when the number of channels fill the MTU on a given link.
In the future, this needs to be configurable per peer with trunking enabled.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@46195 65c4cc65-6c06-0410-ace0-fbb531ad65f3
appreciated really. (Read the coding guidelines).
I've worked hard to make chan_sip a better place to code in, let's
keep it that way and don't add more stuff without comments.
Thank you.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@46183 65c4cc65-6c06-0410-ace0-fbb531ad65f3
simplifies the manager queue handling as described in
the comment, and will make a lot easier to make further
work on this code.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@46120 65c4cc65-6c06-0410-ace0-fbb531ad65f3
https://origsvn.digium.com/svn/asterisk/branches/1.4
........
r46067 | russell | 2006-10-23 21:27:42 -0400 (Mon, 23 Oct 2006) | 7 lines
In muted.c, check the return value of strdup. In ael_main.c, check the return
value of calloc. (issue #8157)
In passing fix a few minor bugs in ael_main.c. The last argument to strncpy()
was a hard-coded 100, where it should have been 99. I changed this to use
sizeof() - 1.
........
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@46068 65c4cc65-6c06-0410-ace0-fbb531ad65f3
so fix the places where this might happen.
This is also a fix that ought to go into 1.4
[The difference between the two functions is a bit confusing,
and in asterisk i believe all string handling functions
should be able to handl a NULL string as argument,
but changing the API in trunk and not in 1.4 would make
backporting harder.]
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@46045 65c4cc65-6c06-0410-ace0-fbb531ad65f3
As discussed on the mailing lists, 0 is a legal value
for Cseq, so there is no point to treat it specially.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@46035 65c4cc65-6c06-0410-ace0-fbb531ad65f3
for NULL is certainly wrong and usually disables the
checks that we want to make instead.
This commit fixes a number of the above bugs where the result
of get_header() is immediately checked for NULL.
This is certainly a candidate for merging into 1.4
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@46026 65c4cc65-6c06-0410-ace0-fbb531ad65f3
On passing, remove a wrong comment (that probably I wrote
myself!) and introduce a temporary variable to avoid a
misleading cast.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@46003 65c4cc65-6c06-0410-ace0-fbb531ad65f3
simplifies its usage.
+ add another client for parse_uri, in handling Contact: strings
(on passing, document the content of the "fullcontact" field);
+ in register_verify(), mark with XXX what i believe is another
misinterpretation on the URI format when '@' is missing.
No code changed here, so no fixes applied.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@45977 65c4cc65-6c06-0410-ace0-fbb531ad65f3
fix parse_uri() to interpret a missing userinfo
section as a domain-only URI, and comment a wrong
interpretation of the above in check_user_full().
The function has been patched to preserve the existing
behaviour (in what admittedly is a corner case, but
could be received under attacks).
Hopefully the From: based matching will go away soon!
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@45976 65c4cc65-6c06-0410-ace0-fbb531ad65f3
before splitting around the @, otherwise the refer_to_domain
might contain arguments as well, causing failures.
I think this is a true bug that ought to be fixed in 1.4 as well.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@45966 65c4cc65-6c06-0410-ace0-fbb531ad65f3
introducing the function parse_uri() that splits
a URI in its components.
Right now use it only in one place, because the custom
parsing that is done here and there sometimes has
bugs that i want to figure out first.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@45965 65c4cc65-6c06-0410-ace0-fbb531ad65f3
apart from a small (but disabled by default) new option.
In detail:
+ introduce a new value for enum check_auth_result, AUTH_DONT_KNOW,
used (read below) when a function does not have a conclusive response.
Possibly this is the same as AUTH_NOT_FOUND, but need to check further.
+ move the large blocks (checking in the users list and in the peers
list, respectively) from check_user_full() to separate functions.
They return AUTH_DONT_KNOW in case they don't find a match, so
the caller know that it has to try the next method.
There is still some duplication of code here, but i
have not tried yet to remove it.
+ [new option] a new option in sip.conf, match_auth_username,
has been introduced, and disabled by default.
If set, and the incoming request carries authentication info,
the username to match in the users list is taken from there
rather than from the From: field.
This change is easy to identify, being made of
- one line to declare the variable match_auth_username
- a block of 15 lines in check_user_full()
- one line in sip list settings
- two lines for parsing the config file.
check_user_full() is now a lot cleaner - basically a sequence of
checks that are applied to the request. This will help future
work with new matching schemes.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@45945 65c4cc65-6c06-0410-ace0-fbb531ad65f3
A large block needs reindentation now, but we don't do that because
it can be moved to a separate function.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@45905 65c4cc65-6c06-0410-ace0-fbb531ad65f3
and support linux as well (using fopencookie(), which should
be available in glibc).
Update configure.ac to check for funopen (BSD) and fopencookie(glibc),
and while we are at it also for gethostbyname_r
(the generated files need to be updated, or you need
to run bootstrap.sh yourself).
Document the new options in http.conf.sample
(names are only tentative, better ones are welcome).
At this point we can safely enable the option.
Anyone willing to try this on Sun and Apple platforms ?
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@45892 65c4cc65-6c06-0410-ace0-fbb531ad65f3
The changes are not large. Most of the diff comes from putting the
global variables describing an accept session into a structure, so
we can reuse the existing code for running multiple accept threads
on different ports.
Once this is done, and if your system has the funopen() library
function (and ssl, of course), it is just a matter of calling
the appropriate functions to set up the ssl connection on the
existing socket, and everything works on the secure channel now.
At the moment, the code is disabled because i have not implemented yet
the autoconf code to detect the presence of funopen(), and add -lssl
to main/Makefile if ssl libraries are present. And a bit of documentation
on the http.conf arguments, too.
If you want to manually enable https support, that is very simple
(step 0 1 2 will be eventually detected by ./configure, the
rest is something you will have to do anyways).
0. make sure your system has funopen(3). FreeBSD does, linux probably
does too, not sure about other systems.
1. uncomment the following line in main/http.c
// #define DO_SSL /* comment in/out if you want to support ssl */
2. add -lssl to AST_LIBS in main/Makefile
3. add the following options to http.conf
sslenable=yes
sslbindport=4433 ; pick one you like
sslcert=/tmp/foo.pem ; path to your certificate file.
4. generate a suitable certificate e.g. (example from mini_httpd's Makefile:
openssl req -new -x509 -days 365 -nodes -out /tmp/foo.pem -keyout /tmp/foo.pem
and here you go:
https://localhost:4433/asterisk/manager
now works.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@45869 65c4cc65-6c06-0410-ace0-fbb531ad65f3
reply back to http clients.
Use fprintf/fwrite instead, since we are already using a FILE *
to read the input.
If you wonder why, this is because it makes it trivial to
implement https support (as long as your system has funopen()).
And this is what i am going to put in with the next few commits...
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@45858 65c4cc65-6c06-0410-ace0-fbb531ad65f3
lock these data structures.
This improve readability, and also hides the underlying
locking mechanism so it is a lot easier to add diagnostic
code, or move the object locks somewhere else, etc.
On passing, rename the lock field in sip_pvt to pvt_lock,
also for ease of readability.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@45786 65c4cc65-6c06-0410-ace0-fbb531ad65f3
to minimize diffs with future modifications.
The current implementation is problematic for the following reasons:
+ all insertions are O(N) because the event list does not have a tail
pointer;
+ there is only a single lock protecting both session and users queues.
+ the implementation of the queue itself is not documented.
I think i have figured it out, more or less, but am unclear on
whether there is proper locking in place
The rewrite (which i have working locally) uses a tailq so insertions
are O(1), separate locks for the event and session queues, and has
a documented implementation so hopefully we can figure out if/where
bug exist.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@45753 65c4cc65-6c06-0410-ace0-fbb531ad65f3
