result. It keeps retrying:


Aug 21 10:29:18 ProfileManager[10232] <Error>: Retry #3 of database migration...

Aug 21 10:29:18 ProfileManager[10232] <Info>: Running rake command: '/usr/bin/rake --trace db:migrate'

Aug 21 10:29:20 ProfileManager[10232] <Info>: Output of rake command '/usr/bin/rake --trace db:migrate:

rake aborted!

could not connect to server: No such file or directory

Is the server running locally and accepting

connections on Unix domain socket "/Library/Server/PostgreSQL For Server Services/Socket/.s.PGSQL.5432"?


In the WikiConfig.log I get the following:


E, [2013-08-21T09:34:43.510928 #9119] ERROR -- : Timed out waiting for collabd service.

W, [2013-08-21T09:34:48.003350 #9162] WARN -- : Retrying connection to collabd service...

W, [2013-08-21T09:34:50.007130 #9162] WARN -- : Retrying connection to collabd service...

W, [2013-08-21T09:34:52.010615 #9162] WARN -- : Retrying connection to collabd service...

W, [2013-08-21T09:34:54.014274 #9162] WARN -- : Retrying connection to collabd service...

W, [2013-08-21T09:34:56.018076 #9162] WARN -- : Retrying connection to collabd service...

W, [2013-08-21T09:34:58.021360 #9162] WARN -- : Retrying connection to collabd service...

W, [2013-08-21T09:35:00.024217 #9162] WARN -- : Retrying connection to collabd service...

W, [2013-08-21T09:35:02.027679 #9162] WARN -- : Retrying connection to collabd service...

W, [2013-08-21T09:35:04.030850 #9162] WARN -- : Retrying connection to collabd service...

E, [2013-08-21T09:35:04.033394 #9162] ERROR -- : Timed out waiting for collabd service.


Now, I'm wondering if this is caused by me telling Server.app to move the server data to a different volume. I did this because the boot volume is an SSD with limited space. So I told Server.app to save the data in an internal RAID volume so I can get enough space for NetInstall images and Software Updates.


I compared the (original) /Library/Server/PostgreSQL directory in the boot SSD to the same directory in the RAID, and two folders are missing in the latter, the "Config" and "Socket."

Reply
User profile for user: Sylnsr
Sylnsr
User level: Level 1
0 points

Nov 9, 2013 10:43 AM in response to Halloween_Jack

Halloween_Jack, I think this will solve your issue: http://sylnsr.blogspot.com/2013/11/os-x-mountain-lion-server-profile.html


I had basically the same issue because the volume where my service data used to be, crashed. After I specified a new volume in Server.app, the configs were still pointing to the old unavailable volume. I had to manually update the config files that Server.app failed to update.

Reply
User profile for user: Halloween_Jack
Halloween_Jack
User level: Level 1
4 points

May 15, 2014 8:57 AM in response to ThePulsingEye

It didn't work.


What happened is that when I updated the OS (eg. 10.8.4 to 10.8.5) and the Server app, for some reason it now sees the RAID volume where the /Library/Server/ directory is as "STORAGE 2" instead of just "STORAGE".


I looked at the instruction in the link that Sylnr provided, but when I checked the xml files they seem to have the current path using "STORAGE 2". So, I don't know what else I'm supposed to check for.


The only way I was able to fix this issue before was to just completely re-do the server because even restoring the directory and settings from Time Machine didn't work.


So, basically, it seems one is supposed to just use the boot drive for everything even though it's small. I'm not even running the Wiki server or Profile Manager because of this stupid problem that can't be fixed easily. The other reason is the limitations of Profile Manager so it's not a tool we can use anyway. However, if we would've run an internal Wiki, this would've completly broken the thing.


Apparently the problem still exists in Mavericks server, so this is something that Apple needs to just fix.

Reply

Profile Manager Error prevents startup

Welcome to Apple Support Community
A forum where Apple customers help each other with their products. Get started with your Apple ID.
Learn more Sign up