tm Module

Bogdan-Andrei Iancu

Jiri Kuthan


Edited by

Jiri Kuthan

Edited by

Bogdan-Andrei Iancu

Edited by

Ovidiu Sas

Revision History
Revision $Revision: 8740 $$Date$

Table of Contents

1. Admin Guide
1.1. Overview
1.1.1. Per-Branch flags
1.1.2. Timeout-Based Failover
1.1.3. DNS Failover
1.1.4. Anycast Scenario
1.1.5. Usage Scope
1.2. Dependencies
1.2.1. OpenSIPS Modules
1.2.2. External Libraries or Applications
1.3. Exported Parameters
1.3.1. fr_timeout (integer)
1.3.2. fr_inv_timeout (integer)
1.3.3. wt_timer (integer)
1.3.4. delete_timer (integer)
1.3.5. T1_timer (integer)
1.3.6. T2_timer (integer)
1.3.7. ruri_matching (integer)
1.3.8. via1_matching (integer)
1.3.9. unix_tx_timeout (integer)
1.3.10. restart_fr_on_each_reply (integer)
1.3.11. tw_append (string)
1.3.12. pass_provisional_replies (integer)
1.3.13. syn_branch (integer)
1.3.14. onreply_avp_mode (integer)
1.3.15. disable_6xx_block (integer)
1.3.16. enable_stats (integer)
1.3.17. minor_branch_flag (string/integer)
1.3.18. timer_partitions (integer)
1.3.19. auto_100trying (integer)
1.3.20. tm_replication_cluster (integer)
1.3.21. cluster_param (string)
1.3.22. cluster_auto_cancel (boolean)
1.4. Exported Functions
1.4.1. t_relay([flags])
1.4.2. t_relay(proto:host:port,[flags])
1.4.3. t_reply(code, reason_phrase)
1.4.4. t_reply_with_body(code, reason_phrase, body)
1.4.5. t_newtran()
1.4.6. t_check_trans()
1.4.7. t_check_status(re)
1.4.8. t_local_replied(reply)
1.4.9. t_was_cancelled()
1.4.10. t_cancel_branch([flags])
1.4.11. t_new_request(method,RURI,from,to[,body[,ctx]])
1.4.12. t_on_failure(failure_route)
1.4.13. t_on_reply(reply_route)
1.4.14. t_on_branch(branch_route)
1.4.15. t_inject_branches(source[,flags])
1.4.16. t_wait_for_new_branches()
1.4.17. t_add_hdrs("sip_hdrs")
1.4.18. t_add_cancel_reason("Reason_hdr")
1.4.19. t_replicate(URI,[flags])
1.4.20. t_write_req(info,fifo) t_write_unix(info,sock)
1.4.21. t_flush_flags()
1.4.22. t_anycast_replicate()
1.5. Exported variables
1.5.1. $T_branch_idx
1.5.2. $T_reply_code
1.5.3. $T_fr_timeout
1.5.4. $T_fr_inv_timeout
1.5.5. $T_ruri
1.5.6. $bavp(name)
1.5.7. $T_id
1.6. Exported MI Functions
1.6.1. t_uac_dlg
1.6.2. t_uac_cancel
1.6.3. t_hash
1.6.4. t_reply
1.7. Exported statistics
1.7.1. received_replies
1.7.2. relayed_replies
1.7.3. local_replies
1.7.4. UAS_transactions
1.7.5. UAC_transactions
1.7.6. 2xx_transactions
1.7.7. 3xx_transactions
1.7.8. 4xx_transactions
1.7.9. 5xx_transactions
1.7.10. 6xx_transactions
1.7.11. inuse_transactions
2. Developer Guide
2.1. Functions
2.1.1. load_tm(*import_structure)
3. Frequently Asked Questions

List of Examples

1.1. Set fr_timeout parameter
1.2. Set fr_inv_timeout parameter
1.3. Set wt_timer parameter
1.4. Set delete_timer parameter
1.5. Set T1_timer parameter
1.6. Set T2_timer parameter
1.7. Set ruri_matching parameter
1.8. Set via1_matching parameter
1.9. Set unix_tx_timeout parameter
1.10. Set restart_fr_on_each_reply parameter
1.11. Set tw_append parameter
1.12. Set pass_provisional_replies parameter
1.13. Set syn_branch parameter
1.14. Set onreply_avp_mode parameter
1.15. Set disable_6xx_block parameter
1.16. Set enable_stats parameter
1.17. Set minor_branch_flag parameter
1.18. Set timer_partitions parameter
1.19. Set auto_100trying parameter
1.20. Set tm_replication_cluster parameter
1.21. Set the cluster_param parameter
1.22. Set the cluster_auto_cancel parameter
1.23. t_relay usage
1.24. t_relay usage
1.25. t_reply usage
1.26. t_reply_with_body usage
1.27. t_newtran usage
1.28. t_check_trans usage
1.29. t_check_status usage
1.30. t_local_replied usage
1.31. t_was_cancelled usage
1.32. t_cancel_branch usage
1.33. t_new_request usage
1.34. t_on_failure usage
1.35. t_on_reply usage
1.36. t_on_branch usage
1.37. t_inject_branches usage
1.38. t_wait_for_new_branches usage
1.39. t_add_hdrs usage
1.40. t_add_cancel_reason usage
1.41. t_replicate usage
1.42. t_write_req/unix usage
1.43. t_flush_flags usage
1.44. t_anycast_replicate usage

Chapter 1. Admin Guide

1.1. Overview

TM module enables stateful processing of SIP transactions. The main use of stateful logic, which is costly in terms of memory and CPU, is some services inherently need state. For example, transaction-based accounting (module acc) needs to process transaction state as opposed to individual messages, and any kinds of forking must be implemented statefully. Other use of stateful processing is it trading CPU caused by retransmission processing for memory. That makes however only sense if CPU consumption per request is huge. For example, if you want to avoid costly DNS resolution for every retransmission of a request to an unresolvable destination, use stateful mode. Then, only the initial message burdens server by DNS queries, subsequent retransmissions will be dropped and will not result in more processes blocked by DNS resolution. The price is more memory consumption and higher processing latency.

From user's perspective, the major function is t_relay(). It setup transaction state, absorb retransmissions from upstream, generate downstream retransmissions and correlate replies to requests.

In general, if TM is used, it copies clones of received SIP messages in shared memory. That costs the memory and also CPU time (memcpys, lookups, shmem locks, etc.) Note that non-TM functions operate over the received message in private memory, that means that any core operations will have no effect on statefully processed messages after creating the transactional state. For example, calling record_route after t_relay is pretty useless, as the RR is added to privately held message whereas its TM clone is being forwarded.

TM is quite big and uneasy to program--lot of mutexes, shared memory access, malloc and free, timers--you really need to be careful when you do anything. To simplify TM programming, there is the instrument of callbacks. The callback mechanisms allow programmers to register their functions to specific event. See t_hooks.h for a list of possible events.

Other things programmers may want to know is UAC--it is a very simplistic code which allows you to generate your own transactions. Particularly useful for things like NOTIFYs or IM gateways. The UAC takes care of all the transaction machinery: retransmissions , FR timeouts, forking, etc. See t_uac prototype in uac.h for more details. Who wants to see the transaction result may register for a callback.

1.1.1. Per-Branch flags

First what is the idea with the branch concept: branch route is a route to be execute separately for each branch before being sent out - changes in that route should reflect only on that branch.

There are several types of flags in OpenSIPS :

  • message/transaction flags - they are visible everywhere in the transaction (in all routes and in all sequential replies/request).

  • branch flags - flags that are visible only from a specific branch - in all replies and routes connected to this branch.

  • script flags - flags that exist only during script execution. They are not store anywhere and are lost once the top level route was left.

For example: I have a call parallel forking to GW and to a user. And I would like to know from which branch I will get the final negative reply (if so). I will set a branch route before relaying the calls (with the 2 branches). The branch route will be separately executed for each branch; in the branch going to GW (I can identified it by looking to RURI), I will set a branch flag. This flag will appear only in the onreply route run for replied from GW. It will be also be visible in failure route if the final elected reply belongs to the GW branch. This flags will not be visible in the other branch (in routes executing replies from the other branch).

For how to define branch flags and use via script, see Section 1.4.14, “ t_on_branch(branch_route) and the setbflag(), resetbflag() and isbflagset() script functions.

Also, modules may set branch flags before transaction creation (for the moment this feature is not available in script). The REGISTRAR module was the first to use this type of flags. The NAT flag is pushed in branch flags instead in message flags

1.1.2. Timeout-Based Failover

Timeouts can be used to trigger failover behavior. E.g. if we send a call to a gateway and the gateway does not send a provisional response within 3 seconds, we want to cancel this call and send the call to another gateway. Another example is to ring a SIP client only for 30 seconds and then redirect the call to the voicemail.

The transaction module exports two types of timeouts:

  • fr_timeout - used when no response was received yet. If there is no response after fr_timeout seconds, the timer triggers (and failure route will be executed if t_on_failure() was called). For INVITE transactions, if a provisional response was received, the timeout is reset to fr_inv_timeout seconds and RT_T2 for all other transactions. Once a final response is received, the transaction has finished.

  • fr_inv_timeout - this timeout starts counting down once a provisional response was received for an INVITE transaction.

For example: You want to have failover if there is no provisional response after 3 seconds, but you want to ring for 60 seconds. Thus, set the fr_timeout to 3 and fr_inv_timeout to 60.

1.1.3. DNS Failover

DNS based failover can be use when relaying stateful requests. According to RFC 3263, DNS failover should be done on transport level or transaction level. TM module supports them both.

Failover at transport level may be triggered by a failure of sending out the request message. A failure occurs if the corresponding interface was found for sending the request, if the TCP connection was refused or if a generic internal error happened during send. There is no ICMP error report support.

Failover at transaction level may be triggered when the transaction completed either with a 503 reply, either with a timeout without any received reply. In such a case, automatically, a new branch will be forked if any other destination IPs can be used to deliver the requests. The new branch will be a clone of the winning branch.

The set of destinations IPs is step-by-step build (on demand) based on the NAPTR, SRV and A records available for the destination domain.

DNS-based failover is by default applied excepting when this failover is globally disabled (see the core parameter disable_dns_failover) or when the relay flag (per transaction) is set (see the t_relay() function).

1.1.4. Anycast Scenario

Doing a load balancing scenario using Anycast IPs, one might run into an issue where a transaction request comes on one instance, and the reply (or replies) comes on different ones. This would normaly break the transaction state, because the local transaction will start re-transmissios and would eventually timeout. Moreover, from UA's perspective, the reply whould have been sent, but since it reaches a proxy that is not aware of that transaction, it will not be forwarded (nor ACKed in case of INVITES). And from this point things can escalade quickly.

To sort out these problems, the module uses a distributed mechanism to figure out where the transaction for a specific reply was created. When an instance receives a reply that does not have an associated transaction, it replicates it to be handled by the instance that owns it. This is achieved using the clusterer module support.

Setting up an anycast scenario is very simple: all the instances that are part of an anycast secnario must be set up in a cluster (more info at the tm_replication_cluster param). When a transaction is created, a special identifier is appended to the branch parameter, namely the instance that created the transaction. When a reply comes in, the transaction module checks who owns the transaction. If the identifier is the instance's own id, then the reply is processed locally. Otherwise it is replicated to the node indicated by the id. Replication is done in a very efficient manner, using the proto_bin transport.

Special handling is applied to CANCEL and ACK methods. Due to the fact that these methods do not contain the special identifier in the branch parameter (since they are generated by the UAC and not by us), there is no way to determine who owns the transaction. Therefore, if we do not find a local transaction for these requests, we broadcast them to all the other instances using the t_anycast_replicate function. Again, this is done in a very efficient manner using the proto_bin transport.

1.1.5. Usage Scope

Transaction functions and variables are only designed to be called on SIP request messages where a transaction can be created, or in routes that are transaction aware, such as branch_route[name], failure_route[name] or onreply_route[name]. Using TM functtions or variables in a route that is not transaction aware, such as the generic onreply_route, error_route or timer_route[name, timer] may lead to undefined behavior, and most of the time in bogus or malformed signalling. Therefore it is strongly recommended to avoid using them in non-tm context aware routes.

1.2. Dependencies

1.2.1. OpenSIPS Modules

The following modules must be loaded before this module:

  • clusterer module, if the anycast scenario is enabled (see tm_replication_cluster param for more information).

1.2.2. External Libraries or Applications

The following libraries or applications must be installed before running OpenSIPS with this module loaded:

  • None.

1.3. Exported Parameters

1.3.1. fr_timeout (integer)

Timeout which is triggered if no final reply for a request or ACK for a negative INVITE reply arrives (in seconds).

Default value is 30 seconds.

Example 1.1. Set fr_timeout parameter

modparam("tm", "fr_timeout", 10)

1.3.2. fr_inv_timeout (integer)

Timeout which is triggered if no final reply for an INVITE arrives after a provisional message was received (in seconds). This timeout starts counting down once the first provisional response is received. Thus, fast failover (no 100 trying from gateway) can be achieved by setting fr_timeout to low values. See example below.

Default value is 120 seconds.

Example 1.2. Set fr_inv_timeout parameter

modparam("tm", "fr_inv_timeout", 200)

1.3.3. wt_timer (integer)

Time for which a transaction stays in memory to absorb delayed messages after it completed; also, when this timer hits, retransmission of local cancels is stopped (a puristic but complex behavior would be not to enter wait state until local branches are finished by a final reply or FR timer--we simplified).

For non-INVITE transaction this timer relates to timer J of RFC 3261 section 17.2.2. According to the RFC this timer should be 64*T1 (= 32 seconds). But this would increase memory usage as the transactions are kept in memory very long.

Default value is 5 seconds.

Example 1.3. Set wt_timer parameter

modparam("tm", "wt_timer", 10)

1.3.4. delete_timer (integer)

Time after which a to-be-deleted transaction currently ref-ed by a process will be tried to be deleted again.

Default value is 2 seconds.

Example 1.4. Set delete_timer parameter

modparam("tm", "delete_timer", 5)

1.3.5. T1_timer (integer)

Retransmission T1 period, in milliseconds.

Default value is 500 milliseconds.

Example 1.5. Set T1_timer parameter

modparam("tm", "T1_timer", 700)

1.3.6. T2_timer (integer)

Maximum retransmission period, in milliseconds.

Default value is 4000 milliseconds.

Example 1.6. Set T2_timer parameter

modparam("tm", "T2_timer", 8000)

1.3.7. ruri_matching (integer)

Should be request-uri matching used as a part of pre-3261 transaction matching as the standard wants us to do so? Turn only off for better interaction with devices that are broken and send different r-uri in CANCEL/ACK than in original INVITE.

Default value is 1 (true).

Example 1.7. Set ruri_matching parameter

modparam("tm", "ruri_matching", 0)

1.3.8. via1_matching (integer)

Should be top most VIA matching used as a part of pre-3261 transaction matching as the standard wants us to do so? Turn only off for better interaction with devices that are broken and send different top most VIA in CANCEL/ACK than in original INVITE.

Default value is 1 (true).

Example 1.8. Set via1_matching parameter

modparam("tm", "via1_matching", 0)

1.3.9. unix_tx_timeout (integer)

Send timeout to be used by function which use UNIX sockets (as t_write_unix).

Default value is 2 seconds.

Example 1.9. Set unix_tx_timeout parameter

modparam("tm", "unix_tx_timeout", 5)

1.3.10. restart_fr_on_each_reply (integer)

If true (non null value), the final response timer will be re-triggered for each received provisional reply. In this case, final response timeout may occur after a time longer than fr_inv_timeout (if UAS keeps sending provisional replies)

Default value is 1 (true).

Example 1.10. Set restart_fr_on_each_reply parameter

modparam("tm", "restart_fr_on_each_reply", 0)

1.3.11. tw_append (string)

List of additional information to be appended by t_write_req and t_write_unix functions.

Default value is null string.

Syntax of the parameter is:

  • tw_append = append_name':' element (';'element)*

  • element = ( [name '='] variable)

Each element will be appended per line in name: value format. Element $rb (message body) is the only one which does not accept name; the body it will be printed all the time at the end, disregarding its position in the definition string.

Example 1.11. Set tw_append parameter

modparam("tm", "tw_append",
   "test: ua=$hdr(User-Agent) ;avp=$avp(avp);$rb;time=$Ts")

1.3.12. pass_provisional_replies (integer)

Enable/disable passing of provisional replies to FIFO applications.

Default value is 0.

Example 1.12. Set pass_provisional_replies parameter

modparam("tm", "pass_provisional_replies", 1)

1.3.13. syn_branch (integer)

Enable/disable the usage of stateful synonym branch IDs in the generated Via headers. They are faster but not reboot-safe.

Default value is 1 (use synonym branches).

Example 1.13. Set syn_branch parameter

modparam("tm", "syn_branch", 0)

1.3.14. onreply_avp_mode (integer)

Describes how the AVPs should be handled in reply route:

  • 0 - the AVPs will be per message only; they will not interfere with the AVPS stored in transaction; initially there will be an empty list and at the end of the route, all AVPs that were created will be discarded.

  • 1 - the AVPs will be the transaction AVPs; initially the transaction AVPs will be visible; at the end of the route, the list will attached back to transaction (with all the changes)

In mode 1, you can see the AVPs you set in request route, branch route or failure route. The side efect is performance as more locking is required in order to keep the AVP's list integrity.

Default value is 0.

Example 1.14. Set onreply_avp_mode parameter

modparam("tm", "onreply_avp_mode", 1)

1.3.15. disable_6xx_block (integer)

Tells how the 6xx replies should be internally handled:

  • 0 - the 6xx replies will block any further serial forking (adding new branches). This is the RFC3261 behaviour.

  • 1 - the 6xx replies will be handled as any other negative reply - serial forking will be allowed. Logically, you need to break RFC3261 if you want to do redirects to announcement and voicemail services.

Default value is 0.

Example 1.15. Set disable_6xx_block parameter

modparam("tm", "disable_6xx_block", 1)

1.3.16. enable_stats (integer)

Enables statistics support in TM module - If enabled, the TM module will internally keep several statistics and export them via the MI - Management Interface.

Default value is 1 (enabled).

Example 1.16. Set enable_stats parameter

modparam("tm", "enable_stats", 0)

1.3.17. minor_branch_flag (string/integer)

A branch flag index to be used in script to mark the minor branches ( before t_relay() ).

A minor branch is a branch OpenSIPS will not wait to complete during parallel forking. So, if the rest of the branches are negativly replied OpenSIPS will not wait for a final answer from the minor branch, but it will simply cancel it.

Main applicability of minor branch is to fork a branch to a media server for injecting (via 183 Early Media) some pre-call media - of course, this branch will be transparanent for the rest of the call branches (from branch selection point of view).

Default value is none (disabled).

Example 1.17. Set minor_branch_flag parameter

modparam("tm", "minor_branch_flag", "MINOR_BFLAG")

1.3.18. timer_partitions (integer)

The number of partitions for the internal TM timers (retransmissions, delete, wait, etc). Partitioning the timers increase the throughput under heavly load by handling timer events in parallel, rather than all serial.

Recomanded range for timer partitions is max 16 (soft limit).

Default value is 1 (disabled).

Example 1.18. Set timer_partitions parameter

# Enable two timer partitions
modparam("tm", "timer_partitions", 2)

1.3.19. auto_100trying (integer)

This parameter controls if the TM module should automatically generate an 100 Trying stateful reply when an INVITE transaction is created.

You may want to disable this behavior if you want to control from script level when the 100 Trying is to be sent out.

Default value is 1 (enabled).

Example 1.19. Set auto_100trying parameter

# Disable automatic 100 Trying
modparam("tm", "auto_100trying", 0)

1.3.20. tm_replication_cluster (integer)

This parameter should be used in an anycast setup, and specifies the cluster id of all the nodes that use an anycast IP.

Check out the tm_anycast section for more details.

Anycast replication is disabled by default.

Example 1.20. Set tm_replication_cluster parameter

# replicate anycast messages in cluster 1
modparam("tm", "tm_replication_cluster", 1)

1.3.21. cluster_param (string)

This parameter should be used in an anycast setup, and specifies the name of the parameter used in the VIA branch param to specifiy the instance id that created the transaction.

Check out the tm_anycast section for more details.

Default value is cid.

Example 1.21. Set the cluster_param parameter

modparam("tm", "cluster_param", "tid")

1.3.22. cluster_auto_cancel (boolean)

This parameter should be used in an anycast setup, and specifies whether a CANCEL message received on a listener that is marked as anycast should be automatically handled, or should get in the OpenSIPS script. If this parameter is enabled (default), CANCEL messages received on an anycast listener will never enter the script, thus making the script cleaner.

Check out the tm_anycast section for more details.

Default value is yes (enabled).

Example 1.22. Set the cluster_auto_cancel parameter

# disable auto-cancel handling
modparam("tm", "cluster_auto_cancel", no)

1.4. Exported Functions

1.4.1.  t_relay([flags])

Relay a message statefully to destination indicated in current URI. (If the original URI was rewritten by UsrLoc, RR, strip/prefix, etc., the new URI will be taken). Returns a negative value on failure--you may still want to send a negative reply upstream statelessly not to leave upstream UAC in lurch.

The coresponding transaction may or may not be already created. If not yet created, the function will automatically create it.

The function may take as parameter an optional set of flags for controlling the internal behaviour. The flags may be given in decimal or hexa format; supported flags are:

  • 0x01 - deprecated, not used any more

  • 0x02 - do not internally send a negative reply in case of forward failure (due internal error, bad RURI, bad message, etc). When a forward failure occurs, no SIP request is relayed and therefore no negative reply or timeout will show up on the failure_route (if one is set). It applies only when the transaction is created. By default one negative reply is sent. Useful if you want to implement a serial forking in case of failure.

  • 0x04 - disable the DNS failover for the transaction. Only first IP will be used. It disables the failover both at transport and transaction level.

  • 0x08 - If the request is a CANCEL, trust and pass further the Reason header from the received CANCEL - shortly, will propagate the Reason header.

In case of error, the function returns the following codes:

  • -1 - generic internal error

  • -2 - bad message (parsing errors)

  • -3 - no destination available (no branches were added or request already cancelled)

  • -4 - bad destination (unresolvable address)

  • -5 - destination filtered (black listed)

  • -6 - generic send failed

This function can be used from REQUEST_ROUTE, FAILURE_ROUTE.

Example 1.23. t_relay usage

if (!t_relay()) {

1.4.2.  t_relay(proto:host:port,[flags])

Relay a message statefully to a fixed destination. The destination is specified as [proto:]host[:port]. If a destination URI $du for this message was set before the function is called then this value will be used as the destination instead of the function parameter.

The function may take as parameter an optional set of flags for controlling the internal behaviour - for details see the above t_relay([flags]) function.

This functions can be used from REQUEST_ROUTE, FAILURE_ROUTE.

Example 1.24. t_relay usage


1.4.3.  t_reply(code, reason_phrase)

Sends a stateful SIP reply to the currently processed requests. Note that if the transaction was not created yet, it will automatically created by internally using the t_newtran function.

Meaning of the parameters is as follows:

  • code - Reply code number.

  • reason_phrase - Reason string.

Both parameters accept any kind of variables.

This function can be used from REQUEST_ROUTE, FAILURE_ROUTE.

Example 1.25. t_reply usage

t_reply("404", "Use $rU not found");

1.4.4.  t_reply_with_body(code, reason_phrase, body)

Sends a stateful SIP reply with a body to the currently processed requests. Note that if the transaction was not created yet, it will automatically created by internally using the t_newtran function.

Meaning of the parameters is as follows:

  • code - Reply code number.

  • reason_phrase - Reason string.

  • body - Reply body.

All parameters accept any kind of variables.

This function can be used from REQUEST_ROUTE, FAILURE_ROUTE.

Example 1.26. t_reply_with_body usage

		append_to_reply("Contact: $var(contact)\r\n"
				"Content-Type: application/sdp\r\n");
		t_reply_with_body("200", "Ok", "$var(body)");

1.4.5.  t_newtran()

Creates the SIP transaction for the currently processed SIP requests. Basically you switch to a stateful processing from this point furhter, with automatic detection of retransmission and auto sending of 100 Trying reply on INVITEs.

This function can be used from REQUEST_ROUTE.

Example 1.27. t_newtran usage

t_newtran();  # 100 Trying is fired here
xlog("doing my complicated routing logic");
t_relay(); # send the call further

1.4.6.  t_check_trans()

Returns true if the current request is associated to a transaction. The relationship between the request and transaction is defined as follows:

  • non-CANCEL/non-ACK requests - if the request belongs to a transaction (it's a retransmision), the function will do a standard processing of the retransmission and will break/stop the script. The function returns false if the request is not a retransmission.

  • CANCEL request - true if the cancelled INVITE transaction exists.

  • ACK request - true if the ACK is a hop-by-hop ACK (to a negative reply) corresponding to an previous INVITE transaction. IMPORTANT: end-to-end ACKs (to 2xx replies) are not detected by this function as end-to-end ACKs form a different transaction.

Note: To detect retransmissions using this function you have to make sure that the initial request has already created a transaction, e.g. by using t_relay(). If the processing of requests may take long time (e.g. DB lookups) and the retransmission arrives before t_relay() is called, you can use the t_newtran() function to manually create a transaction.

This function can be used from REQUEST_ROUTE and BRANCH_ROUTE.

Example 1.28. t_check_trans usage

if ( is_method("CANCEL") ) {
	if ( t_check_trans() )

1.4.7.  t_check_status(re)

Returns true if the regualr expression re match the reply code of the response message as follows:

  • in routing block - the code of the last sent reply.

  • in on_reply block - the code of the current received reply.

  • in on_failure block - the code of the selected negative final reply.


Example 1.29. t_check_status usage

if (t_check_status("(487)|(408)")) {
    log("487 or 408 negative reply\n");

1.4.8.  t_local_replied(reply)

Returns true if all or last (depending of the parameter) reply(es) were local generated (and not received).

Parameter may be all or last.


Example 1.30. t_local_replied usage

if (t_local_replied("all")) {
	log ("no reply received\n");

1.4.9.  t_was_cancelled()

Retuns true if called for an INVITE transaction that was explicitly cancelled by UAC side via a CANCEL request.

This function can be used from ONREPLY_ROUTE, FAILURE_ROUTE.

Example 1.31. t_was_cancelled usage

if (t_was_cancelled()) {
    log("transaction was cancelled by UAC\n");

1.4.10.  t_cancel_branch([flags])

This function is to be call when a reply is received for cancelling a set of branches (see flags) of the current call.

Meaning of the parameters is as follows:

  • flags - (optional) - set of flags (char based flags) to control what branches to be cancelled:

    • a - all - cancel all pending branches

    • o - others - cancel all the other pending branches except the current one

    • empty - current - cancel only the current branch

This function can be used from ONREPLY_ROUTE.

Example 1.32. t_cancel_branch usage

onreply_route[3] {
	if (t_check_status("183")) {
		# no support for early media

1.4.11.  t_new_request(method,RURI,from,to[,body[,ctx]])

This function generates and sends out a new SIP request (in a stateful way). The new request is completly unrelated to the currently processed SIP message.

Meaning of the parameters is as follows (all do accept variables):

  • method - the SIP method

  • RURI - the SIP Request URI (the request will be sent out to this destination)

  • from - the SIP From hdr information as "[display ]URI"

  • from - the SIP To hdr information as "[display ]URI"

  • body - (optional) the SIP body content starting with the content type string: "conten_type body"

  • ctx - a context string that will be added to the new transaction as an AVP with name "uac_ctx" (it may be visible in local route)

Example 1.33. t_new_request usage

	# send a MESSAGE request
	t_new_request("MESSAGE","sip:alice@","BOB","ALICE","text/plain Hello Alice!")) {

1.4.12.  t_on_failure(failure_route)

Sets reply routing block, to which control is passed after a transaction completed with a negative result but before sending a final reply. In the referred block, you can either start a new branch (good for services such as forward_on_no_reply) or send a final reply on your own (good for example for message silo, which received a negative reply from upstream and wants to tell upstream 202 I will take care of it).

As not all functions are available from failure route, please check the documentation for each function to see the permissions. Any other commands may result in unpredictable behavior and possible server failure.

Only one failure_route can be armed for a request. If you use many times t_on_failure(), only the last one has effect.

Note that whenever failure_route is entered, RURI is set to value of the winning branch.

Meaning of the parameters is as follows:

  • failure_route - Reply route block to be called.


Example 1.34. t_on_failure usage

route { 

failure_route[1] {

1.4.13.  t_on_reply(reply_route)

Sets reply routing block, to which control is passed each time a reply (provisional or final) for the transaction is received. The route is not called for local generated replies! In the referred block, you can inspect the reply and perform text operations on it.

As not all functions are available from this type of route, please check the documentation for each function to see the permissions. Any other commands may result in unpredictable behavior and possible server failure.

If called from branch route, the reply route will be set only for the current branch - that's it, it will be called only for relies belonging to that particular branch. Of course, from branch route, you can set different reply routes for each branch.

When called from a non-branc route, the reply route will be globally set for tha current transaction - it will be called for all replies belonging to that transaction. NOTE that only one> onreply_route can be armed for a transaction. If you use many times t_on_reply(), only the last one has effect.

If the processed reply is provisionla reply (1xx code), by calling the drop() function (exported by core), the execution of the route will end and the reply will not be forwarded further.

Meaning of the parameters is as follows:

  • reply_route - Reply route block to be called.


Example 1.35. t_on_reply usage

route {
	seturi("");  # first branch
	append_branch(""); # second branch

	t_on_reply("global"); # the "global" reply route 
	                      # is set the whole transaction


branch_route[1] {
	if ($rU=="alice")
		t_on_reply("alice"); # the "alice" reply route
		                      # is set only for second branch

onreply_route[alice] {
	xlog("received reply from alice\n");

onreply_route[global] {
	if (t_check_status("1[0-9][0-9]")) {
		log("provisional reply received\n");
		if (t_check_status("183"))

1.4.14.  t_on_branch(branch_route)

Sets a branch route to be execute separately for each branch of the transaction before being sent out - changes in that route should reflect only on that branch.

As not all functions are available from this type of route, please check the documentation for each function to see the permissions. Any other commands may result in unpredictable behavior and possible server failure.

Only one branch_route can be armed for a request. If you use many time t_on_branch(), only the last one has effect.

By calling the drop() function (exported by core), the execution of the branch route will end and the branch will not be forwarded further.

Meaning of the parameters is as follows:

  • branch_route - Branch route block to be called.


Example 1.36. t_on_branch usage

route { 

branch_route[1] {
	if ($ru=~"bad_uri") {
		xlog("dropping branch $ru \n");
	if ($ru=~"GW_uri") {

1.4.15.  t_inject_branches(source[,flags])

The function adds new SIP branches (destinations) to an existing transaction and fires them (sends them out). The transaction may already have ongoing branches (like in ringing state), which will not be affected by the injection of the new branches. Also it is possible for the transaction not to have any ongoing branches at the moment of the injection (still, the transaction must wait for new branches, even if all existing ones are completed - see the t_wait_for_new_branches function for this).

The main usage scenario for this function (and also what makes it different from t_relay is the ability to add new branches to an ongoing transaction from script routes not related to the transaction ( like timer route, event route, notification route, and other). In such routes, other functions/module used before the injection will point to the transaction to be affected by this injection - see the event_routing module.


  • source - where to take the description for the new branches to be injected. It can be

    • event - the branch will be taken from the event attributes exposed in an event notification route (see event_routing module).
    • msg - the branches will be taken from the RURI of the SIP message and from the additional branches (created by append_branch() function or similar).

  • flags - (optional) some additional flags related to the injection process:

    • cancel - cancel all the ongoing existing branches from the transaction before injecting the new branches.

Example 1.37. t_inject_branches usage

route[event_notification] {

1.4.16.  t_wait_for_new_branches()

This function instructs the existing SIP transaction to wait for new branches to be injected even after the completion of the existing branches. This waiting will be done until the Final Response INVITE timer (fr_inv_timeout) will hit for the transaction; of course, the waiting will be terminated if the transaction gets a 2xx final reply from one of the branches.

Normally if you have a transaction with two branches and they get, let's say, a 404 and 486 replies, the branches will be completed and transaction terminated by sending the 404 reply to the caller. Still, if you do t_wait_for_new_branches before relaying the transaction, the transaction will not terminate upon the completion of the branches and not send the 404 to the caller - it will wait for new branches to be injected (see t_inject_branches function) until the fr_inv timer hits.

Example 1.38. t_wait_for_new_branches usage


1.4.17.  t_add_hdrs("sip_hdrs")

Attach a set of headers to the existing transaction - these headers will be appended to all requests related to the transaction (outgoing branches, local ACKS, CANCELs).

Example 1.39. t_add_hdrs usage


1.4.18.  t_add_cancel_reason("Reason_hdr")

This function is used to enforce from the script level a custom "Reason" header into a CANCEL request. Normally, the Reason header is inherited form the received CANCEL (note that CANCEL propagates in a hop-by-hop manner - it is re-generated at each hop), but this function can overwrite it. It must be called before relaying the CANCEL request and its input must be a fully formated Reason header with name, body and CRLF.

Example 1.40. t_add_cancel_reason usage

t_add_cancel_reason('Reason: SIP ;cause=200;text="Call completed elsewhere"\r\n');

1.4.19.  t_replicate(URI,[flags])

Replicates a request to another destination. No information due the replicated request (like reply code) will be forwarded to the original SIP UAC.

The destination is specified by a SIP URI. If multiple destinations are to be used, the additional SIP URIs have to be set as branches.

The function may take as parameter an optional set of flags for controlling the internal behaviour - for description see the above t_relay([flags]) function. Note that only 0x4 is applicable here.

This functions can be used from REQUEST_ROUTE.

Example 1.41. t_replicate usage


1.4.20.  t_write_req(info,fifo) t_write_unix(info,sock)

Write via FIFO file or UNIX socket a lot of information regarding the request. Which information should be written may be control via the tw_append parameter.

This functions can be used from REQUEST_ROUTE, FAILURE_ROUTE and BRANCH_ROUTE.

Example 1.42. t_write_req/unix usage


1.4.21.  t_flush_flags()

Flush the flags from current request into the already created transaction. It make sense only in routing block if the transaction was created via t_newtran() and the flags have been altered since.

This function can be used from REQUEST_ROUTE and BRANCH_ROUTE .

Example 1.43. t_flush_flags usage


1.4.22.  t_anycast_replicate()

This function is used in an anycast setup to replicate a CANCEL or ACK method for whom there are no local transactions found. The function broadcasts the message to all the other nodes in the cluster, but only the owner of the transaction will be albe to handle it.

Example 1.44. t_anycast_replicate usage

if (is_method("ACK|CANCEL") && !t_check_trans()) {

1.5. Exported variables

Exported variables are listed in the next sections.

1.5.1. $T_branch_idx

$T_branch_idx - the index (starting with 0 for the first branch) of the currently proccessed branch. This index makes sense only in BRANCH and REPLY routes (where there is a concept of per branch processing). In all the other types of routes, the value of this index will be NULL.

1.5.2. $T_reply_code

$T_reply_code - the code of the reply, as follows: in request_route will be the last stateful sent reply; in reply_route will be the current processed reply; in failure_route will be the negative winning reply. In case of no-reply or error, '0' value is returned.

1.5.3. $T_fr_timeout

$T_fr_timeout (R/W) - the timeout for the final reply to the current transaction

With each different request received, $T_fr_timeout will initially be equal to the fr_timeout parameter.

"$T_fr_timeout = NULL;" will reset it to fr_timeout.

1.5.4. $T_fr_inv_timeout

$T_fr_inv_timeout (R/W) - the timeout for the final reply to an INVITE request, after a 1XX reply was received

With each different request received, $T_fr_inv_timeout will initially be equal to the fr_inv_timeout parameter.

"$T_fr_inv_timeout = NULL;" will reset it to fr_inv_timeout.

1.5.5. $T_ruri

$T_ruri - the ruri of the current branch; this information is taken from the transaction structure, so you can access this information for any sip message (request/reply) that has a transaction.

1.5.6. $bavp(name)

$bavp(name) - a particular type of avp that can have different values for each branch. They can only be used in BRANCH, REPLY and FAILURE routes. Otherwise NULL value is returned.

1.5.7. $T_id

$T_id - returns the ID of the current transaction. The ID is an opaque hexa string, unique for each transaction. If there is no current transaction, NULL value is returned.

1.6. Exported MI Functions

1.6.1.  t_uac_dlg

Generates and sends a local SIP request.


  • method - request method

  • RURI - request SIP URI

  • NEXT HOP - next hop SIP URI (OBP); use . if no value.

  • socket - local socket to be used for sending the request; use . if no value.

  • headers - set of additional headers to be added to the request; at least From and To headers must be specify)

  • body - (optional, may not be present) request body (if present, requires the Content-Type and Content-length headers)

1.6.2.  t_uac_cancel

Generates and sends a CANCEL for an existing SIP request.


  • callid - callid of the INVITE request to be cancelled.

  • cseq - cseq of the INVITE request to be cancelled.

1.6.3.  t_hash

Gets information about the load of TM internal hash table.


  • none

1.6.4.  t_reply

Generates and sends a reply for an existing inbound SIP transaction.


  • code - reply code

  • reason - reason phrase.

  • trans_id - transaction identifier (has the hash_entry:label format)

  • to_tag - To tag to be added to TO header

  • new_headers - extra headers to be appended to the reply; use a dot (.) char only if there are no headers;

  • body - (optional, may not be present) reply body (if present, requires the Content-Type and Content-length headers)

1.7. Exported statistics

Exported statistics are listed in the next sections. All statistics except inuse_transactions can be reset.

1.7.1. received_replies

Total number of total replies received by TM module.

1.7.2. relayed_replies

Total number of replies received and relayed by TM module.

1.7.3. local_replies

Total number of replies local generated by TM module.

1.7.4. UAS_transactions

Total number of transactions created by received requests.

1.7.5. UAC_transactions

Total number of transactions created by local generated requests.

1.7.6. 2xx_transactions

Total number of transactions completed with 2xx replies.

1.7.7. 3xx_transactions

Total number of transactions completed with 3xx replies.

1.7.8. 4xx_transactions

Total number of transactions completed with 4xx replies.

1.7.9. 5xx_transactions

Total number of transactions completed with 5xx replies.

1.7.10. 6xx_transactions

Total number of transactions completed with 6xx replies.

1.7.11. inuse_transactions

Number of transactions existing in memory at current time.

Chapter 2. Developer Guide

2.1. Functions

2.1.1.  load_tm(*import_structure)

For programmatic use only--import the TM API. See the cpl_c, acc or jabber modules to see how it works.

Meaning of the parameters is as follows:

  • import_structure - Pointer to the import structure - see struct tm_binds in modules/tm/tm_load.h

Chapter 3. Frequently Asked Questions


What happened with old cancel_call() function

The function was replace (as functionality) by cancel_branch("a") - cancel all braches.


How can I report a bug?

Please follow the guidelines provided at: