- launchd-dev - lists.macosforge.org

What does the "_dirty" key mean?
by Axel Luttgens 08 Apr '16

08 Apr '16

Hello, When running: sudo launchctl print-disabled system on 10.11.3, I get: disabled services = { "_dirty" => true […] } What does that "_dirty" key mean? I’m asking, because I’m wondering if I did something wrong… TIA, Axel

3 7

LaunchAgent best practice for turning "off" sockets
by Mitchell Laurren-Ring 31 Mar '16

31 Mar '16

I have a LaunchAgent that is configured via Sockets in the plist to listen on several IP+port number endpoints. I would like to make it possible to turn "off" some listening endpoints while the LaunchAgent is running. Here’s a section of my plist: <key>Sockets</key> <dict> <key>RemoteSocket</key> <dict> <key>SockServiceName</key> <string>26666</string> <key>SockFamily</key> <string>IPv4</string> <key>Bonjour</key> <string>myService</string> </dict> My process doesn’t have privileges so I can’t modify the plist. I could simply ignore connection attempts from my kevent code but, as you can see, I’m advertising my service via Bonjour. What is the best practice for doing this? -Mitchell J Laurren-Ring

2 3

Strange XPC errors in log
by Rainer Brockerhoff 11 Mar '16

11 Mar '16

Hi! I'm debugging an XPC service with the ultimate aim to self-update one of my apps. The app itself is sandboxed, the service is not (which is OK as it's Developer-ID signed). So, the service runs well, downloads and installs the update if necessary, then launches the new version while quitting the old. All this works well, too, and I plan to release it soon, but... However, I see two strange errors in the Console log: 1) on my first call to the XPC service: ... com.apple.xpc.launchd[1]: (com.apple.xpc.launchd.domain.pid.<appName>.45747) Ignoring lax activation attempt for unsupported domain: <serviceBundleId> 2) when launching the updated app: ... com.apple.xpc.launchd[1]: (com.apple.xpc.launchd.domain.pid.(premature birth).45747) Could not resolve origin of domain. XPC services in this domain's bundle will not be bootstrapped: error = 3: No such process, taint = (untainted) This happens regardless of the value of "JoinExistingSession" in the XPC service's Info.plist. Note that both PIDs are the same, too! Any hints welcomed. TIA, -- Rainer Brockerhoff <rainer(a)brockerhoff.net> Belo Horizonte, Brazil "In the affairs of others even fools are wise In their own business even sages err." http://brockerhoff.net/blog/

1 0

(re)configure a launchd created UNIX domain socket after Fast User Switching (on El Capitan)
by Ludovic Rousseau 01 Mar '16

01 Mar '16

Hello, I am using launchd to create a Unix domain socket /tmp/foobar.socket so that launchd starts my agent when someone connects to the sockets. It works fine except when Fast User Switching is involved. My problem is quite similar to the problem "Listening socket and fast user switching" [1] discussed on this mailing list in 2008. The difference is that I am using a UNIX domain socket and not a TCP socket. The problematic scenario is as follows: 1. User A logs in 2. User A can connect to the socket 3. User B logs in using the Fast User Switching 4. User B can connect to the socket 5. User B logs out 6. User A comes back and fast user switch to his session. Since user A does not _logs in_ again the socket is not changed/recreated and is still only usable by user B At this step user A gets: $ telnet /tmp/foobar.socket Trying /tmp/foobar.socket... /tmp/foobar.socket: Connection refused The socket is still present in /tmp/ but it belongs to user B and is no more usable but user A. What I would like is the ability for launchd to (re)configure the Unix domain socket each time a user authenticates to the console, either the first time or using the fast user switching method. - I use a UNIX domain socket so that only local processes can talk to the socket and use my agent. I don't want to open my agent to other computer on the network. - I use launchd to start my agent on request only. That is a very nice feature. I can't detect the user switching since my agent code is, most of the time, not running. - I can't use a different socket name per user since the client of my agent does not know who is currently logged. My client application is talking to the socket configured for whoever is logged at that time. My /Library/LaunchAgents/com.foobar.plist file contains: <?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.foobar</string> <key>ProgramArguments</key> <array> <string>/usr/local/lib/foobar_exe</string> </array> <key>Sockets</key> <dict> <key>Listeners</key> <dict> <key>SockPathName</key> <string>/tmp/foobar.socket</string> </dict> </dict> <key>ProcessType</key> <string>Interactive</string> </dict> </plist> Does launchd provides a solution to my problem? Should I open a bug at Apple to ask for this new feature in launchd? Thanks [1] https://lists.macosforge.org/pipermail/launchd-dev/2008-May/000298.html -- Dr. Ludovic Rousseau

4 5

Cause of "Service cannot load in requested session"?
by James Bucanek 16 Nov '15

16 Nov '15

Greetings, I have a user agent installed in the "Background" session that isn't starting up on one system. My customer is running OS X 10.11.1. I have this software running on hundreds of El Capitan systems, but for some reason this user agent won't start when the user logs in. I've found two things that are suspicious. First, a message like this appears in the console log: system.log: Oct 31 17:45:05 UsersMac otherbsd[488]: Failed to bootstrap agent: path = /Users/theuser/Library/LaunchAgents/com.qrecall.scheduler.plist, error = 134: Service cannot load in requested session I was really hoping that someone could tell me what error 134 means. The other, really strange, thing is that when I look at the log output of the com.qrecall.scheduler process (when it does get started, usually as a result of a launchctl load command), I sometimes find that the very first thing it logs is the receipt of a SIGTERM signal. That means that it has received a SIGTERM even before its run loop has started. The result is that the agent starts up and then immediately shuts down. Context (if it's important): This service is installed as an alway running user agent, in the "Background" session. This is so the process will continue to run even after the user has logged out. (The service can schedule backup and other maintenance tasks to run while the user is logged out.) To make sure the process is running even before the user logs in for the first time, I install a tiny "kicker" system daemon. The "kicker" has a list of UIDs of the users that have installed the background scheduler. At startup, the "kicker" simply make a bootstrap call that causes the bootstrap for that user to be created. This, in turn, initializes the user's bootstrap, which involves starting all of its background processes. I suspect that console message is being logged around this time (but it's hard to tell exactly just by looking at the logs). I've had the user switch to using a regular user agent ("Aqua" session, and no kicker daemon), and things seem to be working. Any insight would be helpful. TIA, James

4 16

Re: [launchd-dev] Cause of "Service cannot load in requested session"?
by Damien Sorresso 06 Nov '15

06 Nov '15

On 5 Nov, 2015, at 16:46, Mihai Moldovan <ionic(a)ionic.de> wrote: > On 06.11.2015 01:41 AM, Damien Sorresso wrote: >> Perhaps the best way to answer would be the statement that the only time a home directory guaranteed to be available is when the user is logged in at the GUI console. Once that GUI login session goes away, there is no guarantee of home directory availability. > > I think console sessions should also be included here. > > It's probably a very unknown feature that a console session can be spawned by logging with the user name ">console" (no password) and then logging on with the user credentials on the console, but at least on 10.9 this is still supported. Not sure about later versions. > > I'd like to add that I never tried that for a user with a non-local, already mounted home directory. That thing still works? News to me! ;) In all seriousness, I'm not sure that it will mount a network home for you as it just authenticates through PAM, and I'm not sure if PAM does the network home magic. I'm pretty sure it doesn't do the OG FileVault magic though. -damien

1 0

Remove "disabled" services; Migrate from "old" launchd
by Jerry Krinock 27 Aug '15

27 Aug '15

Ever since 10.5, my apps have used long-running launchd agents to detect filesystem changes, and these in turn schedule one-shot agents to perform tasks. They do this by writing and removing plist files in ~/Library/LaunchAgents, and by running launchctl with subcommands ‘load’ and ‘unload’. I just discovered that running this command: launchctl print-disabled gui/`id -u` shows hundreds of such services, probably every one which my app has created for the past year or more. Apparently, my agents are not being completely removed, only “disabled”. How do I remove them? I tried ‘unbootstrap’, but got: Command is not yet implemented. * * * More broadly, is there an article which explains the recent changes in launchd, so that I can properly update my apps? The current launchctl man page is incomprehensible to me until I get to the LEGACY SUBCOMMANDS section. No more “agents", it seems; everything is a “service”.

4 7

Re: [launchd-dev] Remove "disabled" services; Migrate from "old" launchd
by Jerry Krinock 24 Aug '15

24 Aug '15

> That's strange. In the past, launchd would forget about just about everything except the installed .plists after a restart. But I suspect the new launchd is now configured by a persistent database. Hmmm. I have not tried a restart. > Next, I'd try the new print-cache command and see if your service names are in there. They are not. > If they are, you might try the new uncache <service-name> command to see if you can get launchd to forgettaboutit. Tried it anyhow. Result: Command is not yet implemented. After all that, I tried a restart, launchctl uncache and launchctl remove again. Result: No change. I also counted the number of disabled services my app has using "grep -c". Result: 3060. That’s three thousand sixty. Eeek. What are these doing? How can I remove them?

2 1

launchd logging?
by Nathan Bailey 30 Jul '15

30 Jul '15

Hi all, All of my launchd jobs log to err and out files, but I'd like to get a better sense of how they are running (or not). I saw there used to be launchd logging levels: http://apple.stackexchange.com/questions/169404/set-launchd-log-level but these have gone away in 10.10. Is there any plans to reintroduce them? thanks! Nathan

1 0

Yosemite not starting a system launch agent
by James Bucanek 23 Jul '15

23 Jul '15

Greetings, Here's my problem in a nutshell: I have a simple scheduler daemon that runs in the background for logged in users. The problem is that it won't start when installed in Yosemite (seems to start in all previous versions of OS X that use launchd). Here's what my program does: 1) I install the com.qrecall.sheduler.plist in /Library/LaunchAgents/ (with the correct ownership and access privileges). 2) launchctl load -S Background /Library/LaunchAgents/com.qrecall.scheduler.plist (successful) 3) launchctl start com.qrecall.scheduler (exits with status 1) And the scheduler process is NOT started. All attempts to get it started through launchctl have failed. However, I know that it's installed because if you simply restart the system the scheduler deamon starts running like a champ. Now ... I know what you're going to say: "Dude! You should be using the new launchctl commands introduced in 10.10!" Tried that. I can't seem to get anywhere with it. Basically, I don't know what I'm doing. I've read the man page for the new launchctl about 20 times and I'm still not sure how I should be using it. For per-user agents, it seems pretty obvious that you would bootstrap and address the service in the user/login/gui domain (i.e. launchctl enable user/501/com.qrecall.scheduler). But I don't understand how to treat a system-wide user agent that should be installed in /Library/LaunchAgents. Through trial and error, I've discovered that I can address the current user's instance of the scheduler process for commands like "kill". So this command works: launchctl kill TERM user/501/com.qrecall.scheduler It sends a TERM signal and the daemon restarts. However, I can't find anyway to use the bootstrap command to install it, the enable command won't start it, and the kickstart command says there's no such service. It would seem to make sense that a system-wide agent would have to be installed in the system domain, but this command: sudo launchctl bootstrap system /Library/LaunchAgents/com.qrecall.scheduler results in the error "/Library/LaunchAgents/com.qrecall.scheduler.plist: Service cannot load in requested session" Attempts to use any other domain in the command result in a syntax error, so I'm completely stumped as to how the service should be installed. I've figured out the I can enable and disable the per-user instance, but it has no effect on the running process. In other words, disabling won't halt the service and enabling it won't start it. And don't get me started on unbootstrap, which doesn't appear to be implemented. So, at this point I have my feet planted firmly in mid-air... James Bucanek

3 11