Time to search PATH varies by HOME directory path ?

I have a rather strange question, as I am running in a somewhat strange environment. I have a SunBlade running Solaris connected via Cisco VPN to my corporate network. After I start the VPN I run a number of mount commands to make my local environment look something like my environment in the office, including mounting my home directory onto my local workstation. I then change the value of the HOME variable to point to the nfs mount for my home directory and update the PATH so that my local (non-remote) bin directories are before the ones on my remote mounts (I use Solaris at work as well).
I've been running this way for a couple of years and it works well, but all of a sudden trying to start an application from my local system started to take a really long time. I narrowed it down to:
if I set HOME=/export/home/localuser it works fine
if I set HOME=/export/home/remoteuser (which is where my home directory is mounted) it does not
It is not the PATH itself as I have the problem even when i don't change it at all. I have a copy of dtpad in /export/home/localuser/bin and that path is the first value in the PATH variable. With HOME set the first way "time which dtpad" takes 0.269s and with it set the second way it takes 4.530s.
This does not make any sense to me. Can someone that understands how the PATH searching algorithm works shed some light on this ?
Thanks

One more thing. I delete the copy of dtpad in my home directory and let it find the real one using the PATH. With HOME set to the local directory it took 0.527s to find it. With HOME set to the remote directory it took 2.139s. Same PATH in both cases. So how does $HOME fit into the $PATH search ?

Similar Messages

Time Machine not backing up my home directory

One day, I came in and booted up my Mac. I began my usual practice of pairing my Bluetooth mouse & keyboard, but this time, it failed to pair. The Bluetooth prefPane was acting weird (beach-balling, garbage characters appearing in the "is the name Bluetooth devices use to connect to this computer" title). I rebooted, and much to my dismay, Bluetooth was no longer detected by the system (both in System Profiler and System Preferences).
I assumed this meant the Bluetooth module was failing. I contacted Apple support, and we concluded that it was in fact the Bluetooth module that was failing, and no the connector cable. I filed a GSX ticket to have the entire display assembly replaced (since that's how these new aluminum uni-body MacBook Pros are built). The Bluetooth module has been failing intermittently, so I cancelled the GSX ticket to do more research (and to make sure it wasn't a software problem with my OS). I planned on doing an Archive & install to see if a new OS fixed it.
I ran a Time Machine backup to make sure I had the latest and greatest copy of my data, and ran an Archive & Install. As expected, the Bluetooth behaved identically in the clean OS, so I rebooted again, and ran a Restore from Time Machine, selected the most recent snapshot I had just created. After the restore completed, the machine rebooted, and an old desktop & Dock greeted me, much to my dismay. I checked and double-checked, and determined that Time Machine had been maintaining a week old copy of my home directory files, but had been continuing to update the rest of my computer (so my Applications folder was up-to-date).
I don't know how Time Machine could think that my Home directory hadn't changed in a week, but now I'm running an extended S.M.A.R.T. scan on the drive to make sure it's not failing. And yes, I made sure my Home directory was not in Time Machine's exclusion list. The reason I think the drive is failing is because when backupd tried to run after I restored from Time Machine, it ground a whole lot, and took forever to backup 90 MB. Like, a lot of grinding to get from 30 KB to 70 KB.
Does anyone know why/how this might've happened?
Message was edited by: Link Dupont
Message was edited by: Link Dupont

Link Dupont wrote:
I'd plug in and run a Time Machine backup about once a day, on average. I've been scouring the logs, and the only thing I notice is the "deep traversal" message, which I'd expect occasionally as it needs to reindex files (especially after a restore).
This also happens if the TM disk isn't dismounted, etc., exactly the way TM expects; or if there's an abnormal power-down. It means TM isn't sure about just exactly what is where, so it looks at every file and folder to be sure.
I've run long S.M.A.R.T. scans on the drive (via smartctl on a Linux LiveCD and SeaTools via a Windows installation), I've run Repair Disk (via Disk Utility) and I've defragmented the filesystem (iDefrag). All the diagnostics come back healthy, and iDefrag had little to do. My gut tells me the drive is failing, but I can't actually prove it. My next step is to reformat the drive and create a new filesystem; I'll see if Time Machine is happy with a clean start.
I'm a MobileMe subscriber, so I downloaded and set up the old .Mac Backup.app, and will be running that daily to shove my important files up to my iDisk (and probably to various other locations), in case backupd decides to choke again, at least I have a backup to my backup.
Good plan. I run CCC daily, in addition to TM's hourly backups, and use Backup to iDisk daily for fairly small stuff, plus others to DVD/RWs kept in my safe deposit box. May seem like overkill, but I've learned, usually the hard way, that when it comes to backups, paranoia is prudent!

Time Machine always switches to my home directory. What to do?

D'oh, need to restore a file, which happens to be in a root folder of my iMac's HD.   No prob, Time Machine is on duty!
The usual procedure: In Finder, bring the containing folder to the front, select the file, go to the clock icon at he upper right, pull down, and choose "Enter Time Machine". TM takes over the display with the distinctive Star-Field background and offers to restore in that folder.
Up until today that that's worked in every version of MacOS that includes Time Machine.
Today: No matter what I do --including after a system restart-- Time Machine switches the front-most window to display my home directory, first, THEN goes into Star-Field mode.
Note 1: Yeah, I can work-around by navigating to my HD and then to the specific file I need to restore.
Note 2: This problem occurs even when the file I need to restore is in the hierarchy of my home directory.
So: Why does TM now insist on switching to my home directory and what to do about it?   Without endangering my existing files or TM backups?
TIA

Link Dupont wrote:
I'd plug in and run a Time Machine backup about once a day, on average. I've been scouring the logs, and the only thing I notice is the "deep traversal" message, which I'd expect occasionally as it needs to reindex files (especially after a restore).
This also happens if the TM disk isn't dismounted, etc., exactly the way TM expects; or if there's an abnormal power-down. It means TM isn't sure about just exactly what is where, so it looks at every file and folder to be sure.
I've run long S.M.A.R.T. scans on the drive (via smartctl on a Linux LiveCD and SeaTools via a Windows installation), I've run Repair Disk (via Disk Utility) and I've defragmented the filesystem (iDefrag). All the diagnostics come back healthy, and iDefrag had little to do. My gut tells me the drive is failing, but I can't actually prove it. My next step is to reformat the drive and create a new filesystem; I'll see if Time Machine is happy with a clean start.
I'm a MobileMe subscriber, so I downloaded and set up the old .Mac Backup.app, and will be running that daily to shove my important files up to my iDisk (and probably to various other locations), in case backupd decides to choke again, at least I have a backup to my backup.
Good plan. I run CCC daily, in addition to TM's hourly backups, and use Backup to iDisk daily for fairly small stuff, plus others to DVD/RWs kept in my safe deposit box. May seem like overkill, but I've learned, usually the hard way, that when it comes to backups, paranoia is prudent!

Multiple remote home directory users logged into one machine?

Hello,
Just wondering if it's possible to have two or more users logged into the same machine at the same time. Each user has a home directory located on an XServe RAID connected to an XServe.
We have a new employee who is needing to toggle back and forth between his account and an account that belongs to a previous employee.
So far I've only had success logging in one remote home directory user, and a user account local to the machine. Making.. two users logged in on the same machine at the same time.
Anyone able to offer some insight? thanks!

The way afp works, you won't be able to log in more than one user with a remote home directory on any one machine at a time. This is a known limitation.
If you need access to more than one account at a time, you could copy the home directory of the old user onto the local machine and set up a local account with it. Or you could give the new user a copy of the old home directory if it's just a matter of file access.
Hope this helps,
Andrew

Change username/home directory

Hi,
I want to change my username and home directory. My Mac was gifted to me and while setting it up for me, they mispelt my name. I have been using it that way for quite some time, and now I want to change it. I am following this article: http://support.apple.com/kb/ht1428 and it seems quite easy.
If I rename the home directory, will Time Machine back up all stuff in my home directory once again? My home directory is quite huge ~70% of my hard disk and so I will lose a lot of old back ups due to this. Any workarounds?
Thanks.

tartarsauce wrote:
If I rename the home directory, will Time Machine back up all stuff in my home directory once again? My home directory is quite huge ~70% of my hard disk and so I will lose a lot of old back ups due to this. Any workarounds?
Yes. Time Machine will back up your home directory again.
Your choices are:
1) Leave the name as is, and you won't have to worry about the large back up for a while, or
2) If you are out of space on your current backup disk, buy another external disk, rename the user and home directory on your Mac and start a new round of backups, keeping the old one around until you no longer need the files,
3) Rename the home directory and let Time Machine delete the oldest of your backed up files as it runs out of disk space.
Have I missed any?

User home directory at /var/imap is unavailable

We're running a single 10.5.7 server and recently migrated our mail to a new location from the default /var/imap and var/spool/imap. Since that time, I've been seeing the following log entries pop up when a user attempts to log in to the mail server. An example:
Feb 5 10:19:37 ServerName imapd: [xxx.xxx.xxx.xxx] username
[6608]: CFPreferences: user home directory at /var/imap is unavailable. User domains will be volatile.
We migrated the mail using a process nearly identical to the one found at:
http://discussions.apple.com/thread.jspa?messageID=6600016&#6600016
I have checked the settings and paths in:
cyrus.conf
imapd.conf
(postfix) main.cf
(postfix) master.cf
All of the paths point to the correct new paths, which are:
/Volumes/Resources/Mail/imap
/Volumes/Resources/Mail/spool/imap
All users are able to access their accounts and use the mail server. What we're seeing is periodic periods of high latency (messages taking minutes to send) and occasionally odd problems with users' Sent folders (a sent message appearing twice with the same timestamp, even though the outgoing mail filter indicates it was only sent once). I'm making the assumption that these intermittent problems are related to the user home pointing to the wrong path.
I've looked through the forums and googled the phrase, but haven't found anything that would point me toward finding the setting in the mail services that remains incorrect.
We also intermittently receive the following mail-related message in the mailaccess.log:
Feb 5 10:36:07 Rivendell master[865]: can't open com file: /var/imap/.smd.imap.com (No such file or directory)
This appears to be a similar problem, but again, the location of the setting I need to change eludes me.
Thanks for any help or pointers.

Here's the solution that finally fixed this problem and removed latency problems that seem to have been associated with it.
(1) Stopped mail services with terminal.
(2) Opened the Server Admin application.
(3) Under Mail/Settings/Advanced/Database I changed the path to the original location, /var/imap and /var/spool/imap.
(4) Clicked "Save."
(5) In the same spot, changed the paths back to the new location.
(6) Clicked "Save."
(7) Restarted the mail server.
I have no idea what hidden setting that's not in the conf files that this changed, but it removed the error messages and the latency.

Forte WS6U1 - Search path SunWS_Cache directory needs write permission

Forte 6 U1, needs write permission to search path SunWS_Cache directory.
When I build a file locally checked out from the baseline directory, and the baseline SunWS_Cache directory needs write permission.
Lets say,
/home/zeebra/baseline/util - is the directory where the baseline is.
Having local workarea as /home/user/work/util
The local work has the baseline in search path. But the local -user does not have write permission on baseline/util/SunWS_Cache.
The local compilation goes thro fine, but when linking it just hangs. As soon as I give the write permission to baseline/util/SunWS_Cache, the linking completes.
Is it a bug? Do we have patch already for this problem for Forte6 Update 1?
Thanks!

The SunWS_Cache directory must have write permission. The cache contains state information that must be locked when a compiler is being run, to avoid trashing the state information if another compilation runs in the same directory at the same time. (Sun C++ supports dmake and gmake for running multiple compilations in parallel.)
A non-writeable template cache is not an available option.
Starting with C++ 5.5 (Sun ONE Studio 8 Compiler Collection), the template cache is no longer required, and by default is not used. If you are having problems involving the cache, you should consider upgrading to this new release.

Is there a file in which OSX stores the home directory path, where to find it and how to edit it in Windows 7?

Hi everyone,
I have a SSD with the System and Apps on it, and another HDD containing the home directory.
I made the mistake to rename the Volume the home directory is on whilst being logged in, the system reacted accordingly and prompted some Password requests,
and I then tried to choose the newly named Volume in the Advanced options dialogue in Users & Groups (example seen below).
I tried to change the name back to what it was in the hope of it all being back to normal again, but instead it's completely messed up now and I cannot login anymore (prompt: "You are unable to log in to the user account "xxxx" at this time. Logging in to the account failed because an error occurred")
All I can do now is boot my Windows 7 and try to fix this from there, I can read/write the OSX Volumes thanks to MacDrive.
So now the big question is:
IS THERE A FILE containing the path to the home directory (red arrow)?
IF SO, WHERE can I find it?
And is there a way to edit it in Windows?
My guess is that if I rename the Volume, AND change this path to something identical again, it might solve the issue.
Thanks!

Download THIS
 Cheers, Tom

WGM Home directory full path not populating properly

I'm setting up WGM with mobility as all of my network users are on laptops. For some reason some of the users are working and others not. When I try to log in with some I just get told that the user couldn't log on because an error occured. As far as I can work out this is because the home directory for the user on the network can't be found and so a local one can't be created either.
The only difference I can see between the users (bearing in mind this is a brand new network set up so I haven't made many changes) is that for the users that do work when I select the SharePoint I want to use under the home tab in user settings, the Full Path has ...er ...well the full path in it: /network/servers/mynetwork/Volumes/myharddrive/myfolder/myuser
whereas ones that don't work just have the folder name: myuser
This not only happens when I use the auto populated list as well as when I create one from scratch.

You're right, DNS implementation in Leopard server is very poor, especially considering how critical it is.
Time Machine for every user is a bit of a tricky one - I've been thrashing this one out in another thread, and there is an answer, but it's not ideal.
Here's the thread: http://discussions.apple.com/thread.jspa?threadID=2034769&start=0&tstart=0
The answer on how to do it is at the end of the thread, but make sure Time Machine isn't used as the only method of Disaster Recovery on a corporate network, it's really only good for a useful extra.
As for an internal and external DNS server, the answer I'm afraid is "not really". The way you have it set up is pretty good, from a redundancy point of view. However, if you have to use an external DNS server, it's always going to be slower, because of the extra distance over slower links that the DNS request has to travel.
Bear in mind, it has to go through your router and/or firewall, all the way out to a busy DNS server somewhere out on the net (probably over a number of routers on the way), then it has to come back the same way before your client can actually resolve the URL to an IP address.
Compare that to internal where it has to travel a matter of metres over a (I assume) 100mbps network to a DNS server that's not processing very many requests.
The only workaround I could suggest would be to employ a second DNS server (perhaps a proxy?) within your network assuming you can separate it from whatever causes your primary DNS server to go down. The costs aren't prohibitive - you could do it on a Mac Mini + cost of a second Leopard Server license.
Oh, and your English is just fine, lol.
Message was edited by: MattLucas1505

Open Directory, third party LDAP search path problem on Snow Leopard

Happy new year folks,
I ran into an interesting problem this past week in regards to a third party LDAP directory in the Search path (which used to work on previous versions). The issue brings the server to its knees eventually. I'm still digging through the logs, but here's the general breakdown...
1. Add third-party LDAP to the OD node list. This has always worked on previous versions, and appears to still work at the most basic level. I can navigate the node with DSCL, read records, etc.
1. Add third-party LDAP to the OD search path.
2. Wait a few minutes....
3. The server begins to slow down. Apache, SSH, ServerAdmin service stop responding. I'm able to run "top" briefly, which shows an increase of threads.
4. Restart the server and quickly remove the directory from the OD search path
5. Server goes back to being rock solid with very nice response times for Apache, SSH, ServerAdmin, etc.
If anyone has any debugging suggestions, or has seen this before, let me know.
Jaime
--- Below is some console output leading up to the chaos. Before adding to search path, everything looks good --------------------
bash-3.2# dscl
Entering interactive mode... (type "help" for commands)
read /LDAPv3/ldap.itd.umich.edu/Users/jaimelm cn
dsAttrTypeNative:cn:
Jaime Magiera
Jaime L Magiera 1
Jaime L Magiera
--- Add to Search Path, which hangs ------------------------------------------------------------------------------
bash-3.2# dscl /Search -append / CSPSearchPath /LDAPv3/ldap.itd.umich.edu
--- DSCL in debug mode contains the following ----------------------------------------------
2010-01-01 19:26:25 EST - T[0x00000001037A5000] - Client: ipfw, PID: 1097, API: libinfo, Server Used : libinfomig DAR : Procedure = getprotobynumber (13) : Result code = 0
2010-01-01 19:26:25 EST - T[0x00000001037A5000] - Client: sso_util, PID: 1103, API: dsFindDirNodes(), Server Used : DAR : 1 : Dir Ref = 16779669 : Requested nodename = /Search
2010-01-01 19:26:25 EST - T[0x00000001037A5000] - Plug-in call "dsDoPlugInCustomCall()" failed with error = -14292.
2010-01-01 19:26:25 EST - T[0x00000001037A5000] - Port: 27151 Call: dsDoPlugInCustomCall() == -14292
2010-01-01 19:26:36 EST - T[0x00000001037A5000] - Client: dscl, PID: 1114, API: dsFindDirNodes(), Server Used : DAR : 1 : Dir Ref = 16779
707 : Requested nodename = /LDAPv3/ldap.itd.umich.edu
2010-01-01 19:26:36 EST - T[0x00000001037A5000] - Client: dscl, PID: 1114, API: dsFindDirNodes(), Server Used : DAR : 2 : Dir Ref = 16779707 : Result code = 0
2010-01-01 19:26:36 EST - T[0x00000001037A5000] - Client: dscl, PID: 1114, API: dsVerifyDirRefNum(), Server Used : DAC : Dir Ref 167797072010-01-01 19:26:36 EST - T[0x00000001037A5000] - Client: dscl, PID: 1114, API: dsVerifyDirRefNum(), Server Used : DAR : Dir Ref 16779707
: Result code = 0
2010-01-01 19:26:36 EST - T[0x00000001037A5000] - Client: dscl, PID: 1114, API: dsFindDirNodes(), Server Used : DAC : Dir Ref 16779707 :
Data buffer size = 1282010-01-01 19:26:36 EST - T[0x00000001037A5000] - Client: dscl, PID: 1114, API: dsFindDirNodes(), Server Used : DAR : 1 : Dir Ref = 16779
707 : Requested nodename = ConfigNode2010-01-01 19:26:36 EST - T[0x00000001037A5000] - Client: dscl, PID: 1114, API: dsFindDirNodes(), Server Used : DAR : 2 : Dir Ref = 16779
707 : Result code = 0
2010-01-01 19:26:36 EST - T[0x00000001037A5000] - Client: Requesting dsOpenDirNode with PID = 1114, UID = 0, and EUID = 0
2010-01-01 19:26:36 EST - T[0x00000001037A5000] - Client: dscl, PID: 1114, API: dsOpenDirNode(), Configure Used : DAC : Dir Ref = 16779707 : Node Name = /Configure
2010-01-01 19:26:36 EST - T[0x00000001037A5000] - Client: dscl, PID: 1114, API: dsOpenDirNode(), Configure Used : DAR : Dir Ref = 1677970
7 : Node Ref = 33556926 : Result code = 0
2010-01-01 19:26:36 EST - T[0x00000001037A5000] - Client: dscl, PID: 1114, API: dsVerifyDirRefNum(), Server Used : DAC : Dir Ref 16779707
2010-01-01 19:26:36 EST - T[0x00000001037A5000] - Client: dscl, PID: 1114, API: dsVerifyDirRefNum(), Server Used : DAR : Dir Ref 16779707 : Result code = 0
2010-01-01 19:26:36 EST - T[0x00000001037A5000] - Client: dscl, PID: 1114, API: dsGetDirNodeInfo(), Configure Used : DAC : Node Ref = 33556926 : Requested Attrs = dsAttrTypeStandard:OperatingSystemVersion : Attr Type Only Flag = 0
2010-01-01 19:26:36 EST - T[0x00000001037A5000] - Client: dscl, PID: 1114, API: dsGetDirNodeInfo(), Configure Used : DAR : Node Ref = 33556926 : Result code = 0
2010-01-01 19:26:36 EST - T[0x00000001037A5000] - Client: dscl, PID: 1114, API: dsGetDirNodeInfo(), Search Used : DAC : Node Ref = 33556924 : Requested Attrs = dsAttrTypeStandard:LSPSearchPath : Attr Type Only Flag = 0
2010-01-01 19:26:36 EST - T[0x00000001037A5000] - Client: dscl, PID: 1114, API: dsGetDirNodeInfo(), Search Used : DAR : Node Ref = 33556924 : Result code = 0
2010-01-01 19:26:36 EST - T[0x00000001037A5000] - Client: dscl, PID: 1114, API: dsDoPlugInCustomCall(), Search Used : DAC : Node Ref = 33556924 : Request Code = 444
2010-01-01 19:26:36 EST - T[0x00000001037A5000] - Checking for Search Node XML config file:
2010-01-01 19:26:36 EST - T[0x00000001037A5000] - /Library/Preferences/DirectoryService/SearchNodeConfig.plist
2010-01-01 19:26:36 EST - T[0x00000001037A5000] - Have written the Search Node XML config file:
2010-01-01 19:26:36 EST - T[0x00000001037A5000] - /Library/Preferences/DirectoryService/SearchNodeConfigBackup.plist
2010-01-01 19:26:36 EST - T[0x00000001037A5000] - Setting search policy to Custom search
2010-01-01 19:26:36 EST - T[0x00000001037A5000] - CSearchPlugin::SwitchSearchPolicy: switch - reachability of node </LDAPv3/127.0.0.1> retained as <true>
2010-01-01 19:26:36 EST - T[0x000000010070A000] - CSearchPlugin::CheckNodes: checking network node reachability on search policy 0x0000000000002201
2010-01-01 19:26:36 EST - T[0x00000001037A5000] - CCachePlugin::EmptyCacheEntryType - Request to empty all types - Flushing the cache
2010-01-01 19:26:36 EST - T[0x000000010070A000] - Client: Requesting dsOpenDirNode with PID = 0, UID = 0, and EUID = 0
2010-01-01 19:26:36 EST - T[0x000000010070A000] - Internal Dispatch, API: dsOpenDirNode(), LDAPv3 Used : DAC : Dir Ref = 16777216 : Node Name = /LDAPv3/127.0.0.1
2010-01-01 19:26:36 EST - T[0x000000010070A000] - Internal Dispatch, API: dsOpenDirNode(), LDAPv3 Used : DAR : Dir Ref = 16777216 : Node Ref = 33556929 : Result code = 0
2010-01-01 19:26:36 EST - T[0x000000010070A000] - CSearchPlugin::CheckNodes: calling dsOpenDirNode succeeded on node </LDAPv3/127.0.0.1>
2010-01-01 19:26:36 EST - T[0x000000010070A000] - Internal Dispatch, API: dsCloseDirNode(), LDAPv3 Used : DAC : Node Ref = 33556929
2010-01-01 19:26:36 EST - T[0x000000010070A000] - Internal Dispatch, API: dsCloseDirNode(), LDAPv3 Used : DAR : Node Ref = 33556929 : Result code = 0
2010-01-01 19:26:36 EST - T[0x0000000103181000] - mbr_mig - dsFlushMembershipCache - force cache flush (internally initiated)
2010-01-01 19:26:36 EST - T[0x000000010070A000] - Client: Requesting dsOpenDirNode with PID = 0, UID = 0, and EUID = 0
2010-01-01 19:26:36 EST - T[0x0000000103181000] - Membership - dsNodeStateChangeOccurred - flagging all entries as expired
2010-01-01 19:26:36 EST - T[0x000000010070A000] - Internal Dispatch, API: dsOpenDirNode(), LDAPv3 Used : DAC : Dir Ref = 16777216 : Node Name = /LDAPv3/ldap.itd.umich.edu
2010-01-01 19:26:36 EST - T[0x000000010070A000] - CLDAPNodeConfig::InternalEstablishConnection - Node ldap.itd.umich.edu - Connection requested for read
2010-01-01 19:26:36 EST - T[0x000000010070A000] - CLDAPNodeConfig::FindSuitableReplica - Node ldap.itd.umich.edu - Attempting Replica connect to 141.211.93.133 for read
2010-01-01 19:26:36 EST - T[0x0000000102481000] - CCachePlugin::SearchPolicyChange - search policy change notification, looking for NIS
2010-01-01 19:26:36 EST - T[0x0000000102481000] - Internal Dispatch, API: dsGetDirNodeInfo(), Search Used : DAC : Node Ref = 33554436 : Requested Attrs = dsAttrTypeStandard:SearchPath : Attr Type Only Flag = 0
------- From another screen, I do "id jaimelm", which hangs ------------------------------------------------------------------------
: Requested Rec Names = jaimelm : Rec Name Pattern Match:8449 = eDSiExact : Requested Rec Types = dsRecTypeStandard:Users
2010-01-01 19:36:55 EST - T[0x00000001082A2000] - Internal Dispatch, API: dsGetRecordList(), Search Used : DAC : 2 : Node Ref = 33554436 : Requested Attrs = dsAttrTypeStandard:AppleMetaNodeLocation;dsAttrTypeStandard:RecordName;dsAttrTy peStandard:Password;dsAttrTypeStandard:UniqueID;dsAttrTypeStandard:GeneratedUID; dsAttrTypeStandard:PrimaryGroupID;dsAttrTypeStandard:NFSHomeDirectory;dsAttrType Standard:UserShell;dsAttrTypeStandard:RealName;dsAttrTypeStandard:Keywords : Attr Type Only Flag = 0 : Record Count Limit = 1 : Continue Data = 0
2010-01-01 19:37:03 EST - T[0x0000000108325000] - Client: httpd, PID: 157, API: mbr_syscall, Server Used : process kauth result 0x0000000102022B30
2010-01-01 19:37:03 EST - T[0x00000001083A8000] - Client: httpd, PID: 151, API: mbr_syscall, Server Used : process kauth result 0x0000000102022C50
2010-01-01 19:37:05 EST - T[0x000000010842B000] - Client: httpd, PID: 203, API: mbr_syscall, Server Used : process kauth result 0x0000000102022D70
2010-01-01 19:37:15 EST - T[0x00000001084AE000] - Client: httpd, PID: 994, API: mbr_syscall, Server Used : process kauth result 0x0000000102023890
2010-01-01 19:37:26 EST - T[0x0000000108531000] - Client: httpd, PID: 198, API: mbr_syscall, Server Used : process kauth result 0x0000000102023980
2010-01-01 19:37:31 EST - T[0x00000001085B4000] - Client: httpd, PID: 161, API: mbr_syscall, Server Used : process kauth result 0x0000000~

Hi
I'm in agreement with harry here but what I'm struggling to understand is why you are seeing this as a problem? I'm also struggling to see this as being a possibility in a single server environment if I understand your post correctly?
Promotion to OD Master with all that entails absolutely rests on a properly configured and tested internal DNS Service. The Kerberos Realm's foundation (and with that the ability of the server to perform its function as KDC and offer LDAP services) entirely depends on what is configured in the DNS Service. This will include the server name, domain name and tld. The Kerberos Realm automatically configures itself using that information. Likewise the searchbase.
Its more than possible to change the Realm name and with it the LDAP search base (in certain circumstances) and have an OD Master, however Kerberos won't start it won't need to as the KDC will be elsewhere. You generally see this when augmenting Windows AD with MCX. In that situation Realm name and search base will reflect what is set on the Active Directory. Client computers will use what is set there for contact and authentication information before looking at the OD Master for anything else.
Does this help? Tony

Home directory is not being backed up in Time Machine

Problem: It turns out my home directory does not have any backups in Time Machine for the last year. I've never received any error message indicating it wasn't doing the full backup.
Goal: I'd like to get Time Machine to back up my home folder (which contains 100% of the content I care about) again, ideally in a way where I can keep my history from the last time it backed up successfully.
How I found out:
- I opened Time Machine and realized all the "pink bars" were dimmed all the way back up to Sept. 2011 when I navigated to my home folder or any folder within it.
- I cloned a file within my home folder, did a manual Time Machine back up, opened Time Machine and verified that the backup volume did not contain the cloned file.
- I did this again while watching the Console and verified that it did not make a large enough backup to include the cloned file.
- I explored all the backups in Finder and realized that indeed all the backups that corresponded to the dimmed pink bars in Time Machine had no home folder.
- I did an "ls -a" in the backup volume in Terminal and realized that it was not the case that the home directory was present but merely hidden.
What I've tried:
- I created a new admin account, logged into that and did a manual Time Machine backup from there. The new account was backed up, while the original home directory still wasn't. This continues to be true in subsequent backups made while logged into the main account.
- I examined the exclusions list in Time Machine preferences. The only items were the backup volume itself and another partition of that external drive I use for other backups, as expected.
- I did a "Full Reset of Time Machine" (http://pondini.org/TM/A4.html), with a reboot before re-enabling Time Machine. The plist file did change to become much simpler, but the home folder is still not being backed up.
Other notes:
- I run Dropbox, and have those contents in a folder one level under the home directory
- I run Git
- The existing backups have only used 500GB or so of a 1TB partition, so it's not filling up.
- I ran Verify Disk in Disk Utility on both my backup volume and the mac's HD. The backup volume appeared OK, but the mac HD was "found corrupt and needs to be repaired". I could boot into the Recovery HD and repair it, but I feel like this is a risk and I'd like to have a working backup before I do that (I don't have another external HD around to make a temporary backup). Could the corrupted drive be the reason Time Machine has been silently skipping the home folder?
- My home folder is named "apple". Might this be a problem? I'd rename it, but that's a risk and I'd like a verified backup before doing that.
- The Time Machine plist file in /Library/Preferences is pretty hard to read in pico, TextEdit, etc (maybe this is normal?)
I've scoured the forums and pondini's very thorough troubleshooting pages to no avail. Please let me know if anyone has ideas. Thanks.

Latest update: Bought a new external harddrive, set that up as the new Time Machine volume, but got the exact same results (all but my main user directory are backed up). This is after another Time Machine reset (deleting the .plist), starting up in Safe Mode, and trying to initiate the backup from both the account that doesn't get backed up and another administrator account.
Here is some of the output from the Console when filtering messages for "backupd":
10/5/12 2:51:23.572 AM com.apple.backupd[13529]: Backup content size: 481.57 GB excluded items size: 11.31 GB for volume Macintosh HD
10/5/12 2:51:23.572 AM com.apple.backupd[13529]: Found 1437613 files (470.25 GB) needing backup
10/5/12 2:51:23.573 AM com.apple.backupd[13529]: 564.3 GB required (including padding), 1.5 TB available
10/5/12 2:51:23.581 AM com.apple.backupd[13529]: Waiting for index to be ready (100)
10/5/12 3:40:33.297 AM com.apple.backupd[13529]: Copied 763432 files (21.54 GB) from volume Macintosh HD.
Put the relevant parts in bold. No messages about why it only backed up 21.54GB when it found 470.25GB that needed to get backed up.
Does this make sense to anyone??

Change NAME of User Account, Home Directory, Time Capsule and Network

Hi All.
I have a small home network and I would like to do this for all four of these.
I think I have instructions on doing this for User Account, Home Directory and Time Capsule.
Can anyone give me the "it is OK to do this" so I don't have to worry about something going wrong in the future?
Also, can I do this for my network and does anyone know where I find this?
Thanks.

hotwheels 22 wrote:
So I gather I cannot rename the name of the "Macintosh HD" which I guess is the name of my Hard Drive?
Technically, it's the name of the OSX volume or partition, but is often just called your internal hard drive, since most folks only have one partition. But some have multiple partitions, usually for dual booting, such as OSX and Windows; or Lion and Snow Leopard.
Yes, you can rename it easily (but you might check whether that will cause a problem with DejaVu). It won't be a problem for Time Machine.
Just right-click it in a Finder sidebar or on your desktop. Do not use any slashes or colons, however -- you might want to avoid any punctuation or special characters.
And then - In Accounts I have "Full Name" as literally my first and last name.
Yes. Each user account has a full name and a short name. The full name isn't used for much.
the HOME FOLDER is showing a totally different name than the "Full Name".
Correct. It's the same as the short name.
So I set the HOME FOLDER name through some hocus pocus in the article and then what is the relationship between these things?
The short name and the home folder name should be the same. The article shows how to rename the home folder, create a new user account with a matching name, then delete the old account.
If you're not comfortable with the instructions in the article, I'd strongly advise not doing it. You may end up having to do a full system restore. I don't know, but perhaps your local Apple Store will do it for you.
Functionally I have a HARD DRIVE and a HOME FOLDER and then the Full Name (listed in Accounts) is just sort of what I get called by...?
The hard drive (OSX partition) is where your entire Snow Leopard installation is: OSX itself, your Apple and 3rd-party apps, and one or more home folders. (Many folks who share a Mac have separate home folders for each user.)
have I been smoking?
No comment.

How To Find Oracle Home Directory Path...

Hello,
I am working on deployment of java web service on Oracle Application Server(LINUX) 10g, EBS - R12.
I am working on other system and i want to deploy web service remotely on application server.
And i am not getting the oracle home directory path. I want to find the home directory path remotely.
so suggest me how to find out path of that.
or weather i have to install jdevloper on server system and than i have to deploy it,
or any other alternative way is available.
Reply as soon as possible......

Thanks for your reply.
It's precisely the registry key which got tampered, and there's no ORACLE_HOME environmental variable(Or that too is tampered)
Now,how to set it to one of the Oracle home directories?
Regards,
Bhagat
!!!!!!!!!!!!!!!!!!!!

Does anyone know what the file "memeodhelper is? It's preventing me from backing up my home directory. I can't even find the file with search. Hidden somehow?

I would like to backup my Mac by copying my home directory to an external disk. When I try this, it always gets hung up on a file called "memeodhelper" saying I don't have permission to access the file and the paste stops. Can't find the file with a search. Any ideas?

Hi Ray,
Spotlight is no good for finding anything Apple thinks you don't need to find.
You can "fix" that error...
http://www.somewhereville.com/?tag=memeodhelper
Or likely trash it...
https://discussions.apple.com/message/11237019?messageID=11237019&tstart=0#11237 019?messageID=11237019&tstart=0
Find Any File...
http://apps.tempel.org/FindAnyFile/
Hold Option or alt key when selecting Find to Find All.
EasyFind...
http://www.devon-technologies.com/download/
Near the bottom of the page.

Lldb equivalent of gdb "directory" command for specifying source search path?

Looking for the lldb equivalent of the gdb "directory" command to add search paths for finding missing source code files (due to moved directories). Or possibly similar functionality within xcode...
Thanks in advance!

Great, thanks for the pointer although according to the lldb console the "add" command needs a "substitution pair" argument and the "Insert" command needs a new "image search path" and the "target modules list" command only gives a list of dylib modules. I played around a bit and the construction resembles more of a dylib module search path .i.e it looks like you only can revert the module search path but not separate it from the source path.
It would be nice to find out more details about different search-path capabilities within lldb. You don't happen to know where to find more thorough LLDB manuals or other useful documentation besides the "LLDB to GDB Command Map" (http://lldb.llvm.org/lldb-gdb.html) and the "LLDB Tutorial" (lldb.llvm.org/tutorial.html), do you? Also, any idea where to find the official doc pages for lldb. Unfortunately "http://lldb.llvm.org/docs.html" is still empty and I didn't find anything useful in the llvm github.

Time to search PATH varies by HOME directory path ?

Similar Messages

Maybe you are looking for