sr-users April 2005

sr-users@lists.kamailio.org

250 participants
697 discussions

[Serusers] Carrier-grade framework for SER

by Andreas Granig

Hi guys, I'd like to propose another possibility for a highly-available and scalable system design based on SER: The problems I've encountered for scalable systems are: - Distribution of the user location and alias location among the nodes (user location is based on registrations, alias location comes from web interfaces and is used for call forwarding). - Reloading up to date location tables after breakdown and recovery of a node So I'm just thinking loud about the following provisioning system: - Write a client which fulfills the this demands: - Receive one or more locations from SER via a SER module or from a web application and distribute them to other known clients. Take care of retransmissions if a client isn't reachable or reports a temporary failure. - Receive one or more locations from other clients and write them into the SER FIFO. If writing into the FIFO fails, try to write directly into the database (location-table, alias-table etc.). Report a temporary failure if this also fails. Maybe a centralized server should be used which receives the locations from the clients and distributes them to other clients, so that the nodes just know about the server and nothing about other nodes. This would make integration of new nodes easy. On the other hand, it's another single point of failure, so a decentralized solution should be considered. But that would mean that you've to inform every node about the existence of a new node. The protocol used between the nodes should be simple and fast. So I think SOAP drops out here. Maybe XMLRPC or ICE (http://www.zeroc.com/ice.html) could be used. One might think now why not just use replication on SIP layer, but t_replicate only supports one peer and you've no possibility to get locations on a node while it's down. Replication of other location tables like the alias-table is also not possible. I'd be willing to release these parts as GPL for creating an open framework for carrier-grade SER integration, so any feedback, improvements or flames are highly welcome. Cheers, Andy

18 years, 2 months

[Serusers] Modifying "From" header field of an INVITE message

by scm-j＠nuntius.com

I am relatively new to SER. My SIP service provider requires me to present the ANI in the "From" field in my INVITE message header in E.164 format. I am currently sending this information as follows: From: "2125551212" <sip:2125551212@xxx.xxx.xxx.xxx> They want it in E.164 format as follows: From: "+12125551212" <sip:+12125551212@xxx.xxx.xxx.xxx> I see function only to modify the SIP URI but not the other header or SDP fields. Could someone help me with a code snippet that I can use in ser.cfg to do this conversion? Regards, SCM

18 years, 7 months

[Serusers] Authenticate to Active Directory

by Bob McCormick

Does anyone know if there is a way to authenticate SER to MS Active Directory? I'd like to use an existing user database instead of having to manage users myself, and most of our users are in Active Directory. Has anyone tried this? What are most SER users using for authentication? Is anyone here tying into any kind of existing organization user database?

19 years

[Serusers] Over-long fields can crash SER with PostgreSQL

by Karl A. Krueger

We are using SER with the PostgreSQL module, so that SER and our Asterisk PSTN gateway and voice mail system can use the same back-end database for user records. (Anyone who's interested in how we did this, let me know -- it's still a little shaky, but it pretty much works.) Recently we have encountered a problem: it seems that if any database error occurs, SER hangs or crashes. We have so far been able to mostly work around these conditions by being Really Careful with the database, but it is a little unsettling. The database schema we're working from is practically identical to the MySQL schema that ships with SER. Notably for this problem the lengths of all CHARACTER VARYING fields are the same. The User-Agent field is CHARACTER VARYING(50), for instance. So ... today boss got a copy of Xten's "eyeBeam" software, set it up to register to our SER ... and SER promptly locked up. I looked in the PostgreSQL database log and found this: Feb 10 13:39:32 mabell postgres[13058]: [267] ERROR: value too long for type character varying(50) Whoops. SER tried to insert eyeBeam's User-Agent data, PostgreSQL returned a data integrity error, and SER went bye-bye. So I did a little ALTER TABLE and adjusted the field to length 128 ... then put this in ser.cfg: # Disallow long user-agent if (search("^User-Agent: .{100,}") ) { log(1, "LOG: User agent too long\n"); sl_send_reply("479", "User agent too long!"); break; }; But this is obviously a kludge ... and doesn't fix the problem of one of the *other* fields being over-long. Anyone have any thoughts on this? Robustness patches for the PostgreSQL module or usrloc? Better ser.cfg workarounds to protect the world from malformed or over-long SIP header fields? Should I just change all the database fields to TEXT instead of CHARACTER VARYING(whatever) ? -- Karl A. Krueger <kkrueger(a)whoi.edu> Network Security -- Linux/Unix Systems Support -- Etc. Woods Hole Oceanographic Institution

19 years, 4 months

[Serusers] SER with Radius Authentication

by Rafael J. Risco G.V.

Hi, I´ve configured freeradius and SER according to the Radius HOW TO document, Accounting works very well but now I am doing some tests trying to do user authentication however all the authentication requests coming to the freeradius fails and X-lite sipphone is receiving an Unauthorized message from SER, please some advice, thanks rafael PS: config files... in /usr/local/etc/raddb/users : --------- test Auth-Type := Digest, User-Password == "test" Reply-Message = "Hello, test with digest" 6609876 Auth-Type := Digest User-Password := "9876", Digest-Response = "lalalalala", Reply-Message = "Hello, ibm1" 6604321 Auth-Type := Digest User-Password := "4321", Digest-Response = "lalalalala", Reply-Message = "Hello, ibm2" --------- Some relevant data in ser.cfg: ... modparam("group_radius", "use_domain", 0) .... if (uri==myself) { if (method=="REGISTER") { # Uncomment this if you want to use digest authentication if (!radius_www_authorize("")) { www_challenge("", "1"); break; }; if (!save("location")) { sl_reply_error(); }; break; }; lookup("aliases"); if (!uri==myself) { append_hf("P-hint: outbound alias\r\n"); route(1); break; }; # does the user wish redirection on no availability? (i.e., is he # in the voicemail group?) -- determine it now and store it in # flag 4, before we rewrite the flag using UsrLoc if (radius_is_user_in("Request-URI", "voicemail")) { log(1, "requested user is in voicemail group"); setflag(4); }; # native SIP destinations are handled using our USRLOC DB if (!lookup("location")) { # sl_send_reply("404", "Not Found"); log(1,"unable to locate user"); route(4); break; }; }; # End of "if(uri==myself)" .... ------------------RADIUSD -X Output ---------------------------: rad_recv: Access-Request packet from host 127.0.0.1:33187, id=79, length=311 User-Name = "6604321(a)10.0.1.22" Digest-Attributes = 0x0a0936363034333231 Digest-Attributes = 0x010b31302e302e312e3232 Digest-Attributes = 0x022a34323366333163373062336631643261643330383833633238303434316632663133643136613830 Digest-Attributes = 0x040f7369703a31302e302e312e3232 Digest-Attributes = 0x030a5245474953544552 Digest-Attributes = 0x050661757468 Digest-Attributes = 0x090a3030303030303162 Digest-Attributes = 0x08224433343132424232394131453131443939334232303035304241373836433642 Digest-Response = "a6a7812ac0331324f977453c228da2ed" Service-Type = IAPP-Register Sip-URI-User = "6604321" Cisco-AVPair = "call-id=D3412ADB9A1E11D993B20050BA786C6B(a)10.0.1.22" NAS-IP-Address = 127.0.0.1 NAS-Port = 5060 Processing the authorize section of radiusd.conf modcall: entering group authorize for request 8 modcall[authorize]: module "preprocess" returns ok for request 8 modcall[authorize]: module "chap" returns noop for request 8 modcall[authorize]: module "mschap" returns noop for request 8 rlm_digest: Converting Digest-Attributes to something sane... Digest-User-Name = "6604321" Digest-Realm = "10.0.1.22" Digest-Nonce = "423f31c70b3f1d2ad30883c280441f2f13d16a80" Digest-URI = "sip:10.0.1.22" Digest-Method = "REGISTER" Digest-QOP = "auth" Digest-Nonce-Count = "0000001b" Digest-CNonce = "D3412BB29A1E11D993B20050BA786C6B" rlm_digest: Adding Auth-Type = DIGEST modcall[authorize]: module "digest" returns ok for request 8 rlm_realm: Looking up realm "10.0.1.22" for User-Name = "6604321(a)10.0.1.22" rlm_realm: No such realm "10.0.1.22" modcall[authorize]: module "suffix" returns noop for request 8 rlm_eap: No EAP-Message, not doing EAP modcall[authorize]: module "eap" returns noop for request 8 users: Matched DEFAULT at 152 modcall[authorize]: module "files" returns ok for request 8 modcall: group authorize returns ok for request 8 rad_check_password: Found Auth-Type DIGEST auth: type "digest" Processing the authenticate section of radiusd.conf modcall: entering group authenticate for request 8 rlm_digest: Configuration item "User-Password" is required for authentication. modcall[authenticate]: module "digest" returns invalid for request 8 modcall: group authenticate returns invalid for request 8 auth: Failed to validate the user. Delaying request 8 for 1 seconds Finished request 8 Going to the next request --- Walking the entire request list --- Waking up in 1 seconds... rad_recv: Access-Request packet from host 127.0.0.1:33188, id=80, length=311 User-Name = "6609876(a)10.0.1.22" Digest-Attributes = 0x0a0936363039383736 Digest-Attributes = 0x010b31302e302e312e3232 Digest-Attributes = 0x022a34323366333163373062336631643261643330383833633238303434316632663133643136613830 Digest-Attributes = 0x040f7369703a31302e302e312e3232 Digest-Attributes = 0x030a5245474953544552 Digest-Attributes = 0x050661757468 Digest-Attributes = 0x090a3030303030303163 Digest-Attributes = 0x08224433343132424235394131453131443939334232303035304241373836433642 Digest-Response = "50fa695654b20e2eec54a1003fe15d9f" Service-Type = IAPP-Register Sip-URI-User = "6609876" Cisco-AVPair = "call-id=D3412ADE9A1E11D993B20050BA786C6B(a)10.0.1.22" NAS-IP-Address = 127.0.0.1 NAS-Port = 5060 Processing the authorize section of radiusd.conf modcall: entering group authorize for request 9 modcall[authorize]: module "preprocess" returns ok for request 9 modcall[authorize]: module "chap" returns noop for request 9 modcall[authorize]: module "mschap" returns noop for request 9 rlm_digest: Converting Digest-Attributes to something sane... Digest-User-Name = "6609876" Digest-Realm = "10.0.1.22" Digest-Nonce = "423f31c70b3f1d2ad30883c280441f2f13d16a80" Digest-URI = "sip:10.0.1.22" Digest-Method = "REGISTER" Digest-QOP = "auth" Digest-Nonce-Count = "0000001c" Digest-CNonce = "D3412BB59A1E11D993B20050BA786C6B" rlm_digest: Adding Auth-Type = DIGEST modcall[authorize]: module "digest" returns ok for request 9 rlm_realm: Looking up realm "10.0.1.22" for User-Name = "6609876(a)10.0.1.22" rlm_realm: No such realm "10.0.1.22" modcall[authorize]: module "suffix" returns noop for request 9 rlm_eap: No EAP-Message, not doing EAP modcall[authorize]: module "eap" returns noop for request 9 users: Matched DEFAULT at 152 modcall[authorize]: module "files" returns ok for request 9 modcall: group authorize returns ok for request 9 rad_check_password: Found Auth-Type DIGEST auth: type "digest" Processing the authenticate section of radiusd.conf modcall: entering group authenticate for request 9 rlm_digest: Configuration item "User-Password" is required for authentication. modcall[authenticate]: module "digest" returns invalid for request 9 modcall: group authenticate returns invalid for request 9 auth: Failed to validate the user. Delaying request 9 for 1 seconds Finished request 9 Going to the next request --- Walking the entire request list --- Waking up in 1 seconds... --- Walking the entire request list --- Sending Access-Reject of id 79 to 127.0.0.1:33187 Waking up in 1 seconds... rad_recv: Access-Request packet from host 127.0.0.1:33189, id=81, length=311 User-Name = "6609876(a)10.0.1.22" Digest-Attributes = 0x0a0936363039383736 Digest-Attributes = 0x010b31302e302e312e3232 Digest-Attributes = 0x022a34323366333163373062336631643261643330383833633238303434316632663133643136613830 Digest-Attributes = 0x040f7369703a31302e302e312e3232 Digest-Attributes = 0x030a5245474953544552 Digest-Attributes = 0x050661757468 Digest-Attributes = 0x090a3030303030303163 Digest-Attributes = 0x08224433343132424236394131453131443939334232303035304241373836433642 Digest-Response = "e4f68760f2b3eed0ad45942b32542c92" Service-Type = IAPP-Register Sip-URI-User = "6609876" Cisco-AVPair = "call-id=D3412ADE9A1E11D993B20050BA786C6B(a)10.0.1.22" NAS-IP-Address = 127.0.0.1 NAS-Port = 5060 Processing the authorize section of radiusd.conf modcall: entering group authorize for request 10 modcall[authorize]: module "preprocess" returns ok for request 10 modcall[authorize]: module "chap" returns noop for request 10 modcall[authorize]: module "mschap" returns noop for request 10 rlm_digest: Converting Digest-Attributes to something sane... Digest-User-Name = "6609876" Digest-Realm = "10.0.1.22" Digest-Nonce = "423f31c70b3f1d2ad30883c280441f2f13d16a80" Digest-URI = "sip:10.0.1.22" Digest-Method = "REGISTER" Digest-QOP = "auth" Digest-Nonce-Count = "0000001c" Digest-CNonce = "D3412BB69A1E11D993B20050BA786C6B" rlm_digest: Adding Auth-Type = DIGEST modcall[authorize]: module "digest" returns ok for request 10 rlm_realm: Looking up realm "10.0.1.22" for User-Name = "6609876(a)10.0.1.22" rlm_realm: No such realm "10.0.1.22" modcall[authorize]: module "suffix" returns noop for request 10 rlm_eap: No EAP-Message, not doing EAP modcall[authorize]: module "eap" returns noop for request 10 users: Matched DEFAULT at 152 modcall[authorize]: module "files" returns ok for request 10 modcall: group authorize returns ok for request 10 rad_check_password: Found Auth-Type DIGEST auth: type "digest" Processing the authenticate section of radiusd.conf modcall: entering group authenticate for request 10 rlm_digest: Configuration item "User-Password" is required for authentication. modcall[authenticate]: module "digest" returns invalid for request 10 modcall: group authenticate returns invalid for request 10 auth: Failed to validate the user. Delaying request 10 for 1 seconds Finished request 10 Going to the next request Sending Access-Reject of id 80 to 127.0.0.1:33188 Waking up in 1 seconds... --- Walking the entire request list --- Waking up in 1 seconds... --- Walking the entire request list --- Sending Access-Reject of id 81 to 127.0.0.1:33189 Waking up in 2 seconds... --- Walking the entire request list --- Cleaning up request 8 ID 79 with timestamp 423f309b Waking up in 1 seconds... --- Walking the entire request list --- Cleaning up request 9 ID 80 with timestamp 423f309c Waking up in 1 seconds... --- Walking the entire request list --- Cleaning up request 10 ID 81 with timestamp 423f309d Nothing to do. Sleeping until we see a request. -- rrgv

19 years, 6 months

[Serusers] Record-Route and Route parameters patch

by Bogdan-Andrei Iancu

We are all aware of SER limitation due the fact it is *only* transaction statefull at proxy level, any dialog support being missing. There are many features/services which actually requires a dialog awareness, like storing the information in the dialog creation stage, information which will be used during the whole dialog existence. The most urging example is NAT traversal, in dealing with the within the dialog INVITEs (re-INVITEs). When processing the initial INVITE, the proxy detects if the caller or callee is behind some NAT and fixes the signalling and media parts - since not all the detection mechanism are available for within the dialog requests (like usrloc), to be able to fix correspondingly the sequential requests, the proxy must remember that the original request was NAT processed. There are many other cases where dialog awareness fixes or helps....it's not my goal to make a list here.... The solution is to store additional dialog-related information in the routing set, headers which show up in all sequential requests; by routing set I would say the Route hdrs and RURI; the routing set is build based on Record-Route and Contact hdrs. So far, the way to go was using the Contact header (via parameters) as carrier. In most of the cases works, but this mechanism is not RFC reinforced - I had a talk with Juha and he found some phones which actually ignore the Contact hdr parameters, which is perfect allowed by RFC....Not to mention that which Contact hdr (UAC or UAS) appear in routing info depends on the request direction (UAC->UAS or UAS->UAC) and Record-Route parameters are re-infonced by RFC (see 12.1.1 UAS behavior) So the next try is to add this information in the Record-Route header inserted by the proxy. Within the dialog, this information can be found (with no direction dependencies) in Route header (with or proxy address)... And here is a patch for the RR module which adds this functionality: 1) adding parameters to Record-Route; since RR is inserted during req. processing (as lump), no textops function can be apply to it. So use the add_rr_param("param") function - it can be called also before or after the record_route() call; I would say it's more convenient from scripting point of view (no dependencies between doing record_route() and adding params); 2) checking Route parameters: this could be done via textops, but it would be dangerous -> you have to identify the Route header corresponding to your proxy (which will be quite difficult to put as regexp), So use check_route_param("regexp") function which will check the proper Route parameters against the given regexp. Must be call after loose_route(). Basic scenario (example): UAC SER PROXY UAS ------ INVITE ----------> record_route() --------- INVITE --------> add_rr_param(";foo=true") ----- reINVITE ---------> loose_route() -------- reINVITE -------> check_route_param(";foo=true") <---- reINVITE ---------- loose_route() <------- reINVITE -------- check_route_param(";foo=true") <-------- BYE ----------- loose_route() <--------- BYE ----------- check_route_param(";foo=true") The patch doesn't alter any of the previous functionality. Jan take a look and let's put it on CVS. bogdan Index: modules/rr/loose.c =================================================================== RCS file: /cvsroot/ser/sip_router/modules/rr/loose.c,v retrieving revision 1.36 diff -u -r1.36 loose.c --- modules/rr/loose.c 23 Feb 2005 17:16:05 -0000 1.36 +++ modules/rr/loose.c 15 Apr 2005 14:52:55 -0000 @@ -30,6 +30,8 @@ * --------- * 2003-02-28 scratchpad compatibility abandoned (jiri) * 2003-01-27 next baby-step to removing ZT - PRESERVE_ZT (jiri) + * 2005-04-10 check_route_param() and all hooks for keeping reference to + * Route params added (bogdan) */ @@ -57,6 +59,10 @@ #define ROUTE_SUFFIX ">\r\n" #define ROUTE_SUFFIX_LEN (sizeof(ROUTE_SUFFIX)-1) +/* variables used to hook the param part of the local route -bogdan */ +static unsigned int routed_msg_id; +static str routed_params = {0,0}; + /* * Test whether we are processing pre-loaded route set @@ -511,12 +517,16 @@ if (is_myself(&puri.host, puri.port_no)) #endif { - /* if (enable_double_rr && is_2rr(&_ruri->params)) { */ - /* DBG("ras(): Removing 2nd URI of mine: '%.*s'\n", rt->nameaddr.uri.len, ZSW(rt->nameaddr.uri.s)); */ + /*if (enable_double_rr && is_2rr(&_ruri->params)) { + * DBG("ras(): Removing 2nd URI of mine: '%.*s'\n", + * rt->nameaddr.uri.len, ZSW(rt->nameaddr.uri.s)); */ + /* set the hooks for the params -bogdan */ + routed_msg_id = _m->id; + routed_params = puri.params; if (!rt->next) { - /* No next route in the same header, remove the whole header - * field immediately - */ + /* No next route in the same header, remove the whole header + * field immediately + */ if (!del_lump(_m, hdr->name.s - _m->buf, hdr->len, 0)) { LOG(L_ERR, "after_strict: Cannot remove Route HF\n"); return RR_ERROR; @@ -661,11 +671,15 @@ if (is_myself(&puri.host, puri.port_no)) #endif { - DBG("after_loose: Topmost route URI: '%.*s' is me\n", uri->len, ZSW(uri->s)); + DBG("after_loose: Topmost route URI: '%.*s' is me\n", + uri->len, ZSW(uri->s)); + /* set the hooks for the params -bogdan */ + routed_msg_id = _m->id; + routed_params = puri.params; if (!rt->next) { - /* No next route in the same header, remove the whole header - * field immediately - */ + /* No next route in the same header, remove the whole header + * field immediately + */ if (!del_lump(_m, hdr->name.s - _m->buf, hdr->len, 0)) { LOG(L_ERR, "after_loose: Can't remove Route HF\n"); return RR_ERROR; @@ -684,16 +698,16 @@ if (enable_double_rr && is_2rr(&puri.params)) { if (!rt->next) { - /* No next route in the same header, remove the whole header - * field immediately - */ + /* No next route in the same header, remove the whole header + * field immediately + */ if (!del_lump(_m, hdr->name.s - _m->buf, hdr->len, 0)) { LOG(L_ERR, "after_loose: Can't remove Route HF\n"); return RR_ERROR; } res = find_next_route(_m, &hdr); if (res < 0) { - LOG(L_ERR, "after_loose: Error while finding next route\n"); + LOG(L_ERR,"after_loose: Error while finding next route\n"); return RR_ERROR; } if (res > 0) { /* No next route found */ @@ -783,3 +797,32 @@ } } } + + + +int check_route_param(struct sip_msg * msg, char *re, char *foo) +{ + regmatch_t pmatch; + char bk; + + /* check if the hooked params belong to the same message */ + if (routed_msg_id != msg->id) + return -1; + + /* check if params are present */ + if ( !routed_params.s || !routed_params.len ) + return -1; + + /* do the well-known trick to convert to null terminted */ + bk = routed_params.s[routed_params.len]; + routed_params.s[routed_params.len] = 0; + DBG("DEBUG:rr:check_route_param: params are <%s>\n", routed_params.s); + if (regexec( (regex_t*)re, routed_params.s, 1, &pmatch, 0)!=0) { + routed_params.s[routed_params.len] = bk; + return -1; + } else { + routed_params.s[routed_params.len] = bk; + return 1; + } +} + Index: modules/rr/loose.h =================================================================== RCS file: /cvsroot/ser/sip_router/modules/rr/loose.h,v retrieving revision 1.3 diff -u -r1.3 loose.h --- modules/rr/loose.h 24 Aug 2004 09:00:38 -0000 1.3 +++ modules/rr/loose.h 15 Apr 2005 14:52:56 -0000 @@ -40,4 +40,6 @@ int loose_route(struct sip_msg* _m, char* _s1, char* _s2); +int check_route_param(struct sip_msg * msg, char *re, char *foo); + #endif /* LOOSE_H */ Index: modules/rr/record.c =================================================================== RCS file: /cvsroot/ser/sip_router/modules/rr/record.c,v retrieving revision 1.14 diff -u -r1.14 record.c --- modules/rr/record.c 24 Aug 2004 09:00:38 -0000 1.14 +++ modules/rr/record.c 15 Apr 2005 14:52:56 -0000 @@ -29,6 +29,7 @@ * History: * ------- * 2003-04-04 Extracted from common.[ch] (janakj) + * 2005-04-10 add_rr_param() function and all corresponing hooks added (bogdan) */ #include <string.h> @@ -63,6 +64,22 @@ #define INBOUND 1 /* Insert inbound Record-Route */ #define OUTBOUND 0 /* Insert outbound Record-Route */ +#define RR_PARAM_BUF_SIZE 512 + +static unsigned int last_rr_msg; + +/* RR suffix lump and offset - used for inserting RR params + * after RR was done; use a two values array for INBOUND and + * OUTBOUND RR -bogdan */ +static struct lump *rr_suffix_lump[2] = {0,0}; +static int rr_suffix_end_offset[2] = {0,0}; + +/* RR param buffer - usd for storing RR param which are + * added before RR insertion -bogdan */ +static char rr_param_buf_ptr[RR_PARAM_BUF_SIZE]; +static str rr_param_buf = {rr_param_buf_ptr,0}; +static unsigned int rr_param_msg; + /* * Extract username from the Request URI @@ -104,16 +121,20 @@ { char* prefix, *suffix, *crlf, *r2; int suffix_len, prefix_len; + char *p; prefix_len = RR_PREFIX_LEN + (user->len ? (user->len + 1) : 0); prefix = pkg_malloc(prefix_len); if (enable_full_lr) { suffix_len = (_lr ? RR_LR_FULL_TERM_LEN : RR_SR_TERM_LEN) + - ((tag && tag->len) ? (RR_FROMTAG_LEN + tag->len) : 0); + ((tag && tag->len) ? (RR_FROMTAG_LEN + tag->len) : 0) + + rr_param_buf.len; } else { suffix_len = (_lr ? RR_LR_TERM_LEN : RR_SR_TERM_LEN) + - ((tag && tag->len) ? (RR_FROMTAG_LEN + tag->len) : 0); + ((tag && tag->len) ? (RR_FROMTAG_LEN + tag->len) : 0) + + rr_param_buf.len; } + rr_suffix_end_offset[_inbound] = RR_SR_TERM_LEN; suffix = pkg_malloc(suffix_len); crlf = pkg_malloc(2); @@ -145,21 +166,24 @@ #endif prefix[RR_PREFIX_LEN + user->len] = '@'; } - + + p = suffix; if (tag && tag->len) { - memcpy(suffix, RR_FROMTAG, RR_FROMTAG_LEN); - memcpy(suffix + RR_FROMTAG_LEN, tag->s, tag->len); - if (enable_full_lr) { - memcpy(suffix + RR_FROMTAG_LEN + tag->len, _lr ? RR_LR_FULL_TERM : RR_SR_TERM, _lr ? RR_LR_FULL_TERM_LEN : RR_SR_TERM_LEN); - } else { - memcpy(suffix + RR_FROMTAG_LEN + tag->len, _lr ? RR_LR_TERM : RR_SR_TERM, _lr ? RR_LR_TERM_LEN : RR_SR_TERM_LEN); - } + memcpy(p, RR_FROMTAG, RR_FROMTAG_LEN); + p += RR_FROMTAG_LEN; + memcpy(p, tag->s, tag->len); + p += tag->len; + } + if (rr_param_buf.len) { + memcpy( p, rr_param_buf.s, rr_param_buf.len); + p += rr_param_buf.len; + } + if (enable_full_lr) { + memcpy( p, _lr ? RR_LR_FULL_TERM : RR_SR_TERM, + _lr ? RR_LR_FULL_TERM_LEN : RR_SR_TERM_LEN); } else { - if (enable_full_lr) { - memcpy(suffix, _lr ? RR_LR_FULL_TERM : RR_SR_TERM, _lr ? RR_LR_FULL_TERM_LEN : RR_SR_TERM_LEN); - } else { - memcpy(suffix, _lr ? RR_LR_TERM : RR_SR_TERM, _lr ? RR_LR_TERM_LEN : RR_SR_TERM_LEN); - } + memcpy( p, _lr ? RR_LR_TERM : RR_SR_TERM, + _lr ? RR_LR_TERM_LEN : RR_SR_TERM_LEN); } memcpy(crlf, CRLF, 2); @@ -177,6 +201,7 @@ r2 = 0; } if (!(_l2 = insert_new_lump_before(_l2, suffix, suffix_len, 0))) goto lump_err; + rr_suffix_lump[_inbound] = _l2; suffix = 0; if (!(_l2 = insert_new_lump_before(_l2, crlf, 2, 0))) goto lump_err; crlf = 0; @@ -225,6 +250,12 @@ tag = 0; } + if (rr_param_buf.len && rr_param_msg!=_m->id) { + /* rr_params were set for a different message -> reset buffer */ + rr_param_buf.len = 0; + DBG("----> reset rr_param_buf\n"); + } + if (enable_double_rr) { l = anchor_lump(_m, _m->headers->name.s - _m->buf, 0, 0); l2 = anchor_lump(_m, _m->headers->name.s - _m->buf, 0, 0); @@ -256,6 +287,8 @@ return -4; } + /* reset the rr_param buffer */ + rr_param_buf.len = 0; return 0; } @@ -266,8 +299,6 @@ */ static inline int do_RR(struct sip_msg* _m, int _lr) { - static unsigned int last_rr_msg; - if (_m->id == last_rr_msg) { LOG(L_ERR, "record_route(): Double attempt to record-route\n"); return -1; @@ -395,3 +426,72 @@ { return do_RR(_m, 0); } + + +/* + * Appends a string to an existent lump + */ +static inline int append_to_lump( struct lump *l, str *s, int end_offset) +{ + char *p; + + /* update buffer */ + p = pkg_realloc( l->u.value, l->len+s->len); + if (p==0) { + LOG(L_ERR,"ERROR:rr:add_rr_param: no more pkg memory\n"); + return -1; + } + memmove(p+l->len-end_offset+s->len, p+l->len-end_offset, end_offset); + memcpy( p+l->len-end_offset, s->s, s->len); + /* update lump structure */ + l->len += s->len; + l->u.value = p; + return 0; +} + + +/* + * Appends a new Record-Route parameter + */ +int add_rr_param(struct sip_msg* msg, char* param, char* foo) +{ + str *rr_param; + int i; + + rr_param = (str*)param; + + if (last_rr_msg==msg->id) { + /* RR was already done -> have to modify the RR-sufix lump */ + for( i=OUTBOUND ; i<=INBOUND ; i++ ) { + if ( !rr_suffix_lump[i] || !rr_suffix_end_offset[i] ) + continue; + if (append_to_lump( rr_suffix_lump[i], rr_param, + rr_suffix_end_offset[i])!=0) { + LOG(L_ERR,"ERROR:rr:add_rr_param: failed to update lump\n"); + goto error; + } + } + } else { + /* RR not done yet -> store the param in the static buffer */ + if (rr_param_msg!=msg->id) { + /* it's about a different messge -> reset buffer */ + rr_param_buf.len = 0; + rr_param_msg = msg->id; + } + if (rr_param_buf.len+rr_param->len>RR_PARAM_BUF_SIZE) { + LOG(L_ERR,"ERROR:rr:add_rr_param: maximum size of " + "rr_param_buf exceeded\n"); + goto error; + } + memcpy( rr_param_buf.s+rr_param_buf.len, rr_param->s, rr_param->len); + rr_param_buf.len += rr_param->len; + DBG("DEBUG:rr:add_rr_param: rr_param_buf=<%.*s>\n",rr_param_buf.len, + rr_param_buf.s); + } + return 1; + +error: + return -1; +} + + Index: modules/rr/record.h =================================================================== RCS file: /cvsroot/ser/sip_router/modules/rr/record.h,v retrieving revision 1.2 diff -u -r1.2 record.h --- modules/rr/record.h 24 Aug 2004 09:00:38 -0000 1.2 +++ modules/rr/record.h 15 Apr 2005 14:52:56 -0000 @@ -56,4 +56,10 @@ int record_route_strict(struct sip_msg* _m, char* _s1, char* _s2); +/* + * Appends a new Record-Route parameter + */ +int add_rr_param(struct sip_msg* msg, char* param, char* foo); + + #endif /* RECORD_H */ Index: modules/rr/rr_mod.c =================================================================== RCS file: /cvsroot/ser/sip_router/modules/rr/rr_mod.c,v retrieving revision 1.33 diff -u -r1.33 rr_mod.c --- modules/rr/rr_mod.c 24 Aug 2004 09:00:38 -0000 1.33 +++ modules/rr/rr_mod.c 15 Apr 2005 14:52:57 -0000 @@ -33,11 +33,14 @@ * 2003-03-19 all mallocs/frees replaced w/ pkg_malloc/pkg_free (andrei) * 2003-04-01 Added record_route with ip address parameter (janakj) * 2003-04-14 enable_full_lr parameter introduced (janakj) + * 2005-04-10 add_rr_param() and check_route_param() added (bogdan) */ #include <stdio.h> #include <stdlib.h> +#include <regex.h> + #include "../../sr_module.h" #include "../../ut.h" #include "../../error.h" @@ -61,6 +64,7 @@ static int mod_init(void); static int str_fixup(void** param, int param_no); +static int regexp_fixup(void** param, int param_no); /* @@ -74,10 +78,18 @@ * Oh, BTW, have I mentioned already that you shouldn't use strict routing ? */ static cmd_export_t cmds[] = { - {"loose_route", loose_route, 0, 0, REQUEST_ROUTE}, - {"record_route", record_route, 0, 0, REQUEST_ROUTE}, - {"record_route_preset", record_route_preset, 1, str_fixup, REQUEST_ROUTE}, - {"record_route_strict" , record_route_strict, 0, 0, 0 }, + {"loose_route", loose_route, 0, 0, + REQUEST_ROUTE}, + {"record_route", record_route, 0, 0, + REQUEST_ROUTE}, + {"record_route_preset", record_route_preset, 1, str_fixup, + REQUEST_ROUTE}, + {"record_route_strict" , record_route_strict, 0, 0, + 0}, + {"add_rr_param", add_rr_param, 1, str_fixup, + REQUEST_ROUTE}, + {"check_route_param", check_route_param, 1, regexp_fixup, + REQUEST_ROUTE}, {0, 0, 0, 0, 0} }; @@ -149,3 +161,29 @@ return 0; } + + +static int regexp_fixup(void** param, int param_no) +{ + regex_t* re; + + if (param_no==1) { + if ((re=pkg_malloc(sizeof(regex_t)))==0) { + LOG(L_ERR,"ERROR:rr:regexp_fixup: no more pkg memory\n"); + return E_OUT_OF_MEM; + } + if (regcomp(re, *param, REG_EXTENDED|REG_ICASE|REG_NEWLINE) ) { + pkg_free(re); + LOG(L_ERR, "ERROR:rr:regexp_fixup: bad regexp %s\n",(char*)*param); + return E_BAD_RE; + } + /* free string */ + pkg_free(*param); + /* replace it with the compiled re */ + *param=re; + } + return 0; +} + + +

19 years, 6 months

Re: [Serusers] SER Message waiting indicator light Cisco 7940/7960

by Java Rockx

Patrick, This is a bash script that externnotify executes #!/bin/ash # # This script is called by Asterisk every time a voice mail is # left for a subscriber. The CONTEXT, EXTENSION, and VM_COUNT # are passed by Asterisk. # # All we need to do is update a file in /var/spool/asterisk/mwi # to tell our cron job that we have to send a SIP NOTIFY message # to a particular UA # # This file is referenced in /etc/asterisk/voicemail.conf # # NOTE: The $2 argument is passed in to this script in the # following form: MAILBOX@DOMAIN # # example: 4075551212@default # MWI_DIR=/var/spool/asterisk/mwi/ if [ -n ${2} ] ; then if /bin/touch ${MWI_DIR}/${2} ; then exit 0 else exit 1 fi else exit 2 fi The following is executed in CRON every minute. This script actually notifies the subscribers by sending NOTIFY messages to SER. Please note that our asterisk is a modified version that stores mailboxes in a directory hash, so you can modify this to fit your needs. #!/bin/sh # This script send the SIP NOTIFY message to UA. The NOTIFY message # can either enable or disable the UA message indicator. The script # looks for any file in $VM_HOME and creates the NOTIFY message # # The actual NOTIFY message is sent to the SIP proxy by the sipsak # utilty. After the message is send we delete the processed file # from VM_HOME so we don't keep resending the same message. Most # UAs will periodically send a SUBSCRIBE message to the SIP proxy. # When this occurs the SIP proxy will place a file in the location # /var/spool/mwi/delay and a less frequent CRON job will process it. VM_ROOT=/var/spool/asterisk/voicemail VM_HOME=/var/spool/asterisk/mwi SIP_SERVER=71.110.20.33 SIP_PORT=5060 SIP_FROM=vmserver SIP_PASSWORD=heslo DB_NAME=ser DB_SERVER=10.255.237.18 DB_OPTIONS="--skip-column-names --connect_timeout=3" DB_USER=ser DB_PASSWORD=heslo # Allow glob expansions to return NULL shopt -s nullglob # Sanity check that VM_HOME exists, warn if it does not if [ ! -d $VM_HOME ] ; then echo "Cannot chdir to VM_HOME, exiting" exit 1 fi for file in $VM_HOME/* do file=$( basename $file @default ) # The file existance check may seem redundant, but it is theoretically possible # for this script to take longer than 1 cron cycle to run, or for it to be run # by hand while another copy is running, so we really do need this check. if [ -e ${VM_HOME}/${file}@default ] && [ ${#file} -eq 10 ]; then QUERY="SELECT subscriber.domain FROM subscriber, location WHERE subscriber.username='$file' AND subscriber.username = location.username LIMIT 1" DOMAIN=`mysql ${DB_OPTIONS} -h ${DB_SERVER} -u${DB_USER} -p${DB_PASSWORD} -e "$QUERY" ${DB_NAME}` # If the MySQL call fails, then $DOMAIN will hold the error message instead if [ $? -ne 0 ] ; then echo "Error connecting to MySQL: $DOMAIN" exit 2 fi if [ -n "${DOMAIN}" ]; then MAILBOX=$VM_ROOT/default/${file:0:3}/${file:3:3}/${file:6:4} # Sanity check that INBOX exists, warn if it does not if [ ! -d ${MAILBOX} ]; then echo "${MAILBOX} does not exist" continue fi # Delete all messages with a zero length. If the subscriber still # has voice mail files in their INBOX after this step then notify # them even if the total is zero, which will extinguish their MWI for message in ${MAILBOX}/INBOX/*.gsm do if [ ! -s $message ]; then GSM=`basename $message .gsm` rm -f ${MAILBOX}/INBOX/$GSM.* fi done TOTAL_MESSAGES=0 NEW_MESSAGES=0 OLD_MESSAGES=0 if [ -d ${MAILBOX}/INBOX ]; then tmp=( $( echo ${MAILBOX}/INBOX/*.txt ) ) NEW_MESSAGES=${#tmp[*]} else NEW_MESSAGES=0 fi if [ -d ${MAILBOX}/Old ]; then tmp=( $( echo ${MAILBOX}/Old/*.txt ) ) OLD_MESSAGES=${#tmp[*]} else OLD_MESSAGES=0 fi TOTAL_MESSAGES=$(($NEW_MESSAGES + $OLD_MESSAGES)) if [ "$NEW_MESSAGES" == "0" ]; then HAS_NEW="no" else HAS_NEW="yes" fi CONTENT_LENGTH=$(( 34 + ${#HAS_NEW} + ${#NEW_MESSAGES} + ${#OLD_MESSAGES} )) SEQUENCE=$( printf "%06d" $RANDOM ) ( cat <<-EOM NOTIFY sip:${file}@${DOMAIN} SIP/2.0 From: <sip:${SIP_FROM}@${DOMAIN}> To: <sip:${file}@${DOMAIN}> Contact: <sip:${file}@${SIP_SERVER}> Call-ID: ${SEQUENCE}@${SIP_SERVER} CSeq: ${SEQUENCE} NOTIFY Event: message-summary Content-Type: application/simple-message-summary Content-Length: ${CONTENT_LENGTH} Messages-Waiting: ${HAS_NEW} Voicemail: ${NEW_MESSAGES}/${OLD_MESSAGES} EOM ) | /usr/bin/sipsak shoot -a ${SIP_PASSWORD} -s sip:${file}@${DOMAIN} -f - if [ $? -ne 0 ] ; then echo "sipsak error for $file" fi else echo "DOMAIN is null for $file" fi else if [ ${#file} -ne 10 ] ; then echo "Extension was not 10 digits: '$file' (${#file})" else echo "Problem with file: ${VM_HOME}/${file}@default" fi fi rm ${VM_HOME}/${file}@default done exit 0 On Tue, 1 Mar 2005 10:06:56 -0500, Patrick Baker <patricksbaker(a)gmail.com> wrote: > could you show me an example of the sipsak command for the cisco phone > - I can then just make an agi that will call it after vm is left. > > Best regards, > > Patrick B > > > On Tue, 1 Mar 2005 10:03:35 -0500, Java Rockx <javarockx(a)gmail.com> wrote: > > Since the SIP UA is registered with SER you'll need to have Asterisk > > send a NOTIFY message to the user by means of SER. > > > > So you can send the NOTIFY to SER with a <From:> header as asterisk > > and the <To> / R-URI as the Cisco UA. > > > > FYI, we do this using sipsak to send the NOTIFY to ser. This is > > triggered either using the externnotify parameter in the Asterisk > > voicemail.conf file and by CRON for periodic MWI updates. > > > > Regards, > > Paul > > > > > > On Tue, 1 Mar 2005 09:53:43 -0500, Patrick Baker > > <patricksbaker(a)gmail.com> wrote: > > > Hello, > > > > > > I'm sure this may have been asked but, I just wanted clarify how this > > > would be accomplished. I'm looking to have the MWI light on my cisco > > > phone turn on when a message is left in VM. > > > > > > Here is my current situation: > > > > > > sip user -> ser -> asterisk > > > > > > Best regards, > > > > > > Patrick b. > > > > > > _______________________________________________ > > > Serusers mailing list > > > serusers(a)lists.iptel.org > > > http://lists.iptel.org/mailman/listinfo/serusers > > > > > >

19 years, 7 months

[Serusers] UAC module BUG??

by Helge Waastad

Hi, from my debug after trying to use the uac_auth in a failure route, I get the following: Mar 31 16:37:46 sip ser[20394]: BUG:uac:uac_auth: empty reply on picked branch I've modified the cvs version to compile on 0.9.0 Anyone? Mvh, Helge Waastad

19 years, 7 months

[Serusers] SER Proxy with Redirect

by Eric Dean

I have a SER Proxy that is routing calls to a SIP Redirector which then informs SER as to which gateway to send call to. GW1 GW2 | | SERProxy------->SIP Redirector | IP Phone I would like for SER to take a 302 redirect and then proxy the call to the gateway per the redirect..however, SER is currently serving the redirect to the IP Phone..which is unauthorized to access the gateway. How do I configure SER to actually proxy the call per the redirect on behalf of the IP Phone?

19 years, 7 months

[Serusers] RTPProxy Remote mode - startup command?

by Alex Mack

Hi! I want to use rtpproxy in remote mode, i.e. rtpproxy runs on another machine as ser. I've tried to start rtpproxy accordingly to the README.remote with: ./rtpproxy -s udp:213.10.10.10 But I only get this error response: rtpproxy: explicit binding address has to be specified in UDP command mode What am I doing wrong? Alex Mack

19 years, 7 months

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

sr-users April 2005