October 2008 - Darwin-dev - lists.apple.com

launchd and programmatic access
by Jason Coco 03 Oct '08

03 Oct '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:message-id:from:to :content-type:mime-version:subject:date:x-mailer; bh=3DTpok0ZVI6/ixfnRHsWnnK35AuHFOHbSpp42hbBi/4=; b=MGhXMqzxTfr8iJHQr/MyG9PJU+kcqGazQ5RXgXAmA5yPk7ACliJ+cBTLCA6oKmy6RN PlpJtTdl63g2Uz5yg3KpGaDX0twWEMkMXHYueorh7yw8JD9MZVHgi8BKH0YwVIb9lpnQ 6nYxkHz01a+9l2ft7mGz+eTuQZRU935Mb1uEE= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:from:to:content-type:mime-version:subject:date:x-mailer; b=asUo/mKSwpR/FWTuS5Ss2pAb6VmAbr67Or5ec8VDoW5O1KqLm3k2T2tNTe+nHFKwfG QYW5dH3eN2xH0V+AX5n56RXt4nwkWAloUvTbH45zdw2skkUcqbeY0/PYW0bwZRhUunUz r0ZUYt/JRaXBW1auG+yPkpfMDiIf0HLaCCek8= Hello all, I have a question about the launchd API from launch.h... background: Thanks, Jason _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com I have a daemon application that is using launchd but is a cross- platform application so I can't make it totally launchd aware (it's written in Java). It has a plist that lives in /Library/LaunchDaemons and starts up properly without any issues. My problem is that I have created a preference pane that I want the console use to be able to use to get status, set some configuration options and start/stop the service. The pref pane includes a setuid helper application that does proper authorization, etc. however, I can't seem to find my job through the launchd API in order to perform operaitons on it (start/stop, list status, etc). I'm thinking this is a mach-namespace issue... is there any way to resolve this? To force the API to use the root namespace??? I really don't want to write tons of hacks/scripts etc if possible. I am running on 10.5.5. If any other information is needed to assist, please let me know. smime.p7s

1 0

[darwin-dev] Building dyld: /usr/local/system/lib/libc.a ?
by Alexander von Below 02 Oct '08

02 Oct '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hello, I am trying to build dyld from source (1). Yes, /usr/local Thanks for any hints Alex 1) http://www.opensource.apple.com/darwinsource/10.5.5/dyld-96.2/ _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… Forgive me if this is a stupd or silly question, but when just pressing the Build button, the linker looks for /usr/local/system/lib/ libc.a Which prerequisites do I need to build dyld? Who puts libc.a into that location? This email sent to site_archiver(a)lists.apple.com

1 0

Re: launchd daemon, spawning children?
by Terry Lambert 02 Oct '08

02 Oct '08

site_archiver(a)lists.apple.com Delivered-To: Darwin-dev(a)lists.apple.com You probably actually want: As I said before, it gets complicated. -- Terry _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… On Oct 1, 2008, at 10:30 PM, Kevin Van Vechten <kvv(a)apple.com> wrote: On Oct 1, 2008, at 12:21 PM, Terry Lambert wrote: setgid(); syscall(SYS_initgroups, int ngroups, const gid_t *gidset, uid_t gmuid) Calling initgroups(3) is a better idea. It's not recommended to make syscalls directly. Doing so might not be compatible with future releases of Mac OS X. Kevin is of course right about this. But it adds substantially to the complications I was trying to avoid explaining before. The parameters for initgroups(3) have a string name, which requires using getpwuid(3) to convert the uid into a string name so the initgroups(3) can call getpwnam(3) to convert the string name to a uid so it can call the SYS_initgroups call under the covers. There is a fundamental assumption that the two "get" calls will be reflexive for any directory service; that is, a given name does not map to more than one uid, and a given uid reverse maps to only one name. As an implementation detail, the SYS_initgroups itself could end up changing or going away entirely. By taking the string name, it also takes the decision of what supplementary groups other than the primary group will be in the list if 16 out of your hands and gives it to libinfo which gives it to Directory Services. These are in turn the ones which get proxied over the wire for certain network filing protocols, e.g. NFS, and if the server side does not support external group membership resolution (e.g. NetApp filers, or clients in a firewall DMZ talking to a server interior to the firewall between the DMZ and "inside"), or does but, is not a member of the same security association as the one granting the credential via libinfo (including group initialization by the initgroups(3) call), then you can lose on the reordering from the indirect call. This is particularly true if the server is doing the access control enforcement on a file or container object based on group rights and group ID, and it's not your primary effective group at the time of the call (you would potentially be denied access). On top of this, we also reserve the right to change the little dance programs must do to establish sessions from a Minuet to a Foxtrot or The Froog at some point in the future. For interactive sessions, we've thrown all our eggs in the login basket. For non-interactive sessions, like user space network file services daemons, we've had to put the knowledge into the server directly, or punt back to straight POSIX. The flip side is that if you limit yourself to putting users in NGROUPS only, traditional POSIX semantics apply, but you lose the functionality of extended groups lists and of groups of groups that you get by participating in external group resolution. This email sent to site_archiver(a)lists.apple.com

1 0

VFS
by Karan, Cem (Civ, ARL/CISD) 02 Oct '08

02 Oct '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Thread-index: AckkoIgwrJDdn/bXTrCXrR2poLAcMQ== Thread-topic: VFS _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0

Re: launchd daemon, spawning children?
by Kevin Van Vechten 02 Oct '08

02 Oct '08

site_archiver(a)lists.apple.com Delivered-To: Darwin-dev(a)lists.apple.com On Oct 1, 2008, at 12:21 PM, Terry Lambert wrote: You probably actually want: _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… setgid(); syscall(SYS_initgroups, int ngroups, const gid_t *gidset, uid_t gmuid) Calling initgroups(3) is a better idea. It's not recommended to make syscalls directly. Doing so might not be compatible with future releases of Mac OS X. If you do not set the group membership uid, and do it without any other modifications of the credential, save the final setuid(), then you will opt out of external group resolution. This means that if you have put a user into more than 16 groups in DirectoryServices, you wil only see 16 of them, and due to ordering ambiguity, you will only be guaranteed that the primary group ID for the user will be in that list of 16. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Open Firmware on Pismo
by James Little 01 Oct '08

01 Oct '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=talk21.com; h=X-YMail-OSG:Received:X-Mailer:Date:From:Reply-To:Subject:To:MIME-Version:Content-Type:Content-Transfer-Encoding:Message-ID; b=2ExoC97hZ/m6DdQzzOnDHkaw4bgDVbhA/aLkHhy+4xKRwQXnMTK+FV5rfXDi84Q0GsKlc+cXUZfSowGiqvg0aybIqGLumzKUy4ZOg/cdLl4169tWNen/GJAhrbXeIlk06jW2RybMG9YeAUIpaEgdlXMDJxOZUKp0EptzC4k7/G0=; Hi, Thanks very much. Using the commands, after amending the settings for thermal-info I got the fan to turn on straight after booting, by setting them to 01010200 00000000 01010200 00000000. With the open firmware dump below of the values for via-pmu / power-mgt could anyone explain what the values present could likely be. i.e. are the set of 4 hex values likely to be 2 sets of values, or 4 separate ones? I decoded each of the hex values and got 16843264 for hex 01010200 and 1213468963 for hex 48434843. I think these refer to temperature values (perhaps F?). ff918050: /pci@f2000000/mac-io@17/via-pmu@16000/power-mgt compatible 7669612d 706d752d 393900.. ........ "via-pmu-99" device_type 706f7765 722d6d67 7400.... ........ "power-mgt" max-aux-power 00000000 ........ ........ ........ .... name 706f7765 722d6d67 7400.... ........ "power-mgt" prim-info 000000ff 00000060 00003e80 00017fb5 .......`..>..... 0010: 0000d607 00000000 00021300 46000220 ............F.. 0020: 101400.. ........ ........ ........ ... registry-name 65787469 6e742d67 70696f31 00...... extint-gpio1. thermal-info 01010200 00000000 48434843 00000000 ........HCHC.... Any knowledge people may have on this and Open Firmware properties would be much appreciated. Thanks, James. ----- Original Message ---- From: Josh de Cesare <decesare(a)apple.com> To: James Little <jkal(a)talk21.com> Cc: darwin-dev(a)lists.apple.com Sent: Wednesday, 1 October, 2008 2:55:00 PM Subject: Re: Open Firmware on Pismo James, You can build up properties with encode+. The following code: 0 encode-int 1 encode-int encode+ 2 encode-int encode+ 3 encode-int encode+ should leave on the stack the length 4*4 and address of 16 bytes containing the long words 0, 1, 2, and 3. Josh On Oct 1, 2008, at 4:16 AM, James Little wrote: > Hi, > > I hope this is not too off-topic: > > I've been delving around in open firmware on my Pismo to see if I > can control the internal fan, or at least set a better temp. for it > to come on at. > > In the device tree I've found the via-pmu, and under this is a node > called 'power-mgt'. A property of this is called 'thermal-info'. > As far as I know this must be related to the temp sensor in the > Pismo because the G4 chip I have in it doesn't have the thermal > assist temp diode on the chip enabled. > > To check that this was specific to the Pismo I found the same 'power- > mgt' property on a Digital Audio G4, which had everything listed the > same, apart from the 'thermal-info', which was missing. > > I've now hit a problem. There are 4 values of data encoded in open > firmware for the 'thermal-info'. I believe they are the limits for > the fan, as they do not change from boot up o when the computer is > definitely toasty. > > I can write one value by using the following command: > > [code]12345678 encode-int " thermal-info" property[/code] > > However, I need to be able to write all 4 values - and I don't know > the open firmware commands to do this... could anyone help? > > Thanks (and sorry for the rambling post), > > James. > > > > > _______________________________________________ > Do not post admin requests to the list. They will be ignored. > Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) > Help/Unsubscribe/Update your Subscription: > http://lists.apple.com/mailman/options/darwin-dev/decesare%40apple.com > > This email sent to decesare(a)apple.com _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0

Re: launchd daemon, spawning children?
by Terry Lambert 01 Oct '08

01 Oct '08

site_archiver(a)lists.apple.com Delivered-To: Darwin-dev(a)lists.apple.com Hi Terry and List, Terry Lambert wrote: The primary paper on UNIX setuid() is: Setuid Demystified Hao Chen, David Wagner, Drew Dean Proceedings of the 11th USENIX Security Symposium <www.cs.berkeley.edu/~daw/papers/setuid-usenix02.pdf> Thanks a lot, the paper is great! I finally ended up with something like (pseudocode): fork(); if(pid == 0) { setgroups(); setgid(); You probably actually want: setgid(); syscall(SYS_initgroups, int ngroups, const gid_t *gidset, uid_t gmuid) As I said, it gets complicated. setuid(); close_nonstd_file_descriptors(); execve("/path/to/worker"); _exit(); } else { while(1) { waitpid(WNONBLOCK); usleep(1000); } } This works very well, and makes sure the worker process can not gain higher permissions than it should have. Don't mind the while(1) above, in practice I have a network connection open between worker and server to get error-feedback, so the server won't get stuck. Your suggestion to use login seems reasonable. Can someone sketch an outline how an application (daemon) would use login to start a new process as a certain user? Is that what you meant: execve("/bin/login -f worker_user /path/to/worker") -- Terry _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… On Oct 1, 2008, at 3:40 AM, Mario Emmenlauer wrote: On Sep 26, 2008, at 2:37 AM, Mario Emmenlauer wrote: Damien Sorresso wrote: The rules of setuid(2) have evolved into a vastly complex morass with subtle variations on different platforms. Entire papers have been written about this behavior. You name it. I will look for a good tutorial, though Terry's suggestion contained a nice start. If you do not set the group membership uid, and do it without any other modifications of the credential, save the final setuid(), then you will opt out of external group resolution. This means that if you have put a user into more than 16 groups in DirectoryServices, you wil only see 16 of them, and due to ordering ambiguity, you will only be guaranteed that the primary group ID for the user will be in that list of 16. In general, we do not want you to establish your own sessions, and would prefer that you use /bin/login to do session management and support the utmp/utmpx interfaces for process accounting and knowing who is logged into your system, etc.; otherwise, things get more complicated fast, in order to support those interfaces, since you have to (1) maintain privilege in order to access the files to write out the closure records on sessions started by you (login does this - it hangs around until its children go away), and (2) deal with PAM and all the other things login deals with for authenticating you to the system. Typically, it's not used for daemons. daemons are expected to be started by launchd, which will do the session establishing. Establishing your own session is intended to be limited to interactive sessions. If you have more questions about interactive sessions and the login program, you should look at the sources for sshd or ftpd. One last question: If "worker_user" is the current GUI-user, all his applications get killed on logout. Is there a way around this (or would using 'login' also solve this issue?) The logout operation for non-GUI applications sends either a SIGHUP (the correct signal), or a SIGTERM (not the correct signal). You can ignore or catch these signals with a signal handler, although if you ignore SIGTERM, you will delay shutdowns. In general, daemons running as that user are not kiplled unless they are in the process group, or their parent is the per-session launchd. Otherwise, you would not be able to ssh into a machine and log out of a GUI and not have it kill your ssh (which it doesn't do). This email sent to site_archiver(a)lists.apple.com

1 0

Re: locale and file functions
by Mike Mackovitch 01 Oct '08

01 Oct '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: Mutt/1.5.18 (2008-05-17) On Wed, Oct 01, 2008 at 09:53:22AM -0700, Kevin Van Vechten wrote: > > On Oct 1, 2008, at 1:43 AM, Jonas Maebe wrote: > >> What is the extent of the interaction between the current locale >> setting (as in the LANG, LC_TYPE, etc environment variables) and the >> encoding expected/returned by posix functions such as open(), stat() >> etc? Are they always to be encoded using the current LC_TYPE, always >> using UTF-8, or yet something else? >> >> I also recently discovered >> http://developer.apple.com/documentation/CoreFoundation/Reference/CFStringR… >> (CFStringGetFileSystemRepresentation). It's perfectly usable when >> passing stuff to the posix api's, but when getting something back you >> still have to know what the used encoding was if you want to operate >> further on it. > > Mac OS X's posix filesystem API (open(2), stat(2), etc.) always expect > path names to be UTF-8 encoded (the system will normalize them to be > decomposed), and return path names which are UTF-8 encoded (decomposed). With the exception of some file systems which have no way of knowing/enforcing character encoding standards... like NFS, as noted in: > http://developer.apple.com/qa/qa2001/qa1173.html Also, strictly speaking, the VFS layer does no normalization. Any normalization happens in the the file system implementation or in system frameworks. I was hopeful that NFSv4 would at least be bringing NFS to UTF-8. But some implementations (Linux) have expressed that they have no intention of enforcing/implementing that part of the NFSv4 spec. :-( HTH --macko _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0

[Q] valloc bug?
by JongAm Park 01 Oct '08

01 Oct '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=Received:X-YMail-OSG:X-Yahoo-Newman-Property:Message-ID:Date:From:User-Agent:MIME-Version:To:Subject:Content-Type:Content-Transfer-Encoding; b=1zFuHuTtmZYln/KtMiEdpy13Y7h8XkiUARdODVm8Nycp6vqc+lbEtKrMv1nQOM+bi9XS5eHg7TwXl+Aqoi5kzyN5uI2uGrroBTJa9JjW9xdhjxvOVGyx7mGf9nUEDxchVgECwNlekpItgEf5ySwNgSwX9YMtC6xAhiihlsT6VL0= ; User-agent: Thunderbird 2.0.0.17 (Windows/20080914) Hello, all. I think I found a bug of the valloc(). I have lines of code which works like this. Are there any people who also notice this problem? Thank you. _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… 1. A hosting program calls my module when the user initiates "export" from the menu item of the hosting program. 2. The hosting program calls my module. Let's call the called function of the module as function "A". 3. For each slice of the files being exported, the hosting program calls the function "A" 4. My function "A" calls another function "B" to process things, while the "A" itself contains an "Init" function for set up things when it is first call for the file being exported, and "cleanup" functions for the last call for the file. 5. When the call is for the 1st one for the file, it allocates a fresh new memory space, and put the data into the memory space. 6. For every 4 calls, it saves the content of the memory space. This reduces time to save the export procedure. 7. for 2nd, 3rd calls, it "realloc" the memory to increase the memory space to make enough room for the new incoming data. 8. After saing the data, it frees the memory space. 9. Now, if the function is called again because there are still left data to process, it allocates a new memory space. 10. Now take the same procedure described above until the end of the file. When it allocates new memory space, it calls "valloc". One problem I saw happens when it frees old memory space after saving it and make a new memory space and start putting data into the space. When it puts data into the space, it sometimes crash while displaying error message which means, "the code tried modifing memory space which is freed." But I checked the call path and I confirmed that it allocates new memory space using valloc and it accesses the memory space. So, when I put some NSLog() function between the memory allocation and putting data into the space, it did not show the crash problem anymore. So, I replaced the valloc() with malloc() to see if the problem still exists. ( I removed the extra NSLog() function for beign sure. ) Then, it didn't show any crash problem. So, what I guess is.. while compilation the gcc may reorder the instruction and it makes "memcpy" call appear before the "valloc()" related call. ( memcpy() is used to put the data into the memory space. ) I use the Mac OS X 10.5.5 and Xcode 3.1.1. The gcc is the latest one which is installed with the Xcode tools. This email sent to site_archiver(a)lists.apple.com

1 0

Re: locale and file functions
by Jonas Maebe 01 Oct '08

01 Oct '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On 01 Oct 2008, at 18:53, Kevin Van Vechten wrote: More information is available at: http://developer.apple.com/qa/qa2001/qa1173.html http://developer.apple.com/qa/qa2001/qa1235.html Thanks, that's exactly what I wanted to know. Jonas _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… Mac OS X's posix filesystem API (open(2), stat(2), etc.) always expect path names to be UTF-8 encoded (the system will normalize them to be decomposed), and return path names which are UTF-8 encoded (decomposed). This email sent to site_archiver(a)lists.apple.com

1 0