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

Re: ipfw2 and fwd in Darwin
by Giuliano Gavazzi 13 Oct '08

13 Oct '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com to show how the ipfw2 fwd behaviour can be disconcerting (and buggy): on the darwin server with an alias 192.168.202.102 on en1: % netstat -an but this connection does actually correspond to: % w ... root s001 192.168.202.102 21:34 - w that is a connection from itself. % ssh -b 192.168.202.102 88.149.157.x and the firewall rule causing this: Another effect is: even if the system know very well how to get to the next hop: Giuliano _______________________________________________ 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… ... tcp4 0 0 88.149.157.x.22 192.168.202.x.49206 ESTABLISHED tcp4 0 0 192.168.202.102.49206 88.149.157.128.22 ESTABLISHED ... That is, netstat "thinks" we are connected to a remote machine, but we are really connected to the local machine. The command given was: % ipfw add fwd 192.168.202.110 log src-ip 192.168.202.101,192.168.202.102 not dst-ip 192.9.200.0/24,192.168.202.0/24 via en1 Searching the web I found several posts relating ipfw2 fwd problems, but no solution we can apply. Instead in the man page I found: The fwd action does not change the contents of the packet at all. In particular, the destination address remains unmodified, so packets forwarded to another system will usually be rejected by that system unless there is a matching rule on that system to capture them. For packets forwarded locally, the local address of the socket will be set to the original destination address of the packet. This makes the netstat(1) entry look rather weird but is intended for use with transparent proxy servers. so it seems that my case would be explained if the packet was forwarded locally, but this is not the case as 192.168.202.110 is NOT a local address. % traceroute -s 192.168.202.102 88.149.157.x traceroute to 88.149.157.x (88.149.157.x) from 192.168.202.102, 64 hops max, 40 byte packets 1 localhost (127.0.0.1) 0.786 ms 0.159 ms 0.060 ms # traceroute -s 192.168.202.102 192.168.202.110 traceroute to 192.168.202.110 (192.168.202.110) from 192.168.202.102, 64 hops max, 40 byte packets 1 192.168.202.110 (192.168.202.110) 1.094 ms 0.431 ms 0.396 ms This email sent to site_archiver(a)lists.apple.com

1 0

Re: Where to save temporary files?
by Kevin Van Vechten 13 Oct '08

13 Oct '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Oct 12, 2008, at 7:07 PM, Jason Coco wrote: On Oct 12, 2008, at 10:59 , Andre-John Mas wrote: Hi, - Kevin _______________________________________________ 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… I am try to find out what the right place is to save a temporary file on the Mac is. I have looked at the following places: /tmp ~/Library/Caches/<app name> Are there other locations and are there any suggested usage policies? I would use confstr(3) with _CS_DARWIN_USER_TEMP_DIR _CS_DARWIN_USER_TEMP_DIR has similar semantics to /tmp, but is more secure as that portion of the filesystem is owned by the user. Both of these locations get cleaned up periodically and at reboot. Caches typically need to survive past reboot. ~/Library/Caches is a good location for any cache that contains a user's private data. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Problems with aio_read
by Richard Schmitt 13 Oct '08

13 Oct '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com 2. > I downloaded the darwin kernel source May I ask for the link where I can download the kernel source? Not sure why anyone would know this, and not sure why I knew it either, but the kernel source is in the package called xnu. So, for 10.5.5 you can get the source at: http://www.opensource.apple.com/darwinsource/10.5.5/xnu-1228.7.58/ Rich _______________________________________________ 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: Problems with aio_read
by Michael Andronov 13 Oct '08

13 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:date:from:sender :to:subject:cc:in-reply-to:mime-version:content-type :content-transfer-encoding:content-disposition:references :x-google-sender-auth; bh=mLKLi4lsWxfsXzmrrw/+k5sQm9jg/eVaExXjNkpEebo=; b=cE4TsC00yds+Ad30uwayzMKa8dT5Ua9U6Cy9gOExczLRzRXA+JGth6/pa9SSyxgeyF NSAp79OgmeRcmRKqkwrT206gqOJBFFi4sRJ4RDgH+q0JoA1NedutLUuoNw38GB6QdsgP qt1WF3nRYqPy1LslGkUF3Uq/4kOdaubVz4ku0= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:sender:to:subject:cc:in-reply-to:mime-version :content-type:content-transfer-encoding:content-disposition :references:x-google-sender-auth; b=GMT/2Bu/k9Ha2punSsZ2hm7V7zIJFzzKlrC6uJ6tk5G+8+HyAtbLusPI9cF8nlqqfN OGHpMt99d/QG5v+B5HPMoncTChCe0Nh5+/rXIYyU4lNj2BjqFPTAwFhbgXNlA8M/99L3 l8StUbru1KFPB7Ek3+QF/GdZm0CHJsZ93rj/M= Hi, 1. I've run into the same issue on OS X 10.5 - failed to forward the context ( user_siginfo_t . user_sigval) to the SIG handle. ( And to this end failed to know within the handle which aiocb completion was signaled). The work around I did also the same - my own queue of aiocb, signaled by MACH semaphore from the SIG handle. I read the message above and got the filling that in OS X 10.5 may be an easier solution. Am I missing something ? Is there the way to provide a correct user_sigval to the SIG handle? 2. > I downloaded the darwin kernel source May I ask for the link where I can download the kernel source? Thanks. Michael _______________________________________________ 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: Where to save temporary files?
by Jason Coco 12 Oct '08

12 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:cc:message-id:from:to :in-reply-to:content-type:mime-version:subject:date:references :x-mailer; bh=n8z1bEw6Y8DC02NbYKxh9HgvOoFX6yXgJG1ZhC2GZUI=; b=KNfk8w9dpO0d59BcfCGNBgDA5zbgIQPEg1rVMpgz676uqPGA1UDfW5YiaAMQABFCEe OoRpzgh1bd8U2K6aYpoGr0l8SrUY8H89L5fLOrpKaNBGe5DZ2diDMX5JB6RMbpjyIwV5 FJW0ZS//M9zPHzHSg3iU1CtQgbx8nHNUnd/VU= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=cc:message-id:from:to:in-reply-to:content-type:mime-version:subject :date:references:x-mailer; b=d9jOWptJioPE3NyADWgKn7pE5wS6RWm01buH6fT/WK44+YV/7ObUdyTIr2rUn9pbA6 K49s8alnB8AxCQEb8RCHrBPzpfw0XVNHwQ/DoZyozVrmlqJGPUUDvedZeR+Zq0fl4mhP PB5+YaVzGoGHApaplphERNrR5L5RsIcbe858s= On Oct 12, 2008, at 10:59 , Andre-John Mas wrote: Hi, I would use confstr(3) with _CS_DARWIN_USER_TEMP_DIR _______________________________________________ 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 am try to find out what the right place is to save a temporary file on the Mac is. I have looked at the following places: /tmp ~/Library/Caches/<app name> Are there other locations and are there any suggested usage policies? smime.p7s

1 0

ipfw2 and fwd in Darwin
by Giuliano Gavazzi 12 Oct '08

12 Oct '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hello, back in 10.3.9 given this scenario: ipfw add fwd GW_IP ip from ALIAS_BACKUP telnet -s ALIAS_BACKUP www.apple.com 80 Any ideas? but that is the subject for a bug report I just submitted] Giuliano _______________________________________________ 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… Server with two aliases on en1: ALIAS_DEFAULT is a public address and is on the same logical network as the default gateway, ALIAS_BACKUP is a private address and is on the same logical network as a natting adsl router (GW_IP). It is irrelevant here that one is a public address and the other is a private one. What matters is that by setting the simple rule: would forward packets that have source ALIAS_BACKUP to the natting gateway. [please also ignore that I called this a BACKUP, the firewall rules I use actually decide which router to use according to the type of traffic] Well, in Darwin Kernel Version 9.5.0 (10.5.5, but it also seemed not to work in 10.4.x), the packets processed by the rule never appear on the en1 interface. For example if I do a tcpdump -i en1 does not show any packets. The firewall rule is called but it seems to not have any effect. The result to the telnet command is Connection refused, that is a tcp reset has been received, but I can see this RST only in the tcpdump -i lo0 as if it came from the target address, while the target address never sees the packet. [I know there might be other problems in ipfw2 as I can manage to consistently panic with: panic(cpu 0 caller 0x00153E50): "-- unknown opcode 116\n"@/SourceCache/ xnu/xnu-1228.7.58/bsd/netinet/ip_fw2.c:2163 This email sent to site_archiver(a)lists.apple.com

1 0

Re: Where to save temporary files?
by Philip Aker 12 Oct '08

12 Oct '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Oct 12, 2008, at 7:59 AM, Andre-John Mas wrote: Are there other locations and are there any suggested usage policies? Philip Aker echo astwta@lvpc.dslh@nl | tr a-z@. p-za-o.@ Democracy: Two wolves and a sheep voting on lunch. _______________________________________________ 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… I am try to find out what the right place is to save a temporary file on the Mac is. I have looked at the following places: /tmp ~/Library/Caches/<app name> To elaborate: what I am looking at is a file to be downloaded by a web browser, in order to directly be opened up by another application. In order for the other application to have access to it I will have to save it somewhere, though I am not sure where I should be saving it and what sort of life-span this file should have on the user's hard-disk. For the circumstance, I think it would be: defaults read com.apple.Safari DownloadsPath (use realpath on it). Several levels up there would be FSFindFolder() in Folders.h (kDownloadsFolderType, kCachedDataFolderType, etc.). Also there's $TMPDIR but you'd probably want to create your own subfolder in it. I'm not sure if it would be in keeping with standard usage for that folder though. This email sent to site_archiver(a)lists.apple.com

1 0

Where to save temporary files?
by Andre-John Mas 12 Oct '08

12 Oct '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hi, Are there other locations and are there any suggested usage policies? André _______________________________________________ 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… I am try to find out what the right place is to save a temporary file on the Mac is. I have looked at the following places: /tmp ~/Library/Caches/<app name> To elaborate: what I am looking at is a file to be downloaded by a web browser, in order to directly be opened up by another application. In order for the other application to have access to it I will have to save it somewhere, though I am not sure where I should be saving it and what sort of life-span this file should have on the user's hard- disk. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Problems with aio_read
by Richard Schmitt 11 Oct '08

11 Oct '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com I downloaded the darwin kernel source and saw a comment in kern_aio.c: /* validate aiocb.aio_sigevent. at this point we only support sigev_notify * equal to SIGEV_SIGNAL or SIGEV_NONE. this means sigev_value, * sigev_notify_function, and sigev_notify_attributes are ignored. */ So, wow, a comment that answered my question. Too bad this isn't in the man pages for aio. My workaround for 10.4 is to create my own completion queue and when I get a signal, I'll traverse the completion queue looking for aiocb's that completed. What I was hoping to do was get the signal with the aiocb as the sig value. It saves me from having to maintain the completion queue. If there's a better way, I'd like to hear it. I might decide to restrict the application to 10.5, but I was hoping to make it work on 10.4. Thanks for your response. Rich ----- Original Message ---- From: Terry Lambert <tlambert(a)apple.com> To: Richard Schmitt <rschmitt(a)bluepeach.com> Cc: "darwin-dev(a)lists.apple.com" <darwin-dev(a)lists.apple.com> Sent: Saturday, October 11, 2008 12:19:25 PM Subject: Re: Problems with aio_read On Oct 11, 2008, at 5:50 AM, Richard Schmitt <rschmitt(a)bluepeach.com> wrote: > I'm trying to use aio_read in an application and am having problems > with the completion notifications. > > When I use signals, I'll get a signal for the completion, but the > siginfo structure does not have the si_value filled in. > > When I use notify functions, the aio_read call fails with invalid > argument. > > Here's some sample code: > > my_action.sa_sigaction = my_handler ; > my_action.sa_flags = SA_SIGINFO ; > sigaction (SIGIO, &my_action, NULL) ; > > cb = malloc (sizeof (struct aiocb)) ; > memset (cb, '\0', sizeof (struct aiocb)) ; > cb->aio_fildes = open ("sigtest.c", O_NONBLOCK) ; > cb->aio_buf = malloc (1001) ; > cb->aio_nbytes = 1001 ; > cb->aio_offset = 0 ; > > cb->aio_sigevent.sigev_value.sival_ptr = cb ; > cb->aio_sigevent.sigev_signo = SIGIO ; > cb->aio_sigevent.sigev_notify = SIGEV_SIGNAL ; > > ret = aio_read (cb) ; > > When the IO completes, I expect a SIGIO signal and the siginfo > structure passed to my_handler will have the si_value set to the > cb. The example handler is below: > > void my_handler (int sig, siginfo_t *siginfo, void* context) > { > struct aiocb *cb ; > > cb = siginfo->si_value.sival_ptr ; > > cb is always NULL. > > I'm running OSX 10.4.11 The aio implementation in 10.4 is not fully POSIX conformant. If your code needs full POSIX conformance, it should be including <unistd.h> and checking _POSIX_ASYNCHRONUS_IO has a value greater than or equal to 200112L before using the interface. There are a number of workaround that you could use if you wanted to run on something as old as 10.4, but you would need to explain the problem you are trying to solve for us to tell you which ones might be appropriate. -- 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… This email sent to site_archiver(a)lists.apple.com

1 0

Re: Problems with aio_read
by Terry Lambert 11 Oct '08

11 Oct '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Here's some sample code: my_action.sa_sigaction = my_handler ; my_action.sa_flags = SA_SIGINFO ; sigaction (SIGIO, &my_action, NULL) ; cb = malloc (sizeof (struct aiocb)) ; memset (cb, '\0', sizeof (struct aiocb)) ; cb->aio_fildes = open ("sigtest.c", O_NONBLOCK) ; cb->aio_buf = malloc (1001) ; cb->aio_nbytes = 1001 ; cb->aio_offset = 0 ; cb->aio_sigevent.sigev_value.sival_ptr = cb ; cb->aio_sigevent.sigev_signo = SIGIO ; cb->aio_sigevent.sigev_notify = SIGEV_SIGNAL ; ret = aio_read (cb) ; void my_handler (int sig, siginfo_t *siginfo, void* context) { struct aiocb *cb ; cb = siginfo->si_value.sival_ptr ; cb is always NULL. -- 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 11, 2008, at 5:50 AM, Richard Schmitt <rschmitt(a)bluepeach.com> wrote: I'm trying to use aio_read in an application and am having problems with the completion notifications. When I use signals, I'll get a signal for the completion, but the siginfo structure does not have the si_value filled in. When I use notify functions, the aio_read call fails with invalid argument. When the IO completes, I expect a SIGIO signal and the siginfo structure passed to my_handler will have the si_value set to the cb. The example handler is below: I'm running OSX 10.4.11 The aio implementation in 10.4 is not fully POSIX conformant. If your code needs full POSIX conformance, it should be including <unistd.h> and checking _POSIX_ASYNCHRONUS_IO has a value greater than or equal to 200112L before using the interface. There are a number of workaround that you could use if you wanted to run on something as old as 10.4, but you would need to explain the problem you are trying to solve for us to tell you which ones might be appropriate. This email sent to site_archiver(a)lists.apple.com

1 0