Server 3.1 Issues

brucefromamherst,

I think this should stop this problem for one month:

sudo psql -U _devicemgr -d devicemgr_v2m0 -h /Library/Server/ProfileManager/Config/var/PostgreSQL -c "UPDATE devices SET last_update_info_time = dm_current_timestamp() + '1 month'"

Copy the above and paste it into Terminal (or remote ssh shell) on the server as an admin user. You'll almost certainly be prompted to enter your admin password.

Note that until a proper fix is released by Apple, it would appear that enrolling a number of new devices could lead to this problem recurring in less than a month, but you can always run the above command again.

Please let me know if this seems to bring things "under control" for you. (It won't be immediate, but it should show signs of working within a few minutes.)

I would also recommend that you run this command, to keep newly-enrolled devices from causing the same problem after a couple of days:

sudo psql -U _devicemgr -d devicemgr_v2m0 -h /Library/Server/ProfileManager/Config/var/PostgreSQL -c "ALTER TABLE devices ALTER COLUMN last_update_info_time SET DEFAULT dm_current_timestamp() + '1 month'"

Thank You so much for posting the Problem (AND THE SOLUTION)

Had absolutely the same issue with the server after upgrade to Mavericks Server 3.1 . Even before I have noticed that the drive filled with 180GB of log files (in multiple locations) I had problems imaging a computer lab, the Mac would spontanously stop imaging and Quit. - There were multiple postgres_real tasks that would hog 3-4 cpu cores. also PHP-FPM and Bzip processes would hog the server at 100% times. this BZIP led me to find that the Logs are filling up ... under ProcessManager directory. . It was a havoc. I have changed the above command from "1 month" to "1 week" .... but in truth will be looking for official Apple solution soon!

Thanks Again!!!!

I ran in to the same Profile Manager issue. I solved it temporarily by making the logs unwritable:

sudo chmod 000 /Library/Logs/ProfileManager/php.log

I see this exception thrown a lot in php.log:

Caught error 'Undefined index: CommonName' of severity 8 at /Applications/Server.app/Contents/ServerRoot/usr/share/devicemgr/backend/php/ta sk.php:634

Working with Apple on this issue, debugging might be on causing the Logs to swell. Check the Profile Manager logs and see if debugging is indeed on, switch it off with:

sudo defaults write /Library/Preferences/com.apple.ProfileManager debugOutput 0

The other issue that isn't solved for us yet (2 servers) is that Profile Manager's database back up has gone hog wild doing backups every few minutes, swelling the backup directory to 6GB in about 18 hours (and counting) for us with no end in sight.

Check that directory at /Library/Server/ProfileManager/Data/backups

Please let me know if any of you see this backup issue on upgraded servers too (and a fix would be nice if you know one!)

Apple is always very helpful when we call about these issues. I sometimes wonder if they test at all with upgrading a server rather than plopping it on a fresh machine and declaring it working. Most problems seem to happen with people trying to upgrade in a production enviroment rather than clean installs.

Have the same issue here since upgrading to Server 3.1

Apple, please fix your software that we all depend on. There's no excuse for these kinds of bugs in seemly every release of your server app.

Now after a reboot, Profile Manager won't start up and the Server App won't display any interface for managing the service.

Nice work, Apple.

Chris,

Did you run the two commands I posted above? If so, do you have the output of those commands that you can paste here? (You can run them again and paste the output from the 2nd attempt, too.)

Thanks.

I ran the commands but I don't have output. Since Profile Manager won't even start now, I'm screwed. Like so many times before, I'm going to have to blow away the server and rebuilt it. Same old, same old when it comes to Apple's 'server' software.

Backups, backups, backups. I know that doesn't help you now, but...

Same issue here. php.log was growing at a rate of over 50 mb/second and I came in this morning to a full 1TB server.

We're rolling back, since I've added no new devices/settings. When Apple gets their fecal matter consolidated, we'll move forward.

jwzg2 wrote:

Backups, backups, backups. I know that doesn't help you now, but...

Same issue here. php.log was growing at a rate of over 50 mb/second and I came in this morning to a full 1TB server.

We're rolling back, since I've added no new devices/settings. When Apple gets their fecal matter consolidated, we'll move forward.

Hello jwzg2,

which steps did you make to recover to a Backup?

Only restored the old /Applications/Server.app? The hole System? Or just the Database?

I am missing exactly instructions how to do this and keeping the database and the "connected/enrolled" devices and policies.

Thank you, ...

https://discussions.apple.com///thread/4940719

Check the last post for the link to restoring the ODM.

If you aren't getting at least a Time Machine backup daily, and a weekly image, you are playing with fire.

I had the same issues as the OP after upgrading from 3.0.2 to 3.1. After having gotten burned upgrading to Mavericks/3.0 from Mountain Lion/2.2.2 I now make a practice of taking a full disk image backup of the server before doing an upgrade, using Carbon Copy Cloner.

I boot from an external drive with CCC and dump the entire server HD (using block copy to a compressed disk image on the attached promise RAID). It does require taking the server offline, but the entire process takes < 5 minutes (backing up ~35GB of data from an the server's boot drive - an SSD). I also have a script that does nightly database dumps of the wiki server, profile manager, OD, and a couple of custm apps I have running on the server. If my full image is a bit out of date, I can still restore and then bring it up to date by restoring wiki or PM databases as needed.

I just got off the phone with Apple Enterprise support about the Server 3.1 Profile Manager log issue (my server had a 455GB php.log this morning!)

From Apple Enterprise:

sudo psql -U _devicemgr -d devicemgr_v2m0 -h /Library/Server/ProfileManager/Config/var/PostgreSQL -c "UPDATE devices SET last_update_info_time = dm_current_timestamp() + '1 month'"

sudo psql -U _devicemgr -d devicemgr_v2m0 -h /Library/Server/ProfileManager/Config/var/PostgreSQL -c "ALTER TABLE devices ALTER COLUMN last_update_info_time SET DEFAULT dm_current_timestamp() + '1 month'"

A fix is in the works allegedly, the above commands take care of it for 1 month. I haven't compared specifically, but it looks like the same two commands posted by mscott_mdm. I didn't know it, but you can just highlight the command and drag it into a terminal window (thanks Apple support guy!). No copy/paste necessary! At least something positive came of this!

THANKS!

--Dan