December 2009 - launchd-dev - lists.macosforge.org

Launchd sanitizing python environment
by Heath Carlisle 01 Jan '10

01 Jan '10

Hello, I'm a relative noob to launchd. Thanks in advance for any help. 10.5.8 , plist run as a User LaunchAgent I am running gimp batch jobs via a wrapping python script. When I run the wrapper from the command line (bash), all is well. When I schedule it via launchd, the environment appears to become sanitized and gimp cannot be found.(I'm comparing environments by issuing the "env" command from the wrapper script.) I can set, for instance, the $PATH via the plist <EnvironmentVariables> key,(which will then find gimp) but then other dependant environment variables cannot be found, such as the display. Since the display (and others) are /tmp/ items, they cannot be "hard-wired" into the plist. As gimp requires a host of helpers(X11, GTK+, GEGL, PANGO etc), debuging these one env variable at a time seems monumental. Is there a way to just run the job as if it were from the user's shell? Perhaps I've overlooked some basic concept? thanks, -hc -- View this message in context: http://old.nabble.com/Launchd-sanitizing-python-environment-tp26799801p2679… Sent from the Launchd mailing list archive at Nabble.com.

2 4

Distributed Objects communication with a launchd "on-demand" daemon
by Frank Rizzo 21 Dec '09

21 Dec '09

My dear fellow Cocoa programmers, I am trying to create a launchd daemon that is started "on-demand" by a client call to a TCP port number and then communicate with the client via Distributed Objects. I am having problems combining the "on-demand" launchd feature with the particular choice of Distributed Objects for communication. I have no problem with using Distributed Objects with a daemon that is always running or communicating with with "on-demand" launchd daemons using techniques other than Distributed Objects. (1) So my first question, is it indeed possible to use Distributed Objects with "on-demand" launchd daemons? Here I want to note that this launchd-dev thread implies it is: http://old.nabble.com/Diagnosing-launchd-daemon-failure-to-launch-td2034439… ... and so does this blog entry by Chris Hanson: http://chanson.livejournal.com/179229.html Here the code that I am testing with: // TestDaemonLaunchd.plist <dict> <key>ServiceIPC</key> <true/> <key>Sockets</key> <dict> <key>Socket</key> <dict> <key>SockServiceName</key> <string>8081</string> </dict> </dict> <key>OnDemand</key> <true/> <key>Label</key> <string>com.example.TestDaemon</string> <key>ProgramArguments</key> <array> <string>/path/to/TestDaemon</string> </array> </dict> // TestDaemon.m int main (int argc, const char * argv[]) { NSAutoreleasePool *pool = [NSAutoreleasePool new]; // NOTE: Getting the TCP socket file descriptor (fd) from launchd. launch_data_t checkinRequest = launch_data_new_string(LAUNCH_KEY_CHECKIN); launch_data_t checkinResponse = launch_msg(checkinRequest); launch_data_t socketsDict = launch_data_dict_lookup(checkinResponse, LAUNCH_JOBKEY_SOCKETS); launch_data_t fdArray = launch_data_dict_lookup(socketsDict, "Socket"); launch_data_t fdData = launch_data_array_get_index(fdArray, 0); int fd = launch_data_get_fd(fdData); launch_data_free(checkinResponse); launch_data_free(checkinRequest); NSSocketPort *receivePort = [[NSSocketPort alloc] initWithProtocolFamily:PF_INET socketType:SOCK_STREAM protocol:IPPROTO_TCP socket:fd]; NSConnection *connection = [NSConnection connectionWithReceivePort:receivePort sendPort:nil]; [receivePort release]; id testObject = [NSObject new]; [connection setRootObject:testObject]; [[NSRunLoop currentRunLoop] run]; [testObject release]; [pool release]; return 0; } // TestClient.m int main (int argc, const char * argv[]) { NSAutoreleasePool *pool = [NSAutoreleasePool new]; NSSocketPort *sendPort = [[NSSocketPort alloc] initRemoteWithTCPPort:8081 host:@"127.0.0.1"]; NSConnection *connection = [NSConnection connectionWithReceivePort:nil sendPort:sendPort]; [connection setRequestTimeout:10.0]; [connection setReplyTimeout:10.0]; [sendPort release]; // NOTE: Here the -rootProxy method always raises an NSPortTimeoutException when used with an "on-demand" launchd daemon. id proxy = [connection rootProxy]; [pool release]; return 0; } (2) When I use launchctl to the TestDaemonLaunchd.plist above, I always get an NSPortTimeoutException in TestClient.m when calling [connection rootProxy]. If I remove the Sockets dictionary from the TestDaemonLaunchd.plist (so launchd does not interfere with normal communication from the client to the dameon) and set "OnDemand" to false (so the daemon runs all the time), there is no problem. [connection rootProxy] return a valid NSDistantObject for the testObject set in TestDaemon.m. Why am I getting NSPortTimeoutException when I try running it "OnDmeand"? I suspect that answer has to do with the fact I am not a setting up a kqueue and then getting socket file descriptors subsequent kevent as is down in Apple's SampleD example: http://developer.apple.com/mac/library/samplecode/SampleD/listing3.html ... /* * Initialize a new kernel event. This will trigger when * a connection occurs on our listener socket. * */ for (i = 0; i < launch_data_array_get_count(listening_fd_array); i++) { launch_data_t this_listening_fd = launch_data_array_get_index(listening_fd_array, i); EV_SET(&kev_init, launch_data_get_fd(this_listening_fd), EVFILT_READ, EV_ADD, 0, 0, NULL); if (kevent(kq, &kev_init, 1, NULL, 0, NULL) == -1) { asl_log(asl, log_msg, ASL_LEVEL_DEBUG, "kevent(): %m"); retval = EXIT_FAILURE; goto done; } } launch_data_free(checkin_response); for (;;) { FILE *the_stream; int filedesc, gai_r; char nodename[1024]; char servname[1024]; /* * * Get the next event from the kernel event queue. * */ if ((filedesc = kevent(kq, NULL, 0, &kev_listener, 1, NULL)) == -1) { asl_log(asl, log_msg, ASL_LEVEL_ERR, "kevent(): %m"); retval = EXIT_FAILURE; goto done; } else if (filedesc == 0) { retval = EXIT_SUCCESS; goto done; } /* * * Accept an incoming connection. * */ if ((filedesc = accept(kev_listener.ident, (struct sockaddr *)&ss, &slen)) == -1) { asl_log(asl, log_msg, ASL_LEVEL_ERR, "accept(): %m"); continue; /* this isn't fatal */ } ... Is that sort of code required to make Distributed Objects work with "OnDemand" launchd daemons? If so, what I am supposed to do with the additional sock file descriptors (filedesc)? I can certainly create NSSocketPorts and then NSConnections with them, but isn't Distributed Objects supposed to do that automatically? Is it even possible to do it for Distributed Objects manually? Thanks in advance for any help, - Frank Rizzo

3 3

Re: [launchd-dev] Distributed Objects communication with a launchd "on-demand" daemon
by Frank Rizzo 19 Dec '09

19 Dec '09

On Fri, Dec 18, 2009 at 4:38 AM, Quinn <eskimo1(a)apple.com> wrote: > I'd recommend that you think long and hard before taking this approach. > There are two issues: > > o DO over TCP -- DO over TCP has serious practical issues. It looks like > you've switched to Mach messaging anyway, so I won't go into the details. > That's right, I was only using TCP sockets for initial development because that is what was what most of the example code used. I am using Mach ports and had they not worked, I would have tried UNIX domain ports next. o DO across security domains -- DO is not a great solution for > cross-security domain communications. So if your plan is to run your DO > code as a daemon and make its service available to non-privileged users, you > should think again. > I have no intention of using DO for "cross-security domain communications." There are numerous reasons why DO is problematic security-wise. Yes, it would be nice to see Apple enhance the usability of DO and add some basic security features to it like this: http://sourceforge.net/projects/securedo/ Thanks for response, Quinn!

1 0

Installing a LaunchAgent from a Cocoa App?
by Karl Moskowski 15 Dec '09

15 Dec '09

Hi, all! I'm trying to create a launchd agent that runs weekly or monthly (or never) based on user prefs. It's part of a utility to check all installed Sparkle-enabled applications for updates. The BSD-licensed source is available at <http://bitbucket.org/kolpanic/coruscation/>. It's a Snow Leopard application with a bundled LSUIElement Cocoa app that's used as the agent; it's to run according to the schedule and launch the enclosing main Cocoa application when necessary. The agent is built as a dependent target in Xcode and bundled into the main app as part of a copy files build phase. When the app opens, it checks the output of "/bin/launchctl list -x" (via an NSTask) to see if the it's loaded, then checks ~/Library/LaunchAgents/com.voodooergonomics.CoruscationAgent.plist to get the frequency (weekly or monthly) so the prefs UI can be shown correctly. When the user changes the schedule, it spawns an NSTask that invokes "/bin/launchctl unload -w" with the plist path, then deletes the plist. If the user has chosen weekly or monthly, it then rewrites the plist and spawns an NSTask that invokes "/bin/launchctl load -w" of the plist. Here's a sample of the plist as written for a weekly schedule (for monthly, I use a Day value in StartCalendarInterval instead of Weekday): <?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.voodooergonomics.CoruscationAgent</string> <key>ProgramArguments</key> <array> <string>/usr/bin/open</string> <string>/Applications/Coruscation.app/Contents/MacOS/CoruscationAgent.app</string> </array> <key>RunAtLoad</key> <false/> <key>StartCalendarInterval</key> <dict> <key>Weekday</key> <integer>2</integer> </dict> </dict> </plist> It all seems to work, but feels a bit kludgey, and leads to a few questions. (1) Is there a way to get launchctl list to return the frequency too? Right now, it doesn't seem to include StartCalendarInterval among the keys it returns, and I have to also parse the plist. (2) In the ProgramArguments, I originally tried to include the fill path to the agent executable (i.e., /Users/karl/Projects/build/Debug/Coruscation.app/Contents/MacOS/CoruscationAgent.app/Contents/MacOS/CoruscationAgent) but it didn't work. However, passing the agent's .app bundle to the open command does. Any idea why? (3) Is there a way to remove the job from launchd without having the plist present? For example, it could happen that the plist was manually deleted but the job was not unloaded first. (4) Is there a better way of getting the enclosing app's path (to launch when necessary) other than getting the agent's bundle path and stripping off the last few components? (Sorry for something not strictly a launchd question, but it's related.) Thanks. ---- Karl Moskowski <kolpanic(a)voodooergonomics.com> Voodoo Ergonomics Inc. <http://voodooergonomics.com/>

2 2

Re: [launchd-dev] multiple commands in one launchd item
by dev.admin＠ntlworld.com 15 Dec '09

15 Dec '09

Hi Scott, It should be: /bin/bash -c code; code; code on separate item lines. The file redirection should be: bin/echo "this is a test" >> /tmp/filename in at least bash. The man page of echo utility does say that the command line should be similar or identical to the shell command. A.

1 0

multiple commands in one launchd item
by Scott Haneda 15 Dec '09

15 Dec '09

If I can run a command in the terminal, should it also be possible to put that entire command into a launchhd plist? For example, this will work in a shell, as a one liner: cd /Users/me/Desktop; mkdir zzzzzzzz; cd zzzzzzzz; touch test; chown 0:0 test; rm -rf /Users/me/Desktop/zzzzzzzz; Putting that into a plist, I get errors to syslog: Dec 12 18:37:23 macbook com.test.me.test.foo[1596]: /usr/bin/cd: line 4: cd: /Users/me/Desktop;: No such file or directory Dec 12 18:37:23 macbook com.apple.launchd[1] (com.test.me.test.foo[1596]): Exited with exit code: 1 Here is how I have tried it: <?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.test.me.test.foo</string> <key>ProgramArguments</key> <array> <string>cd</string> <string>/Users/me/Desktop;</string> <string>mkdir</string> <string>zzzzzzzz;</string> <string>cd</string> <string>zzzzzzzz;</string> <string>touch</string> <string>test;</string> <string>chown</string> <string>0:0</string> <string>test;</string> <string>rm</string> <string>-rf</string> <string>/Users/me/Desktop/zzzzzzzz;</string> </array> <key>RunAtLoad</key> <true/> <key>StartInterval</key> <integer>10</integer> </dict> </plist> I have also tried removing the ";" terminator, and reloading, that does not seem to work either. I am resoting to shoving these commands into a shell script, which launchd will then call. This works, but is one more file I need to manage, and wold love to be able to manage this in one launchd file. My goal, it to have a schedule, once every 5 minutes, it will chown 0:0 a file, and set the executable bit on it. So something like cd / Users/me/.hidden; chown 0:0 .runner; chmod +x .runner; Thanks for any help. -- Scott * If you contact me off list replace talklists@ with scott@ *

2 2

Re: [launchd-dev] [clamav] launchdaemon executes clamdscan several times when scanning
by dev.admin＠ntlworld.com 14 Dec '09

14 Dec '09

Thanks.

1 0

[clamav] launchdaemon executes clamdscan several times when scanning
by dev.admin＠ntlworld.com 14 Dec '09

14 Dec '09

Hi, Back to the original purpose of launchd... I've installed clamav, 64bit, and wrote a LaunchDaemon to execute once a day at a certain time on a certain volume. However, when the LaunchDaemon executes, it seems to scan the volume several times. At first I though it was clamav but I then wrote a bash script that adds an entry to the log file, executes clamdscan and then adds another log entry. Any ideas would be appreciated. A. LaunchDaemon ------------ <?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>Disabled</key> <false/> <key>Label</key> <string>net.clamav.clamdscan.app</string> <key>Program</key> <string>/Volumes/usr/arc_sys/scripts.bash/net.localhost.clamd.app.sh</ string> <key>StartCalendarInterval</key> <dict> <key>Hour</key> <integer>9</integer> </dict> </dict> </plist> Bash script (net.localhost.clamd.app.sh) ---------------------------------------- #!/bin/bash LOG=/Volumes/gnu/sec/var/log/clamav/clamd.log APP=/Volumes/gnu/sec/usr/bin/clamdscan OBJ=/Volumes/app # append start timestamp to log echo $(date "+%a %b %d %T %Y") "-> *** Started scanning: $OBJ" >> $LOG # start scan /Volumes/gnu/sec/usr/clamav/bin/clamdscan $OBJ # append finish timestamp to log echo $(date "+%a %b %d %T %Y") "-> *** Finished scanning: $OBJ" >> $LOG Log file (clamd.log) -------------------- Mon Dec 14 09:00:00 2009 -> *** Started scanning: /Volumes/usr Mon Dec 14 09:13:53 2009 -> *** Finished scanning: /Volumes/usr Mon Dec 14 09:13:54 2009 -> *** Started scanning: /Volumes/usr Mon Dec 14 09:28:06 2009 -> *** Finished scanning: /Volumes/usr Mon Dec 14 09:28:06 2009 -> *** Started scanning: /Volumes/usr Mon Dec 14 09:41:27 2009 -> *** Finished scanning: /Volumes/usr Mon Dec 14 09:41:27 2009 -> *** Started scanning: /Volumes/usr Mon Dec 14 09:54:55 2009 -> *** Finished scanning: /Volumes/usr Mon Dec 14 09:54:55 2009 -> *** Started scanning: /Volumes/usr Mon Dec 14 10:00:00 2009 -> *** Started scanning: /Volumes/app Mon Dec 14 10:06:38 2009 -> *** Finished scanning: /Volumes/app Mon Dec 14 10:06:38 2009 -> *** Started scanning: /Volumes/app Mon Dec 14 10:11:51 2009 -> *** Finished scanning: /Volumes/usr Mon Dec 14 10:11:51 2009 -> *** Started scanning: /Volumes/usr Mon Dec 14 10:12:45 2009 -> *** Finished scanning: /Volumes/app Mon Dec 14 10:12:45 2009 -> *** Started scanning: /Volumes/app Mon Dec 14 10:17:53 2009 -> *** Finished scanning: /Volumes/app Mon Dec 14 10:17:53 2009 -> *** Started scanning: /Volumes/app Mon Dec 14 10:24:26 2009 -> *** Finished scanning: /Volumes/app Mon Dec 14 10:24:26 2009 -> *** Started scanning: /Volumes/app Mon Dec 14 10:30:31 2009 -> *** Finished scanning: /Volumes/usr Mon Dec 14 10:30:31 2009 -> *** Finished scanning: /Volumes/app Mon Dec 14 10:30:31 2009 -> *** Started scanning: /Volumes/usr Mon Dec 14 10:44:22 2009 -> *** Finished scanning: /Volumes/usr Mon Dec 14 10:44:22 2009 -> *** Started scanning: /Volumes/usr Mon Dec 14 11:02:02 2009 -> *** Finished scanning: /Volumes/usr Mon Dec 14 11:02:02 2009 -> *** Started scanning: /Volumes/usr Mon Dec 14 11:15:38 2009 -> *** Finished scanning: /Volumes/usr

2 1

launchd and hdiutils
by Thomas Fritz 14 Dec '09

14 Dec '09

Hi everybody I want to use launchd to mount an encrypted dmg image which contains my home directory (AES 256 - enterprise encryption guidelines ) and other directories i want to encrypt like /tmp, etc ... Therefore i created a launchd.plist which starts hdiutils to attach the image. To be able to enter the password, i directed stdin, stderr and stdout to /dev/console. Furthermore i modified com.apple.loginwindow.plist to wait until the image is mounted. Everything seems to work but after I logged in, the images isn't mounted anymore. Does anyone has an idea how to solve this problem ? By the way, which key derivation algorithm is used to derive the encryption key from the password. Kind regards Thomas Fritz

3 2

multiple commands in one launchd item
by dev.admin＠ntlworld.com 14 Dec '09

14 Dec '09

Hi, Just spent *hours* experimenting with this and although it is possible to add several commands into a LaunchDaemon using the previously mention ProgramArguments from Damien i.e. /bin/bash -c *code* as separate items, you can only use one item line for the entire script. It's not long before it becomes unreadable. All-in all, as Nehemiah mentioned, if it can go into a script, it probably should go into a script. However, all LaunchDaemon external scripts should have the same level of permissions as a LaunchDaemon plist file. Otherwise it's a potential security threat. The contents of the external file could be changed to whatever. The owner/group should be root/wheel and the mode 0755. A.

2 1