Annoying behavior of Server Admin

Sometimes when I'm done using Server Admin, I accidentally hit the close button on the window instead of quitting the application. Instead of the window closing, it just deletes all the servers that I had listed in the window, and the only way to get them back is to go through and add them all again. This is a huge pain. There doesn't even seem to be a way to add favorite servers or anything that I can find. Is there a way around this problem?

Well, let me clarify, if I quit Server Admin instead of just closing the window, it keeps the servers in the list. Only when I close the window does it lose them. However, if I close the window by accident and the servers disappear, I've found that if I then force-quit Server Admin, the servers will reappear when I launch it again.

Similar Messages

Server Admin suddenly cannot find the host it runs on

I am running mac osx server, snow leopard. It ran fine until just recently. Suddenly Server App is unable to find the host it runs on.
I enter the hosts IP address on the local network, log in with my administrator username and password, and the system comes back telling me
Could not connect to <ip address>
Server Admin was unable to connect to the server at <~ip address>
Server Preferences tells me that this host cannot be found on the network too.
Point is however that both these apps run on the host with that very IP address. It is set up in System Preferences > Network, and is hardcoded, not given by the DHCP server in the router. I can also connect to the web server running on this host using the laptop and an ipad.
This is quite annoying, especially as Server Admin is the only app that lets you manage the host it runs.
Mac mini, OS X Server, version os os x server is 10.6.8

relais:~ locadmin$ sudo changeip -checkhostname
Password:
Primary address     = 192.168.39.200
Current HostName    = relais.local
The DNS hostname is not available, please repair DNS and re-run this tool.
dirserv:success = "success"
and
relais:~ locadmin$ sudo lsof -i -P | grep ':311'
relais:~ locadmin$
The log is telling about a crash in servermgrd, which is restarted and crashing every ten seconds, a bit from the crash report
Exception Type: EXC_BREAKPOINT (SIGTRAP)
Exception Codes: 0x0000000000000002, 0x0000000000000000
Crashed Thread: 0 Dispatch queue: com.apple.main-thread
Thread 0 Crashed: Dispatch queue: com.apple.main-thread
0   com.apple.CoreFoundation                0x00007fff89aa244f __CFTypeCollectionRetain + 191
1   com.apple.CoreFoundation                0x00007fff89aabca5 __CFBasicHashAddValue + 53
2   com.apple.CoreFoundation                0x00007fff89ab3688 CFBasicHashSetValue + 184
3   com.apple.CoreFoundation                0x00007fff89ab34a7 CFDictionarySetValue + 199
4   com.apple.servermgr_dns                 0x00000001031bbd66 addDataToDictionary + 2549
5   com.apple.servermgr_dns                 0x00000001031bc6ec yyparse + 2324
6   com.apple.servermgr_dns                 0x00000001031bd7a6 parseFile + 384
7   com.apple.servermgr_dns                 0x00000001031dac19 optionsParser_lex_destroy + 104800
8   com.apple.servermgr_dns                 0x00000001031d6406 optionsParser_lex_destroy + 86349
9   com.apple.servermgr_dns                 0x00000001031d6468 optionsParser_lex_destroy + 86447
10 com.apple.servermgr_dns                 0x00000001031ce904 optionsParser_lex_destroy + 54859
11 com.apple.servermgr_dns                 0x00000001031cea41 optionsParser_lex_destroy + 55176
12 com.apple.servermgr_dns                 0x00000001031d19c8 optionsParser_lex_destroy + 67343
13 com.apple.servermgr_dns                 0x00000001031c1f78 optionsParser_lex_destroy + 3263
14 com.apple.servermgr_dns                 0x00000001031c67db optionsParser_lex_destroy + 21794
15 com.apple.servermgr_dns                 0x00000001031c254b optionsParser_lex_destroy + 4754
16 libservermgrcommon.dylib                0x000000010003ec18 -[BundleManager init] + 402
17 libservermgrcommon.dylib                0x000000010003dd9d +[BundleManager defaultManager] + 51
18 com.apple.ServerManagerDaemon           0x0000000100003bcb 0x100000000 + 15307
19 com.apple.ServerManagerDaemon           0x00000001000014c0 0x100000000 + 5312
So it looks like there is some t=file that is now not exactly like the parser wants it.

Server Admin 10.6 bug: Realms are completely messed up

After upgrading Server Admin to 10.6, I completely lost control over the Realm configuration with Server Admin. This happens regardless if the Server is 10.5 or 10.6.
Example: For a given Realm the permissions for "Everyone" should read "None" instead nothing is displayed (the field is empty). The permissions for a given Group should read "Browse and Read WebDAV", instead "None" is displayed.
If I try to change permissions, or give permissions to a new Realm, the popup menu shows the following choice:
*Browse Only*
*Browse Only*
None
Normally it should look like this:
*Browse Only*
*Browse and Read WebDAV*
*Browse and Read/Write WebDAV*
None
I can't say for sure if this has consequences for the actual access rights. I found this problem after upgrading to 10.6 and tested it by connecting from a 10.6 client computer via WebDAV as Guest. The Server gave me full access instead of none. Later I found in the logfile, that I have been connected with my current user name instead of being a Guest, with I explicitly choose to be, in the Finders connect dialog on a 10.6 client.
After that I lost all trust into 10.6 Server so far. Also it was late yesterday, so I decided to restart the Server as 10.5 from a backup disk. I'll give 10.6 a second try if I got some answers first.
Remark: I'm using Server Admin in German Language. I have translated the menu items above for this post. Is this a localization bug? Could anyone confirm, this behavior of Server Admin in other language versions?
_Further question:_ Is this "choose to be a Guest, and be connected as the curred user" behavior a bug in 10.6 Finder?
Message was edited by: root 66

This appears to be a problem with the German-language strings present in SnowLeopard Server. You can try fixing them yourself as a workaround. As an administrative user, make a backup copy of the following file (just in case), and open the file with a text editor:
"/Applications/Server/Server Admin.app/Contents/PlugIns/Web.bundle/Contents/Resources/German.lproj/Localizab le.strings"
That file contains lines that map user interface items to displayed strings. For the popup values of the Permissions section of the Realms display, you will find the following incorrect lines:
"kBrowseOnly" = "Ansehen";
"kBrowseAndReadWebDAV" = "Ansehen ";
"kBrowseAndReadWriteWebDAV" = "Ansehen ";
Change each "Ansehen" to a more meaningful German string - I think you can infer a correct string from the context - and save the changes, and then relaunch Server Admin.

Server Admin GUI weird behavior

SLS 10.7 & Server Admin 10.6.5
Very strange GUI issue when I run Server Admin on my MBP and connect to my local SLS running on a Mac Mini. The first line of every window is half-cut off (see image). It is fine when I run it directly on the Server. Any ideas anyone?

Count me in, as well. I've tried connecting from the server itself, as well as another local machine. I can't enter a bogus serial because it's not even recognizing the server... the serial boxes are all greyed out.
(Server 10.4.10)
Please help!
Jamie
Message was edited by: Jamie Pichora

Server Admin's proxy configuration... whacked? (Doesn't seem to work)

I just upgraded to Apache2 so that I could, sensibly, use the Server Admin app to manage all my web sites. Previously, my various httpd conf files were too complicated and I had resorted to hand coding them.
The good news is Server Admin and Apache2 are much closer to functional.
One bit of bad news: I can't get the reverse proxies to work consistently. Once set up, I get the most peculiar mix of success and failure -- almost at random. For example, some URIs load; others don't; sometimes a page loads, but it's style sheet doesn't.
What I want is simply this:
*/confluence (proxy to) localhost:8081/confluence
Using hand-edited config files, I'd usually do something along these lines (and this works great, by the way, if I stick it into 0000any_80www.mydomain.com.conf:
ProxyRequests Off
ProxyPassReverse /confluence http://localhost:8081/confluence
ProxyPass /confluence http://localhost:8081/confluence
So, off to Server Admin to make this work. I tried checking "Enable Reverse Proxy" and specifying a Proxy Path of "/confluence" and a Worker URL of http://localhost:8081/confluence. This ended up created the following entry in my 0000_...conf file:
<IfModule modproxybalancer.c>
ProxyPass /confluence balancer://balancer-group/
ProxyPassReverse /confluence balancer://balancer-group/
<Proxy "balancer://balancer-group">
BalancerMember http://localhost:8081/confluence
</Proxy>
</IfModule>
I wasn't able to figure out what a Sticky Session Identifier, Route or Load Factor would be used for... so I left those blank. Perhaps that's why it's not working... sure would be nice if there was some help in Server Admin...
Hm. Anyhow, Ok, looks alright. I'm not totally sure about the balancer bits (like why they are there), but I don't see why it wouldn't work. But... IT DOES NOT. I get totally inconsistent results. Some pages load, some don't. Sometimes I get some page assets, while others don't come through. Bah. However, if I get rid of the above, and simply put this in:
ProxyRequests Off
ProxyPassReverse /confluence http://localhost:8081/confluence
ProxyPass /confluence http://localhost:8081/confluence
It works great. Only problem is, anytime I edit a server configuration, Server Admin goes and deletes those lines! I think I've found a way around it though. I put the above lines into a "0000.proxy" file (in /etc/apache2/sites/0000.proxy) and now I've added this line:
Include "/etc/apache2/sites/0000.proxy"
At the bottom of my 0000_...conf file. So far, Server Admin hasn't deleted that line... and it seems to be working.
BUT, I'd just LOVE it if someone could tell me how to make this work without hacking around in the command line! Very annoying -- and of course, my proxies are showing up in Server Admin, and I really wish that would work.

Hi there
For the benefit of anyone looking at this thread, what is meant here is that if you choose to open the help for RoboHelp, you have an abiliity to add a comment. That comment may be kept private or shared with others. If you elect to share it with others, it is generally added as a comment into the pages that provide the help. If approved, Adobe may amend the help topic to reflect the new information.
@Jared - Was that information missing in your file? I thought when you elected to use this file that it was automatically populated with all available settings. And unfortunately (IMNSHO) the settings don't immediately reflect the settings in the dialog. (I think they should, as a starting point. At least, that's the behavior *I* expected the first time I played with the feature. I was admittedly taken aback when it didn't work that way.)
Assuming the setting was missing, how did you manage to sort what setting should be added?
Cheers... Rick
Helpful and Handy Links
RoboHelp Wish Form/Bug Reporting Form
Begin learning RoboHelp HTML 7, 8 or 9 within the day!
Adobe Certified RoboHelp HTML Training
SorcerStone Blog
RoboHelp eBooks

Server Admin 10.6 managing a 10.5.8 server

I'm hopeful someone can point out my error here.
I've been running the new 10.6 version of Server Admin on my 10.6 Macbook to mange my 10.5.8 server, and for a while everything seemed fine. When I'm accessing the server remotely (i.e. my client is not on the same LAN as the server) then it is working fine. An d initially this was also true when I was on the same LAN.
But in the last couple of days, and I don't think I changed anything on the server or client machine that might have impacted on this, Server Admin is reduced to a snail's pace when I try to connect to my server, but only when on the same LAN. This clearly suggests to me that DNS may be the issue, but I've looked at it again and again and can't spot anything out of the ordinary.
While it's not critical (when I'm on the LAN I can fire up Server Admin 10.5 on the server itself and work directly, but it is really annoying me nonetheless.
Regards
Graham

I'm hopeful someone can point out my error here.
I've been running the new 10.6 version of Server Admin on my 10.6 Macbook to mange my 10.5.8 server, and for a while everything seemed fine. When I'm accessing the server remotely (i.e. my client is not on the same LAN as the server) then it is working fine. An d initially this was also true when I was on the same LAN.
But in the last couple of days, and I don't think I changed anything on the server or client machine that might have impacted on this, Server Admin is reduced to a snail's pace when I try to connect to my server, but only when on the same LAN. This clearly suggests to me that DNS may be the issue, but I've looked at it again and again and can't spot anything out of the ordinary.
While it's not critical (when I'm on the LAN I can fire up Server Admin 10.5 on the server itself and work directly, but it is really annoying me nonetheless.
Regards
Graham

Server Admin not drawing graphs

I have a fresh install of Leopard that will not display the graphs in Server Admin on the server. I can see the graphs if viewing from my laptop or other machine, but not from the server itself. It gets a few errors like ...
... [0x0-0x1b01b].com.apple.ServerAdmin[1734]: * ERROR: -[GRAxes computeLayout]:1124 - plotRect height = 0.000000 <= 0.0 *
... [0x0-0x1b01b].com.apple.ServerAdmin[1734]: * ERROR: -[GRChartView computeLayout]:1194 - Layout for overlay axes (0x6826d30) failed. *
The hardware is a Xserve, 2 x 2.0GHz PPC G5 with 2.0 GB RAM.
Any help is appreciated as this is an annoying feature.

Hi
I see the same "error" messages in every leopard server install I've done - and I've done a fair few - regardless of what hardware was used. Graphs draw fine for me.
Tony

Server Admin doesn't show Print Queue status

When I look at my print queues (all 17) under Server Admin, it shows queue name, # of jobs, shared via (IPP & SMB), and printer kind (HP LJ 1022n and HP LJ P2015). It has a column for status, but this is always 100% empty/blank. Also, if I go to the Jobs tab, then I can pick "Jobs on Queue" for all the queues, but the line below for "Queue Status" is again always blank.
Is there something I can do to get this info to show up in Server Admin? I know that CUPS knows the printer status, as I can go to localhost:631 and see the status in the Printers list and the Jobs list.
It isn't life threatening. It is just annoying to have to go to the web page when Server Admin -almost- gets me everything I need to diagnose a teacher's printing problem remotely.

Hi Keith
You could always consult the Print Service logs. If you set the logging levels to Debug or Debug2 you should get everything. You have to be a bit careful with this as the log can get quite large. Other than that I don’t know of anything that is available out of the box.
There may be others that can suggest a 3rd party solution that gives you the appropriate feedback you are looking for. Opening port 631 should not be that much of risk I would have thought?
Tony

Graphs in Server Admin isn't updating any more.

Hi all,
The graphs in Server Admin doesn't show correctly. For any given time range, it's always 31% for CPU utilization and 28KB/s for network. However it's definitely not the case when comparing to Activity Monitor.
The graphs for web showing correctly for both throughput and requests.
Not sure if anyone had this problem before. I googled around and found no similar issues.
It's not a big deal but very annoying as I can't see server performance history any more.
Please help.

According to the manual page on the OS X Developer Library, fsck_cs has a bug:
fsck_cs does not perform an exhaustive validation, nor is it able to fix many of the inconsistencies that it does detect.
Somehow my MacBook Air didn't complete some I/O transactions to the SSD and therefore the physical volume of the core storage has "failed". Fixing inconsistencies didn't worked because of the bug mentioned above. Repartitioning disk0s2 and reinstalling the OS solved the problem.
@Link Davis: The SSD has not failed, if so the SMART status would have shown that. That's the most shared oppinion on many Apple related forums. I think you mean the disk, not the drive.
I'll not encrypt the SSD of the MacBook Air with the FileVault 2 until the bug is going to be fixed!

Server Admin runs very slow on Server 10.6.2

After what only could be called a power disaster that brought our XServe down hard, when I run Server Admin on the XServe, it's dog slow...it works, but it takes forever to do anything. However, Server Tools installed on my Macbook runs it just as fast as ever. Does anyone have any suggestions besides reinstalling? As our server is still in the set up stages, a backup of it hadn't been made yet...
Thoughts anyone?

The usual triggers for this (mis)behavior are a bad local DNS configuration within whatever is serving your local DNS, and potentially a firewall issue within the client.
If you've not messed around with your client firewall settings since this last worked, I'll bet on scrozzled DNS.
Verify your forward and reverse DNS translations and the correctness of your local DNS server set-up using the dig tool.
[My DNS notes|http://labs.hoffmanlabs.com/node/1436]; the set-up and verification of DNS on Snow Leopard Server.
As for auxiliary power, APC SMART-UPS 1500 series is a reasonable option for a less-interruptable power supply (LIPS) for an Xserve box and its usual array of companion direct-attached accouterments. And the USB connection from this model is recognized and honored by Mac OS X Server without add-on software; the box will shut down. (I'd look at adding LIPS capacity if you have a FC SAN and Xserve RAID or such, though.)

Server admin beachballing like crazy

I just installed a clean 10.5 Server onto a secondary drive in our 2x867ghz MDD. This server doesn't do much, mostly Software Updates and Netboot. I set up with an advanced config and got everything running. The problem this time around is that Server Admin takes two minutes of beachballing to get to a different config screen. If I want to check on Software Update Server, it takes two minutes. I can check around within the tabs just fine, but trying to check the log or switching to any service outside of it means another extended wait. The little gear wheel in the lower right spins around as well. I fail to see what I could have done to cause this. It was like this right out of the box, so to speak. There's plenty of free RAM, too, and all updates have been run. Checking the console shows A LOT of this:
2/5/08 8:27:42 AM com.apple.launchd[1] (com.apple.dyld) Throttling respawn: Will start in 60 seconds
I recently ran this thing through a diagnostic wringer and didn't find any problems. Tiger Server has been running just fine. This is my second time trying to get Leopard Server to work, and it's going about as well as the first, just with different inexplicable bugs this time.
Oh, and SUS STILL doesn't give you any indication of it's download progress. It just sits there until it's finished. Thanks for that one, Apple. A progress bar of some kind would've taken minimal effort.

The solution for me was to fix the reverse DNS.
The server that provided the reverse DNS for the IP addresses for our server was down. This caused slow behavior with both Server Admin and Remote Access.
I'd call this an Apple bug. The applications should not stop simply because they can't do a reverse lookup.

All logins slow/failed, even local server admin won't work...

I've had an issue last Friday which reoccurred this morning concerning logins. Any logins on the network don't work, but take forever to come up with an error message. So I went to the server locally and tried the admin account there, but it hung as well (pinwheel for a few minutes, then spinning gear indefinitely). On Friday, a hard reboot seemed to fix the issue. Today, after hard rebooting, I could log in locally, but upon opening Server Admin, my server's info never came up (showed connected, but no info came up (in the software/hardware/etc area on the first page) and no services were listed. No network accounts could log in at this time (would take a minute and then the generic "Cannot log into this account due to an error").
Did another restart (via the apple menu this time) and everything is working fine for now. I couldn't find anything in any logs to indicate a problem (although it seems several logs restarted with the server restart).
Most clients are 10.6.2, a few are 10.5.8 and one is 10.4.11.

Happened again this morning. Hard restart seemed to fix it, but the RAID with the home folders on it took (3-drive RAID in the XServe, OS is on a separate solid state drive) about 5 minutes to mount.
Most logs seemed to have reset on the restart. There was no server system log at all or a DCHP system.log.
OD LDAP log's first line was "nse error = 2". Everything after that was after the restart.
OD Password Service Error log was full of "Unknown listener exception error." All entries were since the restart.
Couldn't find any other anomalies in any logs (although, like I said, most of them don't seem to exist anymore.
Now, I don't know if it's related, but the green light on the center RAID drive is off, but it seems to check out fine in System Profiler and Disk Utility. The blue access light works. I just figured it's a bad LED, but seems possibly related if the RAID's losing connection somehow (which could explain logins not working and its slow mounting behavior).
Any ideas would be greatly appreciated as this is really bringing us to a standstill.

Server Admin Graphs

I'm not sure if this is the correct thread to post this issue in, but maybe someone out there could help me out. My graphs for overall CPU usage and Network Throughput in Server Admin have stopped functioning properly. The other night they started showing flat CPU performance at 75% and network performance at about 250 bytes/sec. Other tools (iStat Menus and iStat Server connected to my iPhone) showed correct information. Today however, the graphs have stopped showing any data at all. They both say "There is no data in the selected period". I noticed the odd behavior when my VPN stopped working intermittently. Any suggestions?

Hey Guys, i have the same problem.
Any help is apreciated!
this is something i got out of the error log, i cant figure it out how to solve this:
[Mon Nov 23 21:49:58 2009] [warn] RSA server certificate CommonName (CN) `myserver.example.com' does NOT match server name!?

Server Admin relay host field bug?

I noticed something curious when attempting to specify a relay host via Server Admin 10.4.3 (157.5) application.
If I specify xxx.com in "Relay outgoing mail through host" field, the value will appear as entered in MailServicesOther.plist. However, it will NOT appear in main.cf, there is only a line 'relay='.
When I manually edited main.cf to add 'relay=xxx.com' then relaying started working...however, if I use Server Admin after that to edit the value, it reverts back to an empty value again. Rather annoying, really.
Is this a known bug?

Honestly, I've had trouble with ServerAdmin and Postfix's main.cf from the beginning of 10.3.x I've given up on using SA for anything to do with the mail. I've tried everything from having someone send me a default main.cf and reseting everything in SA.
I can't get it back into sync, and since my backup MX server is running Postfix on FC3 I learned the CLI and got over the SA. I still use SA for looking at Account Quotas under the Maintenance tab, everything else is dorked anyway now.
Xserve G4 1ghz Mac OS X (10.4.3) OSXS 10.4.3

Safari and Server Admin slow to launch on a Network Home Directory

I have an issue with Safari and Server Admin that started two days ago. When you attempt to launch Safari or Server Admin while logged into a Network Home Directory on an Intel Mac it will bounce about 10 times, stop bouncing, and then about a minute later launch and act normal. If you quit and reopen Safari you have to wait again. Server Admin exhibits similar behavior, although the delay is longer and there are long pauses while using the program. The network home folder is on an afp share on an Intel XServe w/ 10.5.5 and 4gb of ram.
Troubleshooting steps already done:
1. Same intel mac with a local user, Safari and Server Admin launches normally.
2. Same intel mac with a portable home directory user, Safari and Server Admin launches normally.
3. Same intel mac with a different network home directory user, Safari and Server Admin have the long delay.
4. Different intel mac with a network home directory user, Safari and Server Admin have the long delay.
5. Different intel mac with a local user, Safari and Server Admin launches normally.
6. Different intel mac with a portable home directory user, Safari and Server Admin launches normally.
7. PPC mac with same network home directory user, Safari and Server Admin launches normally.
8. Checked forward and reverse DNS lookups for the home directory servers.
9. Checked entire DNS zone for forward and reverse lookups.
10. These exact same symptoms began occurring the same day at another site I manage with a separate DNS and Open Directory structure in another city.
11. All other programs appear to behave normally as far as I've been able to tell.
I have the feeling that something strange is going on with DNS and Server Admin and Safari are trying to resolve something, time out, then try to act normally. I don't know why it would just affect intel macs. In addition, there were no setting changes or DNS alterations that happened that I am aware of. These symptoms are also happening at another site that I manage with a completely separate DNS and Open Directory structure. I'm puzzled, any help would be appreciated.

Upgraded to 10.5.6 and the issue disappeared. I don't know what what wrong, but I think it was DNS related...

Annoying behavior of Server Admin

Similar Messages

Maybe you are looking for