Editing & launching launchd agents with Objective C

I am working on an app that will allow the user to set an interval for launching a particular process. I have figured out how to use the Resources to hold templates of the .plist files and copy them to the appropriate folder, but how do I go about actually loading the agent after it is in place?
Additionally, I would also like to provide the option for the user to change the interval, which would require unloading the agent, re-writing the appropriate Dictionary item, and then re-loading the agent.
And yes, I am a rookie, but I am getting the hang of this OOP stuff, and appreciate any advice on where to look for this info.

orangekay:
Thanks, I am familiar with launchd and already use it for some personal stuff on my machine. I had not been able to find any direct interface between Cocoa and launchd (as you mentioned), but am hopeful that someone knows how to programmatically load and unload agents as part of a Cocoa application.
I have been reading up on NSTask, but am not sure if that is the way to go, and if so, how.

Similar Messages

LR4.1 : "Edit-In" "Open as Smart Object with Photoshop..." feature doesn't work with 32bits

Hi there,
I am running LR4.1/CS5/ACR6.7 on a Win7 64bits system.
The feature "Edit-In" > "Open as Smart Object with Photoshop..." does work Ok with raw files (.NEF).
However, when i do try to use the same feature with 32 bits .TIFF files (which are output of the LR4.1's "Edit-In" > "Merge to HDR Pro in Photoshop..." other feature), nothing does happen (the Photoshop application windows does open but w.o the image that i've just selected).
I've used search engines and Adobe online help to see if there was any limit preventing the usage of this feature with 32 bits .TIF files but couldn't see such note.
Thanks for your help !
Albert

Thanks Rikk,
That did work with a DNG file :-)
I am working with very huge .TIF files (400MB), so i am wondering if size could be the main issue (although my PC has lots of HW resources).
I will explore the DNG way, a format which seems anyways to be very promising now with LR4.
In case anyone has an idea why it doesn't work the .TIF files, please shoot !

Add/Edit Form with object

OK, I'm somewhat of a newbie, but I know enough to make me dangerous. Here's my problem.
I have a form that I want to use for both Adding a course and Editing an existing course. I have a Course object. My ideal situation is to check an "action" parameter when the page is loaded. If action == Add, then I would like to instantiate an object with default values, if action == Edit, then get the Course object that corresponds to the CourseID that is also passed in. The form values would be set to Course.getXXX(). That part is pretty basic.
When the form is submitted I want to validate the fields. If they are valid, then update the DB, otherwise, go back to the form with the fields restored to the users input plus error messages. Here's where I am getting tripped up.
First, I would like to save the Course object in the Request scope, so I check that first on the Form page. If a course already exists in the Request scope, then I use that Course object, otherwise I create a new Course object like I described in the first paragraph. The problem is really passing that on to the next page that does the validating. I was told to use a JSP tag and setProp="*" blah blah blah, then call a validate method. The question I have is what happens to Course fields that are set to datatypes such as long or timestamp ??? And what kind of validate method should I write? Should I pass it a long or String? Do I really have to write TWO validation methods for every field I have that isn't a String? What's the most common way to do this? And should I validate BEFORE I try to set properties? Maybe through a static method or something?
The other thought I had was to create a CourseView object in which everything field was a String, then I could validate, then if everything was fine and dandy, I could set or create a new Course object using CourseView and then update the DB. This would mean I would use the CourseView instead of Course for the forms. The problem I have is what's the best way to go about that? I'm not that familiar with abstraction, extending or implementing. What would work best? I would like to have the validation methods with the Course object and just utilize them in CourseView, but I'm sure there are loopholes to that.
Oh, and the other big problem...null values. How do forms treat null, how do validation methods treat nulls and what about objects that require some fields but not others? Or even worse yet, a field that is not required unless another field is not null. There has to be an answer for this already.
Oh yeah, Struts is out of the question for reasons to involved to go into here.
I thank everyone in advance and hope any response help many more people then just I.

Basically lets say I have this:
AddEditForm.jsp
<%@ import com.mycomp.Course, com.mycomp.CourseMgr %>
<%
String action = request.getParameter("action");
Course crs;
if (request.getAttribute("crs") != null){
//if crs exisits in request, then form was submitted and rejected
crs = request.getAttribute("crs");
//otherwise this is the first time the page has been loaded
} else if (action == add){
crs = new Course();
}else if (action == edit){
//course_id is a long
crs = CourseMgr.getCourse(course_id);
%>
<HTML>
<BODY>
<% Custom tag here to display errors from errorMsg hashtable %>
<FORM action="process.jsp">
<input type="text" name="course_id" value="<%=crs.getCourse_id()=%>">
<input type="text" name="status" value="<%=crs.getStatus()=%>">
<input type="text" name="created_by" value="<%=crs.getCreated_by()%>">
<input type="text" name="created_date" value="<%=crs.getCreated_date()%>">
</FORM>
</BODY>
</HTML>
Process.jsp
//Call validation methods and set methods
//If everything validates OK, then call CourseMgr.setCourse(Course crs);
//The question is, do I write the Course class to have all String values
//and then change them to longs before I send to the DB?
//Should I try to validate before I set course_id (my idea)
//or set them and validate before I call setCourse() (someone else's)
//If the later is the case, then the object must be very loose
//and you have to trust that a person is going to call the validation
//method. I would like to not allow sets without validation. The only
//way I see it right now is to have a set for both String and long
//and validations for both. You can still leverage all the code,
//but it still seems stupid. I worked with ColdFusion for a while
//and the loose datatypes were a god send for this kinda thing.
Course.java
import com.mycomp.CourseMgr;
public class Course {
     private long course_id;
     private String status="";
     private String version;
private long created_by;
private Timestamp created_date;
public static Hashtable errorMsg;
     public Course() {
course_id = CourseMgr.getNextCourse_ID();
public long getCourse_id(){
return course_id;
public void setCourse_id(long l){
this.course_id = l;
//I would like to be able to not have to write two sets
//for every long. If the form input is named course_id
//doing a setProp(*) should reflect it.
//But it reflects as a String, doesn't it?
//I also don't know if I should put a try in here.
//A try is part of the validation method and it seems
//redundant. But how do I ensure validation before set?
public void setCourse_id(String s){
this.course_id = Long.parseLong(s);
public boolean validateCourse_id(long vL){
if ( valUtils.isInRange(vL,"Course_id")){
//if it's within a predetirmined range
return true;
}else{
errorMsg.put("course_id","Course ID out of range.");
return false;
//I would love to be able to call only one validation method
//but other objects need to use a long. It's only the
//view that uses Strings all the time. I would like it
//if the request could keep the form field for course_id as
//the type t was meant to be.
public boolean validateCourse_id(String vS){
if ( valUtils.isPositiveLong(vS)){
//if it can be parsedLong and is positive
return validateCourse_id(Long.parseLong(vS));
}else{
errorMsg.put("course_id","Course ID is invalid Long.");
return false;
I really hope this helps. Thank you greatly in advance for any assitance.

Captivate 8 Continually Crashes: When trying to publish, When trying to use the Object Style Manager. Also it is blocked by Windows Firewall (adobecaptivatews) every time it launches. Working with Captivate App Packager the pane to import from Adobe Edge

Captivate 8 Continually Crashes: When trying to publish, When trying to use the Object Style Manager. Also it is blocked by Windows Firewall (adobecaptivatews) every time it launches. Working with Captivate App Packager the pane to import from Adobe Edge is missing, so not clear how to import. Overall it seems unstable. Am I doing something wrong? Trying to evaluate before buying so working on a trial version (14 days left), but a bit concerned with Captivate 8 Performance. Please help! :-

Hi Vikram and Lilybiri,
Thanks for your responses :-)
I'm working on Windows 8.1... I think that I may have found the issue... I was saving the file in a project directory rather than the default My Captivate Projects folder...
Perhaps Captivate was struggling to find resources it needed?
It hasn't crashed for a while now, though it is struggling - I'm working with a 54 slide PPT presentation that is linked and it takes a very long time to interact.
Sometimes it says that I've removed slides, which I haven't?
Best,
Christy.

Launchd agents not working

Hi,
I have created a user launchd agent saved in ~/Library/LaunchAgents and is as follows:
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
<plist version="1.0">
<dict>
<key>Label</key>
<string>com.Keith.ThingsBackupLaunchd</string>
<key>ProgramArguments</key>
<array>
<string>/Library/Scripts/Folder Action Scripts/Copy Things to Dropbox.scpt</string>
</array>
<key>StartInterval</key>
<integer>30</integer>
</dict>
</plist>
Basically it calls a script every 30 secs (only for testing, it will actually only call the script when a file gets modified) which is as follows:
tell application "Finder"
set sfolder to folder "Things" of folder "Cultured Code" of folder "Application Support" of folder "Library" of folder "Keith" of folder "Users" of startup disk
set tfolder to folder "Dropbox" of folder "Keith" of folder "Users" of startup disk
duplicate sfolder to tfolder with replacing
end tell
Copying my Things DB folder to my Dropbox folder. The Script works fine but the Launchd Agent does not.
In console I am getting the following error message:
22/07/2010 22:41:18 com.apple.launchd.peruser.501184 (com.Keith.ThingsBackupLaunchd392) posix_spawn("/Library/Scripts/Folder Action Scripts/Copy Things to Dropbox.scpt", ...): Permission denied
22/07/2010 22:41:18 com.apple.launchd.peruser.501184 (com.Keith.ThingsBackupLaunchd392) Exited with exit code: 1
Actually it doesn't seem to matter what I try to call, even trying to start an app at a specific time with a launchd agent produces the same error.
Any ideas?
Thanks

Thanks for the suggestions,
I have tried lots of variations but still no closer. Currently I have:
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
<plist version="1.0">
<dict>
<key>Label</key>
<string>com.MyLaunchd</string>
<key>ProgramArguments</key>
<array>
<string>/bin/bash</string>
<string>cp</string>
<string>-R</string>
<string>-f</string>
<string>/Users/Keith/Library/Application\ Support/Prog /Users/Keith/Dropbox/Backups/</string>
</array>
<key>WatchPaths</key>
<array>
<string>/Users/Keith/Library/Application Support/Prog</string>
</array>
</dict>
</plist>
This errors in the console with
17/08/2010 21:54:39 com.MyLaunchd[4224] usage: cp [-R [-H | -L | -P]] [-fi | -n] [-apvX] source_file target_file
17/08/2010 21:54:39 com.MyLaunchd[4224] cp [-R [-H | -L | -P]] [-fi | -n] [-apvX] source_file ... target_directory
17/08/2010 21:54:39 com.apple.launchd.peruser.501[268] (com.MyLaunchd[4224]) Exited with exit code: 64
So I definitely know it is call the launchd when the folder is modified.
Any further suggestions?
Message was edited by: Crystal Reef

Is there a way to edit text in a smart object brought in from illustrator?

Is there a way to edit text in a smart object brought in from illustrator? Many thanks for the help!
And can anyone recommend learning resources for someone (me) who is an AI user for a decade who needs to work in PS for web mockups? Something other than a basic PS tutorial. Something geared towards that specific task. I'm not a PS beginner - but have mostly used it for photo correction until now.
Truly -- thanks! BTW - I'm in the CS 6 Suite and CC will not be available to me.

OMG! When I went back into the ai file and made sure that the text was on it's own layer (which I had not originally done) I could indeed edit the text in the file when exported to ps! This is a perfect solution for this situation. As far as the person I'm working with - who fears ai more than anything - I can work in ai - which in many ways is so much less cumbersome than ps - and just export - clicking "preserve" layers & text editing! I hope this helps someone else! Sisham - many thanks for your help!

Batch input with objects

Hi,
Exists a standard class to develop a batch input????
Any ideas???
Is it possible to create a batch input with objects??

Hi there
Why not
just define your class in the normal way
decide what variables you need to pass
in the methods of your class call the fm modules bdc open, bdc dynpro etc etc.
Incidently the following include will have a decent amount of code for the BDC bits.
include bdcrecx1.
A good way is to start sm35 to record what you are doing on the screen. When you've finished go to your recording in SM35 and click on CREATE PROGRAM - it will create most of the code for you including the screen fields -- you'll need to edit a bit of course as you don't need everything that will be shown.
Debugging in the CREATE PROGRAM might show you a decent generic way of generating BDC's which you could easily do via a class.
Note however using BDC's is the LEAST PREFRERRED WAY of performing work and should only be used as a LAST RESORT if there really is no other way. A lot of new BAPI's and Classes are now available which can do away with the need for BDC's in most cases.
job done.
cheers
jimbo

Docs with object status

list of document with object status, can somebody explain how does it work and how to activate a status
how can we activate new user status for documents, pl help
Edited by: varada rajan on Apr 3, 2008 7:31 AM
Edited by: varada rajan on Apr 3, 2008 7:46 AM

Hi,
If you are talking about sales order status, this can be configured through status profile.
SPRO-SD-SALES-SALES DOCUMENTS - DEFINE & ASSIGN STATUS PROFILE.
Based on the user status you have maintained for your sales document, the status can be viewed in table JEST.
For this first go to VBAK - get document condition header, and take this document condition header and go to table JEST.
Here your current status will start with E**** and status inactive flag would be blank.
Hope this helps you.
Reward points for contribution if it does.
Regards
Ravi

Image scaling problem with object handles

Hi,
I am facing scaling problem.I am using custom actionscript component called editImage.EditImage consists imageHolder a flexsprite which holds
image.Edit Image component is uing object handles to resize the image.I am applying mask to the base image which is loaded into imageholder.
My requirement is when I scale the editImage only the mask image should be scaled not the base image.To achieve this I am using Invert matix .With this
the base image is not scaled but it is zoomed to its original size.But the base image should not zoomed and it should fit to the imageholder size always. When I scale the editImage only the mask image should scale.How to solve this.Please help .
With thanks,
Srinivas

Hi Viki,
I am scaling the mask image which is attached to imageHolder.When i scale
the image both the images are scaled.
If u have time to find the below link:
http://fainducomponents.s3.amazonaws.com/EditImageExample03.html
u work with this component by masking the editImage.If u find any solution
for scale only mask ,plz share the same.
thanks,
Srinivas

Windows.Launch contract failed with error: The app didn't start..

We have installed windows 8 Prof X64 Edition; none of the Metro apps are not getting launched; the system is added in domain, the event viewer has the below information for one of the app.
Event Under: Applications
App microsoft.windowscommunicationsapps_8wekyb3d8bbwe!Microsoft.WindowsLive.Chat did not launch within its allotted time.
Event Under: AppHost
The App Host has encountered an unexpected error and will terminate. The error is 0x8007000E.
Event Under: Microsoft-Windows-TWinUI/Operational
Activation of the app microsoft.windowscommunicationsapps_8wekyb3d8bbwe!Microsoft.WindowsLive.Chat for the Windows.Launch contract failed with error: The app didn't start..
Regards Ajinkya Ghare MCITP-Server Administrator | MCTS

Hi,
Try to put a machine disjoin the domain to see if the same issue occurs.
Sometimes the Windows 8 machines cannot use metro app once join the domain, you can try this method:
1- Open "Local Group Policy Editor"
2-Navigate to "Computer Configuration - Administrative Templates - Network Isolation"
3- Open "Internet Proxy Servers for Metro Style Apps" and set the value to your proxy server address like 172.16.0.1:8080.
Regards,
Leo
Huang
Leo Huang
TechNet Community Support

Generate error list with object detail

Dear All,
I have two way (multimaster ) Replication but due to some reason it generated many errors during replicate data so to resolve errors, I want to generate error list (Report) with object details instead viewing them one by one. please guide
Thanks

What do you mean with generate error list (Report) with object details ?
You can get all the errors from DEFERROR view.
There are columns DEFERRED_TRAN_ID and CALLNO.
You can join this columns with DEFCALL (DEFERRED_TRAN_ID and CALLNO).
Joining this two views will provide information about the objects and transaction type involved in the call.
DEFCALL.PACKAGENAME contains the reference to the table_name.
DEFCALL.PROCNAME contains information about the transaction type.
Example:
select e.deferred_tran_id,
       e.callno,
       e.origin_tran_db,
       c.packagename,
       c.procname
from deferror e, defcall c
where e.deferred_tran_id=c.deferred_tran_id
and e.callno=c.callno(below are the links to the 10gR2 docs containing information about DEFERROR and DEFCALL views. If your database is different version, than search http://tahiti.oracle.com for version specific documentation)
DEFERROR
http://download.oracle.com/docs/cd/B19306_01/server.102/b14227/rardeftranviews.htm#sthref2599
DEFCALL
http://download.oracle.com/docs/cd/B19306_01/server.102/b14227/rardeftranviews.htm#sthref2595
Cheers!
Message was edited by:
tekicora

Launchd - agent doesn't work

Hi,
I've created an agent with Lingon. It loads but the script failed with the following errors in system.log
Jan 4 20:02:52 dchambon launchd[3231]: fr.dchambon.MontageAuto: exited abnormally: Broken pipe
Jan 4 20:02:52 dchambon launchd[3231]: fr.dchambon.MontageAuto: 4 more failures without living at least 60 seconds will cause job removal
Here is the script :
#!/bin/sh -x
OutilsPAO=`df -t afpfs | grep -q Outils ; echo $?`
if [ $OutilsPAO == 1 ] ; then
mkdir '/Volumes/Outils PAO'
/sbin/mount_afp "afp://pao:INFOPUB@XServPao/Outils%20Pao" '/Volumes/Outils PAO'
if [ $? -ne 0 ] ; then
echo -e "" | mail -s "Montage Launchd - Echec du montage de \"Outils PAO\"" [email protected]
fi
fi
Echanges=`df -t afpfs | grep -q Echanges ; echo $?`
if [ $Echanges == 1 ] ; then
mkdir '/Volumes/Echanges'
/sbin/mount_afp "afp://pao:INFOPUB@W2SSINFOB01/Echanges" '/Volumes/Echanges'
if [ $? -ne 0 ] ; then
echo -e "" | mail -s "Montage Launchd - Echec du montage de \"Echanges\"" [email protected]
fi
fi
exit 0
Only the first mkdir command is executed
And the plist situated in ~/Library/LaunchAgents :
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE plist PUBLIC "-//Apple Computer//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
<plist version="1.0">
<dict>
<key>Label</key>
<string>fr.dchambon.MontageAuto</string>
<key>ProgramArguments</key>
<array>
<string>/bin/sh</string>
<string>/Users/dchambon/Scripts/Shell/MontageAuto.sh</string>
</array>
<key>RunAtLoad</key>
<true/>
</dict>
</plist>
Any idea ??
Regards
Didier.

Hi Didier,
   I like your MontageAuto.sh script; I've not seen df used that way. If the AFP shares are not mounted, does the script work when run in the command line as dchambon? The only possible syntax error I see is that you didn't use fully-qualified domain names for the servers, XServPao and W2SSINFOB01. It is possible that your dchambon user has a search domain preference. As a consequence, the mount_afp commands may work only after that preference takes effect. Naturally, it's possible that that preference is not set at the time the Launchd job is run.
   I don't think that RunAtLoad is a mode in which you can run this type of job. RunAtLoad is used to start daemons that run continuously. In fact, Launchd is expected to relaunch the daemon if it quits, which is why you get the error message, "fr.dchambon.MontageAuto: exited abnormally: Broken pipe." One would think that the script would run at least once but I've heard reports of other scripts failing under these circumstances.
   Unfortunately, I don't have a good alternative to RunAtLoad. I would try just removing that key and its value. You could also make the script a LoginHook. To that end, you might find the following article useful, Mac OS X: Creating a login hook. Of course the LoginHook would be run as root.
Gary
~~~~
   Emacs, n.:
      A slow-moving parody of a text editor.

How to resolve Agents with pending URL changes

Greetings,
I have run the EMDIAG tool against my 12c OMS and one section of the results hows thie following -
7001. Agents with pending URL changes
LAST_EMD_URL
EMD_URL
https://lepisma.ucdavis.edu:3872/emd/main
https://lepisma.ucdavis.edu:1830/emd/main
1 row selected.
verifyASLM
======================================================================
On this host ./emctl status agent returns -
Oracle Enterprise Manager 12c Cloud Control 12.1.0.1.0
Copyright (c) 1996, 2012 Oracle Corporation. All rights reserved.
Agent Version : 12.1.0.1.0
OMS Version : 12.1.0.1.0
Protocol Version : 12.1.0.1.0
Agent Home : /usr/local/oracle/product/agent12g/agent_inst
Agent Binaries : /usr/local/oracle/product/agent12g/core/12.1.0.1.0
Agent Process ID : 44637
Parent Process ID : 44555
Agent URL : https://lepisma.ucdavis.edu:3872/emd/main/
Repository URL : https://oracle-emrep1.ucdavis.edu:4904/empbs/upload
Started at : 2013-04-20 16:09:06
Started by user : oracle
Last Reload : (none)
Last successful upload : 2013-04-22 13:34:14
Last attempted upload : 2013-04-22 13:34:14
Total Megabytes of XML files uploaded so far : 134.96
Number of XML files pending upload : 1
Size of XML files pending upload(MB) : 0.03
Available disk space on upload filesystem : 62.49%
Collection Status : Collections enabled
Last attempted heartbeat to OMS : 2013-04-22 13:34:11
Last successful heartbeat to OMS : 2013-04-22 13:34:11
Agent is Running and Ready
Examinig the emd.properties file shows no entry for https://lepisma.ucdavis.edu:1830/emd/main and ./emctl config agent listtargets only shows the following for oracle_emd -
[lepisma.ucdavis.edu:3872, oracle_emd]
This was recently upgraded from 11g and prior to the upgrade there were difficulties patching this agent. Because of that it was reinstalled in 12c. I have no clue what is causing the issue let alone how to resolve it. Any suggestions are greatly appreciated.
Thank you.
Bill Wagman

It's ok to have 11g agent and 12c agents up and running simultaneously and we do it all the time during migration from one to the other since they have to run on different ports (12c checks this at install ONLY IF the 11g agent it up and running when the 12c agent is being pushed out to the host).
Also, if EMDIAG is the only place you see 1830 port being reported, it may be becuase its doing a search for any agents currently installed on the server. So if you still have both the 11g agent and 12c agent installed, this should not be an issue that it it reporting that URL/port with 1830 for the 11g agent. Since you can't upgrade 11g agents to 12c, (its required to do a new install of 12c agent) you have to do a complete uninstall of the 11g agent first from the OEM 11g gui console, then at the inventory level where the inital 11g agent was installed, than you may still need to manually remove that 11g directory if your runInstaller did not remove it once its "uninstalled".
Edited by: ora6dba on Apr 25, 2013 7:24 AM

Applet doesn't work with object tags--PLZ HELP!

<NOEMBED>
</NOEMBED>
</EMBED>
</COMMENT>
</OBJECT>

ooops! The applet is not displayed with object tags generated by HtmlConverter utility, The applet is supposed to display a JTable and connect to Oracle DB using JDBC. The web server is Tomcat 4.1.29
<OBJECT
    classid = "clsid:CAFEEFAC-0014-0001-0002-BBCDEFFEDCBA"
    codebase = "http://java.sun.com/products/plugin/autodl/jinstall-1_4_1_02a-windows-i586.cab#Version=1,4,1,21"
    WIDTH = 800 HEIGHT = 500 NAME = "Edit Trunks" >
    <PARAM NAME = CODE VALUE = "cbhtolApplet.class" >
    <PARAM NAME = ARCHIVE VALUE = "trnkapplet.jar,ojdbc14.jar" >
    <PARAM NAME = NAME VALUE = "Edit Trunks" >
    <PARAM NAME = "type" VALUE = "application/x-java-applet;jpi-version=1.4.1_02a">
    <PARAM NAME = "scriptable" VALUE = "false">
    <COMMENT>
        <EMBED
            type = "application/x-java-applet;jpi-version=1.4.1_02a"
            CODE = "cbhtolApplet.class"
            ARCHIVE = "trnkapplet.jar,ojdbc14.jar"
            NAME = "Edit Trunks"
            WIDTH = 800
            HEIGHT = 500
            scriptable = false
           pluginspage = "http://java.sun.com/products/plugin/index.html#download">
            <NOEMBED>
            </NOEMBED>
        </EMBED>
    </COMMENT>
</OBJECT>Appreciate your help!

[SOLVED] systemd: Launch helper exited with unknown return code1

Why might I be seeing messages like the following when trying to set up systemd?
[root@blah ~]# hostnamectl set-hostname foobar
Failed to issue method call: Launch helper exited with unknown return code 1
EDIT: The solution for me was to remove sysvinit and initscripts and go for pure systemd.
Last edited by amadar (2012-11-01 05:47:08)

amadar wrote:
2ManyDogs wrote:Did you search the forum, as I suggested?
https://bbs.archlinux.org/viewtopic.php?id=151633
Yeah, I already found that post but the solution is not there. I'm not trying to use echo to set hostname manually, I'm trying to see why my hostnamectl command (and other *ctl comands) doesn't work, and there's no solution on that thread, just a workaround.
I have systemd 195 and polkit 0.107 installed.
EDIT: I replaced sysvinit with systemd-sysvcompat; then enabled syslog-ng, NetworkManager, and cronie services using systemctl enable; then I uninstalled initscripts. I'm not sure where during that process the commands started working, but they work now. But, that doesn't help people trying to migrate without having to replace sysvinit or remove initscripts first.
This is a problem because the wiki explicitly says to migrate rc.d settings to the new files used by systemd. However, since the *ctl commands don't work at this point, it's impossible for the user to know how to make the config files manually without intimate knowledge of their formats without using *ctl commands.
For example, how is one supposed to complete step one on the following wiki section? https://wiki.archlinux.org/index.php/Sy … stallation
As you can see in those instructions, there is no information on how to create the new config files other than using *ctl commands, which don't even work at that stage. Some of the examples show a sample file, but the problematice ones are the ones that don't show a sample file.
For example, how do you make /etc/adjtime without timedatectl?
The wiki says that timedatectl [...] "will generate /etc/adjtime automatically; no further configuration is required", but then the Failed message in my original post is displayed.
I think I will change the wiki to offer a better walk through when I get the chance to migrate to systemd on my desktop which has many daemons and modules.
the same problem here....

Editing & launching launchd agents with Objective C

Similar Messages

Maybe you are looking for