Internet Engineering Task Force (IETF) V. Hilt Request for Comments: 6357 Bell Labs/Alcatel-Lucent Category: Informational E. Noel ISSN: 2070-1721 AT&T Labs C. Shen Columbia University A. Abdelal Sonus Networks August 2011 Design Considerations for Session Initiation Protocol (SIP) Overload Control Abstract Overload occurs in Session Initiation Protocol (SIP) networks when SIP servers have insufficient resources to handle all SIP messages they receive. Even though the SIP protocol provides a limited overload control mechanism through its 503 (Service Unavailable) response code, SIP servers are still vulnerable to overload. This document discusses models and design considerations for a SIP overload control mechanism. Status of This Memo This document is not an Internet Standards Track specification; it is published for informational purposes. This document is a product of the Internet Engineering Task Force (IETF). It represents the consensus of the IETF community. It has received public review and has been approved for publication by the Internet Engineering Steering Group (IESG). Not all documents approved by the IESG are a candidate for any level of Internet Standard; see Section 2 of RFC 5741. Information about the current status of this document, any errata, and how to provide feedback on it may be obtained at http://www.rfc-editor.org/info/rfc6357.
Copyright Notice Copyright (c) 2011 IETF Trust and the persons identified as the document authors. All rights reserved. This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (http://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License. Table of Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3 2. SIP Overload Problem . . . . . . . . . . . . . . . . . . . . . 4 3. Explicit vs. Implicit Overload Control . . . . . . . . . . . . 5 4. System Model . . . . . . . . . . . . . . . . . . . . . . . . . 6 5. Degree of Cooperation . . . . . . . . . . . . . . . . . . . . 8 5.1. Hop-by-Hop . . . . . . . . . . . . . . . . . . . . . . . . 9 5.2. End-to-End . . . . . . . . . . . . . . . . . . . . . . . . 10 5.3. Local Overload Control . . . . . . . . . . . . . . . . . . 11 6. Topologies . . . . . . . . . . . . . . . . . . . . . . . . . . 12 7. Fairness . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 8. Performance Metrics . . . . . . . . . . . . . . . . . . . . . 14 9. Explicit Overload Control Feedback . . . . . . . . . . . . . . 15 9.1. Rate-Based Overload Control . . . . . . . . . . . . . . . 15 9.2. Loss-Based Overload Control . . . . . . . . . . . . . . . 17 9.3. Window-Based Overload Control . . . . . . . . . . . . . . 18 9.4. Overload Signal-Based Overload Control . . . . . . . . . . 19 9.5. On-/Off Overload Control . . . . . . . . . . . . . . . . . 19 10. Implicit Overload Control . . . . . . . . . . . . . . . . . . 20 11. Overload Control Algorithms . . . . . . . . . . . . . . . . . 20 12. Message Prioritization . . . . . . . . . . . . . . . . . . . . 21 13. Operational Considerations . . . . . . . . . . . . . . . . . . 21 14. Security Considerations . . . . . . . . . . . . . . . . . . . 22 15. Informative References . . . . . . . . . . . . . . . . . . . . 23 Appendix A. Contributors . . . . . . . . . . . . . . . . . . . . 25
1. Introduction As with any network element, a Session Initiation Protocol (SIP) [RFC3261] server can suffer from overload when the number of SIP messages it receives exceeds the number of messages it can process. Overload occurs if a SIP server does not have sufficient resources to process all incoming SIP messages. These resources may include CPU, memory, input/output, or disk resources. Overload can pose a serious problem for a network of SIP servers. During periods of overload, the throughput of SIP messages in a network of SIP servers can be significantly degraded. In fact, overload in a SIP server may lead to a situation in which the overload is amplified by retransmissions of SIP messages causing the throughput to drop down to a very small fraction of the original processing capacity. This is often called congestion collapse. An overload control mechanism enables a SIP server to process SIP messages close to its capacity limit during times of overload. Overload control is used by a SIP server if it is unable to process all SIP requests due to resource constraints. There are other failure cases in which a SIP server can successfully process incoming requests but has to reject them for other reasons. For example, a Public Switched Telephone Network (PSTN) gateway that runs out of trunk lines but still has plenty of capacity to process SIP messages should reject incoming INVITEs using a response such as 488 (Not Acceptable Here), as described in [RFC4412]. Similarly, a SIP registrar that has lost connectivity to its registration database but is still capable of processing SIP messages should reject REGISTER requests with a 500 (Server Error) response [RFC3261]. Overload control mechanisms do not apply in these cases and SIP provides appropriate response codes for them. There are cases in which a SIP server runs other services that do not involve the processing of SIP messages (e.g., processing of RTP packets, database queries, software updates, and event handling). These services may, or may not, be correlated with the SIP message volume. These services can use up a substantial share of resources available on the server (e.g., CPU cycles) and leave the server in a condition where it is unable to process all incoming SIP requests. In these cases, the SIP server applies SIP overload control mechanisms to avoid congestion collapse on the SIP signaling plane. However, controlling the number of SIP requests may not significantly reduce the load on the server if the resource shortage was created by another service. In these cases, it is to be expected that the server uses appropriate methods of controlling the resource usage of
other services. The specifics of controlling the resource usage of other services and their coordination is out of scope for this document. The SIP protocol provides a limited mechanism for overload control through its 503 (Service Unavailable) response code and the Retry-After header. However, this mechanism cannot prevent overload of a SIP server and it cannot prevent congestion collapse. In fact, it may cause traffic to oscillate and to shift between SIP servers and thereby worsen an overload condition. A detailed discussion of the SIP overload problem, the problems with the 503 (Service Unavailable) response code and the Retry-After header, and the requirements for a SIP overload control mechanism can be found in [RFC5390]. In addition, 503 is used for other situations, not just SIP server overload. A SIP overload control process based on 503 would have to specify exactly which cause values trigger the overload control. This document discusses the models, assumptions, and design considerations for a SIP overload control mechanism. The document originated in the SIP overload control design team and has been further developed by the SIP Overload Control (SOC) working group. 2. SIP Overload Problem A key contributor to SIP congestion collapse [RFC5390] is the regenerative behavior of overload in the SIP protocol. When SIP is running over the UDP protocol, it will retransmit messages that were dropped or excessively delayed by a SIP server due to overload and thereby increase the offered load for the already overloaded server. This increase in load worsens the severity of the overload condition and, in turn, causes more messages to be dropped. A congestion collapse can occur [Hilt] [Noel] [Shen] [Abdelal]. Regenerative behavior under overload should ideally be avoided by any protocol as this would lead to unstable operation under overload. However, this is often difficult to achieve in practice. For example, changing the SIP retransmission timer mechanisms can reduce the degree of regeneration during overload but will impact the ability of SIP to recover from message losses. Without any retransmission, each message that is dropped due to SIP server overload will eventually lead to a failed transaction. For a SIP INVITE transaction to be successful, a minimum of three messages need to be forwarded by a SIP server. Often an INVITE transaction consists of five or more SIP messages. If a SIP server under overload randomly discards messages without evaluating them, the chances that all messages belonging to a transaction are
successfully forwarded will decrease as the load increases. Thus, the number of transactions that complete successfully will decrease even if the message throughput of a server remains up and assuming the overload behavior is fully non-regenerative. A SIP server might (partially) parse incoming messages to determine if it is a new request or a message belonging to an existing transaction. Discarding a SIP message after spending the resources to parse it is expensive. The number of successful transactions will therefore decline with an increase in load as fewer resources can be spent on forwarding messages and more resources are consumed by inspecting messages that will eventually be dropped. The rate of the decline depends on the amount of resources spent to inspect each message. Another challenge for SIP overload control is controlling the rate of the true traffic source. Overload is often caused by a large number of user agents (UAs), each of which creates only a single message. However, the sum of their traffic can overload a SIP server. The overload mechanisms suitable for controlling a SIP server (e.g., rate control) may not be effective for individual UAs. In some cases, there are other non-SIP mechanisms for limiting the load from the UAs. These may operate independently from, or in conjunction with, the SIP overload mechanisms described here. In either case, they are out of scope for this document. 3. Explicit vs. Implicit Overload Control The main difference between explicit and implicit overload control is the way overload is signaled from a SIP server that is reaching overload condition to its upstream neighbors. In an explicit overload control mechanism, a SIP server uses an explicit overload signal to indicate that it is reaching its capacity limit. Upstream neighbors receiving this signal can adjust their transmission rate according to the overload signal to a level that is acceptable to the downstream server. The overload signal enables a SIP server to steer the load it is receiving to a rate at which it can perform at maximum capacity. Implicit overload control uses the absence of responses and packet loss as an indication of overload. A SIP server that is sensing such a condition reduces the load it is forwarding to a downstream neighbor. Since there is no explicit overload signal, this mechanism is robust, as it does not depend on actions taken by the SIP server running into overload. The ideas of explicit and implicit overload control are in fact complementary. By considering implicit overload indications, a server can avoid overloading an unresponsive downstream neighbor. An
explicit overload signal enables a SIP server to actively steer the incoming load to a desired level. 4. System Model The model shown in Figure 1 identifies fundamental components of an explicit SIP overload control mechanism: SIP Processor: The SIP Processor processes SIP messages and is the component that is protected by overload control. Monitor: The Monitor measures the current load of the SIP Processor on the receiving entity. It implements the mechanisms needed to determine the current usage of resources relevant for the SIP Processor and reports load samples (S) to the Control Function. Control Function: The Control Function implements the overload control algorithm. The Control Function uses the load samples (S) and determines if overload has occurred and a throttle (T) needs to be set to adjust the load sent to the SIP Processor on the receiving entity. The Control Function on the receiving entity sends load feedback (F) to the sending entity. Actuator: The Actuator implements the algorithms needed to act on the throttles (T) and ensures that the amount of traffic forwarded to the receiving entity meets the criteria of the throttle. For example, a throttle may instruct the Actuator to not forward more than 100 INVITE messages per second. The Actuator implements the algorithms to achieve this objective, e.g., using message gapping. It also implements algorithms to select the messages that will be affected and determine whether they are rejected or redirected. The type of feedback (F) conveyed from the receiving to the sending entity depends on the overload control method used (i.e., loss-based, rate-based, window-based, or signal-based overload control; see Section 9), the overload control algorithm (see Section 11), as well as other design parameters. The feedback (F) enables the sending entity to adjust the amount of traffic forwarded to the receiving entity to a level that is acceptable to the receiving entity without causing overload. Figure 1 depicts a general system model for overload control. In this diagram, one instance of the control function is on the sending entity (i.e., associated with the actuator) and one is on the receiving entity (i.e., associated with the Monitor). However, a specific mechanism may not require both elements. In this case, one of two control function elements can be empty and simply passes along feedback. For example, if (F) is defined as a loss-rate (e.g.,
reduce traffic by 10%), there is no need for a control function on the sending entity as the content of (F) can be copied directly into (T). The model in Figure 1 shows a scenario with one sending and one receiving entity. In a more realistic scenario, a receiving entity will receive traffic from multiple sending entities and vice versa (see Section 6). The feedback generated by a Monitor will therefore often be distributed across multiple Actuators. A Monitor needs to be able to split the load it can process across multiple sending entities and generate feedback that correctly adjusts the load each sending entity is allowed to send. Similarly, an Actuator needs to be prepared to receive different levels of feedback from different receiving entities and throttle traffic to these entities accordingly. In a realistic deployment, SIP messages will flow in both directions, from server B to server A as well as server A to server B. The overload control mechanisms in each direction can be considered independently. For messages flowing from server A to server B, the sending entity is server A and the receiving entity is server B, and vice versa. The control loops in both directions operate independently. Sending Receiving Entity Entity +----------------+ +----------------+ | Server A | | Server B | | +----------+ | | +----------+ | -+ | | Control | | F | | Control | | | | | Function |<-+------+--| Function | | | | +----------+ | | +----------+ | | | T | | | ^ | | Overload | v | | | S | | Control | +----------+ | | +----------+ | | | | Actuator | | | | Monitor | | | | +----------+ | | +----------+ | | | | | | ^ | -+ | v | | | | -+ | +----------+ | | +----------+ | | <-+--| SIP | | | | SIP | | | SIP --+->|Processor |--+------+->|Processor |--+-> | System | +----------+ | | +----------+ | | +----------------+ +----------------+ -+ Figure 1: System Model for Explicit Overload Control
5. Degree of Cooperation A SIP request is usually processed by more than one SIP server on its path to the destination. Thus, a design choice for an explicit overload control mechanism is where to place the components of overload control along the path of a request and, in particular, where to place the Monitor and Actuator. This design choice determines the degree of cooperation between the SIP servers on the path. Overload control can be implemented hop-by-hop with the Monitor on one server and the Actuator on its direct upstream neighbor. Overload control can be implemented end-to-end with Monitors on all SIP servers along the path of a request and an Actuator on the sender. In this case, the Control Functions associated with each Monitor have to cooperate to jointly determine the overall feedback for this path. Finally, overload control can be implemented locally on a SIP server if the Monitor and Actuator reside on the same server. In this case, the sending entity and receiving entity are the same SIP server, and the Actuator and Monitor operate on the same SIP Processor (although, the Actuator typically operates on a pre-processing stage in local overload control). Local overload control is an internal overload control mechanism, as the control loop is implemented internally on one server. Hop-by-hop and end-to-end are external overload control mechanisms. All three configurations are shown in Figure 2.
+---------+ +------(+)---------+ +------+ | | | ^ | | | | +---+ | | +---+ v | v //=>| C | v | //=>| C | +---+ +---+ // +---+ +---+ +---+ // +---+ | A |===>| B | | A |===>| B | +---+ +---+ \\ +---+ +---+ +---+ \\ +---+ ^ \\=>| D | ^ | \\=>| D | | +---+ | | +---+ | | | v | +---------+ +------(+)---------+ (a) hop-by-hop (b) end-to-end +-+ v | +-+ +-+ +---+ v | v | //=>| C | +---+ +---+ // +---+ | A |===>| B | +---+ +---+ \\ +---+ \\=>| D | +---+ ^ | +-+ (c) local ==> SIP request flow <-- Overload feedback loop Figure 2: Degree of Cooperation between Servers 5.1. Hop-by-Hop The idea of hop-by-hop overload control is to instantiate a separate control loop between all neighboring SIP servers that directly exchange traffic. That is, the Actuator is located on the SIP server that is the direct upstream neighbor of the SIP server that has the corresponding Monitor. Each control loop between two servers is completely independent of the control loop between other servers further up- or downstream. In the example in Figure 2(a), three independent overload control loops are instantiated: A - B, B - C, and B - D. Each loop only controls a single hop. Overload feedback received from a downstream neighbor is not forwarded further upstream. Instead, a SIP server acts on this feedback, for example, by rejecting SIP messages if needed. If the upstream neighbor of a server also becomes overloaded, it will report this problem to its
upstream neighbors, which again take action based on the reported feedback. Thus, in hop-by-hop overload control, overload is always resolved by the direct upstream neighbors of the overloaded server without the need to involve entities that are located multiple SIP hops away. Hop-by-hop overload control reduces the impact of overload on a SIP network and can avoid congestion collapse. It is simple and scales well to networks with many SIP entities. An advantage is that it does not require feedback to be transmitted across multiple-hops, possibly crossing multiple trust domains. Feedback is sent to the next hop only. Furthermore, it does not require a SIP entity to aggregate a large number of overload status values or keep track of the overload status of SIP servers it is not communicating with. 5.2. End-to-End End-to-end overload control implements an overload control loop along the entire path of a SIP request, from user agent client (UAC) to user agent server (UAS). An end-to-end overload control mechanism consolidates overload information from all SIP servers on the way (including all proxies and the UAS) and uses this information to throttle traffic as far upstream as possible. An end-to-end overload control mechanism has to be able to frequently collect the overload status of all servers on the potential path(s) to a destination and combine this data into meaningful overload feedback. A UA or SIP server only throttles requests if it knows that these requests will eventually be forwarded to an overloaded server. For example, if D is overloaded in Figure 2(b), A should only throttle requests it forwards to B when it knows that they will be forwarded to D. It should not throttle requests that will eventually be forwarded to C, since server C is not overloaded. In many cases, it is difficult for A to determine which requests will be routed to C and D, since this depends on the local routing decision made by B. These routing decisions can be highly variable and, for example, depend on call-routing policies configured by the user, services invoked on a call, load-balancing policies, etc. A previous message to a target that has been routed through an overloaded server does not necessarily mean that the next message to this target will also be routed through the same server. The main problem of end-to-end overload control is its inherent complexity, since UAC or SIP servers need to monitor all potential paths to a destination in order to determine which requests should be throttled and which requests may be sent. Even if this information is available, it is not clear which path a specific request will take.
A variant of end-to-end overload control is to implement a control loop between a set of well-known SIP servers along the path of a SIP request. For example, an overload control loop can be instantiated between a server that only has one downstream neighbor or a set of closely coupled SIP servers. A control loop spanning multiple hops can be used if the sending entity has full knowledge about the SIP servers on the path of a SIP message. Overload control for SIP servers is different from end-to-end congestion control used by transport protocols such as TCP. The traffic exchanged between SIP servers consists of many individual SIP messages. Each SIP message is created by a SIP UA to achieve a specific goal (e.g., to start setting up a call). All messages have their own source and destination addresses. Even SIP messages containing identical SIP URIs (e.g., a SUBSCRIBE and an INVITE message to the same SIP URI) can be routed to different destinations. This is different from TCP, where the traffic exchanged between routers consists of packets belonging to a usually longer flow of messages exchanged between a source and a destination (e.g., to transmit a file). If congestion occurs, the sources can detect this condition and adjust the rate at which the next packets are transmitted. 5.3. Local Overload Control The idea of local overload control (see Figure 2(c)) is to run the Monitor and Actuator on the same server. This enables the server to monitor the current resource usage and to reject messages that can't be processed without overusing local resources. The fundamental assumption behind local overload control is that it is less resource consuming for a server to reject messages than to process them. A server can therefore reject the excess messages it cannot process to stop all retransmissions of these messages. Since rejecting messages does consume resources on a SIP server, local overload control alone cannot prevent a congestion collapse. Local overload control can be used in conjunction with other overload control mechanisms and provides an additional layer of protection against overload. It is fully implemented within a SIP server and does not require cooperation between servers. In general, SIP servers should apply other overload control techniques to control load before a local overload control mechanism is activated as a mechanism of last resort.
6. Topologies The following topologies describe four generic SIP server configurations. These topologies illustrate specific challenges for an overload control mechanism. An actual SIP server topology is likely to consist of combinations of these generic scenarios. In the "load balancer" configuration shown in Figure 3(a), a set of SIP servers (D, E, and F) receives traffic from a single source A. A load balancer is a typical example for such a configuration. In this configuration, overload control needs to prevent server A (i.e., the load balancer) from sending too much traffic to any of its downstream neighbors D, E, and F. If one of the downstream neighbors becomes overloaded, A can direct traffic to the servers that still have capacity. If one of the servers acts as a backup, it can be activated once one of the primary servers reaches overload. If A can reliably determine that D, E, and F are its only downstream neighbors and all of them are in overload, it may choose to report overload upstream on behalf of D, E, and F. However, if the set of downstream neighbors is not fixed or only some of them are in overload, then A should not activate an overload control since A can still forward the requests destined to non-overloaded downstream neighbors. These requests would be throttled as well if A would use overload control towards its upstream neighbors. In some cases, the servers D, E, and F are in a server farm and are configured to appear as a single server to their upstream neighbors. In this case, server A can report overload on behalf of the server farm. If the load balancer is not a SIP entity, servers D, E, and F can report the overall load of the server farm (i.e., the load of the virtual server) in their messages. As an alternative, one of the servers (e.g., server E) can report overload on behalf of the server farm. In this case, not all messages contain overload control information, and all upstream neighbors need to be served by server E periodically to ensure that updated information is received. In the "multiple sources" configuration shown in Figure 3(b), a SIP server D receives traffic from multiple upstream sources A, B, and C. Each of these sources can contribute a different amount of traffic, which can vary over time. The set of active upstream neighbors of D can change as servers may become inactive, and previously inactive servers may start contributing traffic to D. If D becomes overloaded, it needs to generate feedback to reduce the amount of traffic it receives from its upstream neighbors. D needs to decide by how much each upstream neighbor should reduce traffic. This decision can require the consideration of the amount of traffic
sent by each upstream neighbor and it may need to be re-adjusted as the traffic contributed by each upstream neighbor varies over time. Server D can use a local fairness policy to determine how much traffic it accepts from each upstream neighbor. In many configurations, SIP servers form a "mesh" as shown in Figure 3(c). Here, multiple upstream servers A, B, and C forward traffic to multiple alternative servers D and E. This configuration is a combination of the "load balancer" and "multiple sources" scenario. +---+ +---+ /->| D | | A |-\ / +---+ +---+ \ / \ +---+ +---+-/ +---+ +---+ \->| | | A |------>| E | | B |------>| D | +---+-\ +---+ +---+ /->| | \ / +---+ \ +---+ +---+ / \->| F | | C |-/ +---+ +---+ (a) load balancer (b) multiple sources +---+ | A |---\ a--\ +---+-\ \---->+---+ \ \/----->| D | b--\ \--->+---+ +---+--/\ /-->+---+ \---->| | | B | \/ c-------->| D | +---+---\/\--->+---+ | | /\---->| E | ... /--->+---+ +---+--/ /-->+---+ / | C |-----/ z--/ +---+ (c) mesh (d) edge proxy Figure 3: Topologies Overload control that is based on reducing the number of messages a sender is allowed to send is not suited for servers that receive requests from a very large population of senders, each of which only sends a very small number of requests. This scenario is shown in Figure 3(d). An edge proxy that is connected to many UAs is a typical example for such a configuration. Since each UA typically infrequently sends requests, which are often related to the same session, it can't decrease its message rate to resolve the overload.
A SIP server that receives traffic from many sources, which each contribute only a small number of requests, can resort to local overload control by rejecting a percentage of the requests it receives with 503 (Service Unavailable) responses. Since it has many upstream neighbors, it can send 503 (Service Unavailable) to a fraction of them to gradually reduce load without entirely stopping all incoming traffic. The Retry-After header can be used in 503 (Service Unavailable) responses to ask upstream neighbors to wait a given number of seconds before trying the request again. Using 503 (Service Unavailable) can, however, not prevent overload if a large number of sources create requests (e.g., to place calls) at the same time. Note: The requirements of the "edge proxy" topology are different from the ones of the other topologies, which may require a different method for overload control. 7. Fairness There are many different ways to define fairness between multiple upstream neighbors of a SIP server. In the context of SIP server overload, it is helpful to describe two categories of fairness: basic fairness and customized fairness. With basic fairness, a SIP server treats all requests equally and ensures that each request has the same chance of succeeding. With customized fairness, the server allocates resources according to different priorities. An example application of the basic fairness criteria is the "Third caller receives free tickets" scenario, where each call attempt should have an equal success probability in connecting through an overloaded SIP server, irrespective of the service provider in which the call was initiated. An example of customized fairness would be a server that assigns different resource allocations to its upstream neighbors (e.g., service providers) as defined in a service level agreement (SLA). 8. Performance Metrics The performance of an overload control mechanism can be measured using different metrics. A key performance indicator is the goodput of a SIP server under overload. Ideally, a SIP server will be enabled to perform at its maximum capacity during periods of overload. For example, if a SIP server has a processing capacity of 140 INVITE transactions per second, then an overload control mechanism should enable it to process 140 INVITEs per second even if the offered load is much higher. The delay introduced by a SIP server is another important indicator. An overload control mechanism should ensure that the
delay encountered by a SIP message is not increased significantly during periods of overload. Significantly increased delay can lead to time-outs and retransmission of SIP messages, making the overload worse. Responsiveness and stability are other important performance indicators. An overload control mechanism should quickly react to an overload occurrence and ensure that a SIP server does not become overloaded, even during sudden peaks of load. Similarly, an overload control mechanism should quickly stop rejecting requests if the overload disappears. Stability is another important criteria. An overload control mechanism should not cause significant oscillations of load on a SIP server. The performance of SIP overload control mechanisms is discussed in [Noel], [Shen], [Hilt], and [Abdelal]. In addition to the above metrics, there are other indicators that are relevant for the evaluation of an overload control mechanism: Fairness: Which type of fairness does the overload control mechanism implement? Self-limiting: Is the overload control self-limiting if a SIP server becomes unresponsive? Changes in neighbor set: How does the mechanism adapt to a changing set of sending entities? Data points to monitor: Which and how many data points does an overload control mechanism need to monitor? Computational load: What is the (CPU) load created by the overload "Monitor" and "Actuator"? 9. Explicit Overload Control Feedback Explicit overload control feedback enables a receiver to indicate how much traffic it wants to receive. Explicit overload control mechanisms can be differentiated based on the type of information conveyed in the overload control feedback and whether the control function is in the receiving or sending entity (receiver- vs. sender- based overload control), or both. 9.1. Rate-Based Overload Control The key idea of rate-based overload control is to limit the request rate at which an upstream element is allowed to forward traffic to the downstream neighbor. If overload occurs, a SIP server instructs
each upstream neighbor to send, at most, X requests per second. Each upstream neighbor can be assigned a different rate cap. An example algorithm for an Actuator in the sending entity is request gapping. After transmitting a request to a downstream neighbor, a server waits for 1/X seconds before it transmits the next request to the same neighbor. Requests that arrive during the waiting period are not forwarded and are either redirected, rejected, or buffered. Request gapping only affects requests that are targeted by overload control (e.g., requests that initiate a transaction and not retransmissions in an ongoing transaction). The rate cap ensures that the number of requests received by a SIP server never increases beyond the sum of all rate caps granted to upstream neighbors. Rate-based overload control protects a SIP server against overload, even during load spikes assuming there are no new upstream neighbors that start sending traffic. New upstream neighbors need to be considered in the rate caps assigned to all upstream neighbors. The rate assigned to upstream neighbors needs to be adjusted when new neighbors join. During periods when new neighbors are joining, overload can occur in extreme cases until the rate caps of all servers are adjusted to again match the overall rate cap of the server. The overall rate cap of a SIP server is determined by an overload control algorithm, e.g., based on system load. Rate-based overload control requires a SIP server to assign a rate cap to each of its upstream neighbors while it is activated. Effectively, a server needs to assign a share of its overall capacity to each upstream neighbor. A server needs to ensure that the sum of all rate caps assigned to upstream neighbors does not substantially oversubscribe its actual processing capacity. This requires a SIP server to keep track of the set of upstream neighbors and to adjust the rate cap if a new upstream neighbor appears or an existing neighbor stops transmitting. For example, if the capacity of the server is X and this server is receiving traffic from two upstream neighbors, it can assign a rate of X/2 to each of them. If a third sender appears, the rate for each sender is lowered to X/3. If the overall rate cap is too high, a server may experience overload. If the cap is too low, the upstream neighbors will reject requests even though they could be processed by the server. An approach for estimating a rate cap for each upstream neighbor is using a fixed proportion of a control variable, X, where X is initially equal to the capacity of the SIP server. The server then increases or decreases X until the workload arrival rate matches the actual server capacity. Usually, this will mean that the sum of the rate caps sent out by the server (=X) exceeds its actual capacity,
but enables upstream neighbors who are not generating more than their fair share of the work to be effectively unrestricted. In this approach, the server only has to measure the aggregate arrival rate. However, since the overall rate cap is usually higher than the actual capacity, brief periods of overload may occur. 9.2. Loss-Based Overload Control A loss percentage enables a SIP server to ask an upstream neighbor to reduce the number of requests it would normally forward to this server by X%. For example, a SIP server can ask an upstream neighbor to reduce the number of requests this neighbor would normally send by 10%. The upstream neighbor then redirects or rejects 10% of the traffic that is destined for this server. To implement a loss percentage, the sending entity may employ an algorithm to draw a random number between 1 and 100 for each request to be forwarded. The request is not forwarded to the server if the random number is less than or equal to X. An advantage of loss-based overload control is that the receiving entity does not need to track the set of upstream neighbors or the request rate it receives from each upstream neighbor. It is sufficient to monitor the overall system utilization. To reduce load, a server can ask its upstream neighbors to lower the traffic forwarded by a certain percentage. The server calculates this percentage by combining the loss percentage that is currently in use (i.e., the loss percentage the upstream neighbors are currently using when forwarding traffic), the current system utilization, and the desired system utilization. For example, if the server load approaches 90% and the current loss percentage is set to a 50% traffic reduction, then the server can decide to increase the loss percentage to 55% in order to get to a system utilization of 80%. Similarly, the server can lower the loss percentage if permitted by the system utilization. Loss-based overload control requires that the throttle percentage be adjusted to the current overall number of requests received by the server. This is particularly important if the number of requests received fluctuates quickly. For example, if a SIP server sets a throttle value of 10% at time t1 and the number of requests increases by 20% between time t1 and t2 (t1<t2), then the server will see an increase in traffic by 10% between time t1 and t2. This is even though all upstream neighbors have reduced traffic by 10%. Thus, percentage throttling requires an adjustment of the throttling percentage in response to the traffic received and may not always be able to prevent a server from encountering brief periods of overload in extreme cases.
9.3. Window-Based Overload Control The key idea of window-based overload control is to allow an entity to transmit a certain number of messages before it needs to receive a confirmation for the messages in transit. Each sender maintains an overload window that limits the number of messages that can be in transit without being confirmed. Window-based overload control is inspired by TCP [RFC0793]. Each sender maintains an unconfirmed message counter for each downstream neighbor it is communicating with. For each message sent to the downstream neighbor, the counter is increased. For each confirmation received, the counter is decreased. The sender stops transmitting messages to the downstream neighbor when the unconfirmed message counter has reached the current window size. A crucial parameter for the performance of window-based overload control is the window size. Each sender has an initial window size it uses when first sending a request. This window size can be changed based on the feedback it receives from the receiver. The sender adjusts its window size as soon as it receives the corresponding feedback from the receiver. If the new window size is smaller than the current unconfirmed message counter, the sender stops transmitting messages until more messages are confirmed and the current unconfirmed message counter is less than the window size. Note that the reception of a 100 (Trying) response does not provide a confirmation for the successful processing of a message. 100 (Trying) responses are often created by a SIP server very early in processing and do not indicate that a message has been successfully processed and cleared from the input buffer. If the downstream neighbor is a stateless proxy, it will not create 100 (Trying) responses at all and will instead pass through 100 (Trying) responses created by the next stateful server. Also, 100 (Trying) responses are typically only created for INVITE requests. Explicit message confirmations do not have these problems. Window-based overload control is similar to rate-based overload control in that the total available receiver buffer space needs to be divided among all upstream neighbors. However, unlike rate-based overload control, window-based overload control is self-limiting and can ensure that the receiver buffer does not overflow under normal conditions. The transmission of messages by senders is clocked by message confirmations received from the receiver. A buffer overflow can occur in extreme cases when a large number of new upstream
neighbors arrives at the same time. However, senders will eventually stop transmitting new requests once their initial sending window is closed. In window-based overload control, the number of messages a sender is allowed to send can frequently be set to zero. In this state, the sender needs to be informed when it is allowed to send again and when the receiver window has opened up. However, since the sender is not allowed to transmit messages, the receiver cannot convey the new window size by piggybacking it in a response to another message. Instead, it needs to inform the sender through another mechanism, e.g., by sending a message that contains the new window size. 9.4. Overload Signal-Based Overload Control The key idea of overload signal-based overload control is to use the transmission of a 503 (Service Unavailable) response as a signal for overload in the downstream neighbor. After receiving a 503 (Service Unavailable) response, the sender reduces the load forwarded to the downstream neighbor to avoid triggering more 503 (Service Unavailable) responses. The sender keeps reducing the load if more 503 (Service Unavailable) responses are received. Note that this scheme is based on the use of 503 (Service Unavailable) responses without the Retry-After header, as the Retry-After header would require a sender to entirely stop forwarding requests. It should also be noted that 503 responses can be generated for reasons other than overload (e.g., server maintenance). A sender that has not received 503 (Service Unavailable) responses for a while but is still throttling traffic can start to increase the offered load. By slowly increasing the traffic forwarded, a sender can detect that overload in the downstream neighbor has been resolved and more load can be forwarded. The load is increased until the sender receives another 503 (Service Unavailable) response or is forwarding all requests it has. A possible algorithm for adjusting traffic is additive increase/multiplicative decrease (AIMD). Overload signal-based overload control is a sender-based overload control mechanism. 9.5. On-/Off Overload Control On-/off overload control feedback enables a SIP server to turn the traffic it is receiving either on or off. The 503 (Service Unavailable) response with a Retry-After header implements on-/off overload control. On-/off overload control is less effective in controlling load than the fine grained control methods above. All of
the above methods can realize on-/off overload control, e.g., by setting the allowed rate to either zero or unlimited. 10. Implicit Overload Control Implicit overload control ensures that the transmission of a SIP server is self-limiting. It slows down the transmission rate of a sender when there is an indication that the receiving entity is experiencing overload. Such an indication can be that the receiving entity is not responding within the expected timeframe or is not responding at all. The idea of implicit overload control is that senders should try to sense overload of a downstream neighbor even if there is no explicit overload control feedback. It avoids an overloaded server, which has become unable to generate overload control feedback, from being overwhelmed with requests. Window-based overload control is inherently self-limiting since a sender cannot continue to pass messages without receiving confirmations. All other explicit overload control schemes described above do not have this property and require additional implicit controls to limit transmissions in case an overloaded downstream neighbor does not generate explicit feedback. 11. Overload Control Algorithms An important aspect of the design of an overload control mechanism is the overload control algorithm. The control algorithm determines when the amount of traffic to a SIP server needs to be decreased and when it can be increased. In terms of the model described in Section 4, the control algorithm takes (S) as an input value and generates (T) as a result. Overload control algorithms have been studied to a large extent and many different overload control algorithms exist. With many different overload control algorithms available, it seems reasonable to suggest a baseline algorithm in a specification for a SIP overload control mechanism and allow the use of other algorithms if they provide the same protocol semantics. This will also allow the development of future algorithms, which may lead to better performance. Conversely, the overload control mechanism should allow the use of different algorithms if they adhere to the defined protocol semantics.
12. Message Prioritization Overload control can require a SIP server to prioritize requests and select requests to be rejected or redirected. The selection is largely a matter of local policy of the SIP server, the overall network, and the services the SIP server provides. While there are many factors that can affect the prioritization of SIP requests, the Resource-Priority Header (RPH) field [RFC4412] is a prime candidate for marking the prioritization of SIP requests. Depending on the particular network and the services it offers, a particular namespace and priority value in the RPH could indicate i) a high priority request, which should be preserved if possible during overload, ii) a low priority request, which should be dropped during overload, or iii) a label, which has no impact on message prioritization in this network. For a number of reasons, responses should not be targeted in order to reduce SIP server load. Responses cannot be rejected and would have to be dropped. This triggers the retransmission of the request plus the response, leading to even more load. In addition, the request associated with a response has already been processed and dropping the response will waste the efforts that have been spent on the request. Most importantly, rejecting a request effectively also removes the request and the response. If no requests are passed along, there will be no responses coming back in return. Overload control does not change the retransmission behavior of SIP. Retransmissions are triggered using procedures defined in RFC 3261 [RFC3261] and are not subject to throttling. 13. Operational Considerations In addition to the design considerations discussed above, implementations of a SIP overload control mechanism need to take the following operational aspects into consideration. These aspects, while important, are out of scope for this document and are left for further discussion in other documents. Selection of feedback type: A SIP overload control mechanism can support one or multiple types of explicit overload control feedback. Using a single type of feedback (e.g., loss-based feedback) has the advantage of simplifying the protocol and implementations. Supporting multiple types of feedback (e.g., loss- and rate-based feedback) provides more flexibility; however, it requires a way to select the feedback type used between two servers.
Event reporting: Overload is a serious condition for any network of SIP servers, even if it is handled properly by an overload control mechanism. Overload events should therefore be reported by a SIP server, e.g., through a logging or management interface. 14. Security Considerations This document presents an overview of several overload control feedback mechanisms. These mechanisms and design consideration are presented as input to other documents that will specify a particular feedback mechanism. Specific security measures pertinent to a particular overload feedback mechanism will be discussed in the context of a document specifying that security mechanism. However, there are common security considerations that must be taken into account regardless of the choice of a final mechanism. First, the rate-based mechanism surveyed in Section 9.1 allocates a fixed portion of the total inbound traffic of a server to each of its upstream neighbors. Consequently, an attacker can introduce a new upstream server for a short duration, causing the overloaded server to lower the proportional traffic rate to all other existing servers. Introducing many such short-lived servers will cause the aggregate rate arriving at the overloaded server to decrease substantially, thereby affecting a reduction in the service offered by the server under attack and leading to a denial-of-service attack [RFC4732]. The same problem exists in the windows-based mechanism discussed in Section 9.3; however, because of the window acknowledgments sent by the overloaded server, the effect is not as drastic (an attacker will have to expend resources by constantly sending traffic to keep the receiver window full). All mechanisms assume that the upstream neighbors of an overloaded server follow the feedback received. In the rate- and window-based mechanisms, a server can directly verify if upstream neighbors follow the requested policies. As the loss-based mechanism described in Section 9.2 requires upstream neighbors to reduce traffic by a fraction and the current offered load in the upstream neighbor is unknown, a server cannot directly verify the compliance of upstream neighbors, except when traffic reduction is set to 100%. In this case, a server has to rely on heuristics to identify upstream neighbors that try to gain an advantage by not reducing load or not reducing it at the requested loss-rate. A policing mechanism can be used to throttle or block traffic from unfair or malicious upstream neighbors. Barring such a widespread policing mechanism, the communication link between the upstream neighbors and the overloaded server should be such that the identity of both the servers at the end of each link can be established and logged. The use of Transport
Layer Security (TLS) and mutual authentication of upstream neighbors [RFC3261] [RFC5922] can be used for this purpose. If an attacker controls a server, he or she may maliciously advertise overload feedback to all of the neighbors of the server, even if the server is not experiencing overload. This will have the effect of forcing all of the upstream neighbors to reject or queue messages arriving to them and destined for the apparently overloaded server (this, in essence, is diminishing the serving capacity of the upstream neighbors since they now have to deal with their normal traffic in addition to rejecting or quarantining the traffic destined to the overloaded server). All mechanisms allow the attacker to advertise a capacity of 0, effectively disabling all traffic destined to the server pretending to be in overload and forcing all the upstream neighbors to expend resources dealing with this condition. As before, a remedy for this is to use a communication link such that the identity of the servers at both ends of the link is established and logged. The use of TLS and mutual authentication of neighbors [RFC3261] [RFC5922] can be used for this purpose. If an attacker controls several servers of a load-balanced cluster, he or she may maliciously advertise overload feedback from these servers to all senders. Senders with the policy to redirect traffic that cannot be processed by an overloaded server will start to redirect this traffic to the servers that have not reported overload. This attack can be used to create a denial-of-service attack on these servers. If these servers are compromised, the attack can be used to increase the amount of traffic that is passed through the compromised servers. This attack is ineffective if servers reject traffic based on overload feedback instead of redirecting it. 15. Informative References [Abdelal] Abdelal, A. and W. Matragi, "Signal-Based Overload Control for SIP Servers", 7th Annual IEEE Consumer Communications and Networking Conference (CCNC-10), Las Vegas, Nevada, USA, January 2010. [Hilt] Hilt, V. and I. Widjaja, "Controlling overload in networks of SIP servers", IEEE International Conference on Network Protocols (ICNP'08), Orlando, Florida, October 2008.
[Noel] Noel, E. and C. Johnson, "Novel Overload Controls for SIP Networks", International Teletraffic Congress (ITC 21), Paris, France, September 2009. [RFC0793] Postel, J., "Transmission Control Protocol", STD 7, RFC 793, September 1981. [RFC3261] Rosenberg, J., Schulzrinne, H., Camarillo, G., Johnston, A., Peterson, J., Sparks, R., Handley, M., and E. Schooler, "SIP: Session Initiation Protocol", RFC 3261, June 2002. [RFC4412] Schulzrinne, H. and J. Polk, "Communications Resource Priority for the Session Initiation Protocol (SIP)", RFC 4412, February 2006. [RFC4732] Handley, M., Rescorla, E., and IAB, "Internet Denial-of- Service Considerations", RFC 4732, December 2006. [RFC5390] Rosenberg, J., "Requirements for Management of Overload in the Session Initiation Protocol", RFC 5390, December 2008. [RFC5922] Gurbani, V., Lawrence, S., and A. Jeffrey, "Domain Certificates in the Session Initiation Protocol (SIP)", RFC 5922, June 2010. [Shen] Shen, C., Schulzrinne, H., and E. Nahum, "Session Initiation Protocol (SIP) Server Overload Control: Design and Evaluation, Principles", Systems and Applications of IP Telecommunications (IPTComm'08), Heidelberg, Germany, July 2008.
Appendix A. Contributors Many thanks for the contributions, comments, and feedback on this document to: Mary Barnes (Nortel), Janet Gunn (CSC), Carolyn Johnson (AT&T Labs), Paul Kyzivat (Cisco), Daryl Malas (CableLabs), Tom Phelan (Sonus Networks), Jonathan Rosenberg (Cisco), Henning Schulzrinne (Columbia University), Robert Sparks (Tekelec), Nick Stewart (British Telecommunications plc), Rich Terpstra (Level 3), Fangzhe Chang (Bell Labs/Alcatel-Lucent). Authors' Addresses Volker Hilt Bell Labs/Alcatel-Lucent 791 Holmdel-Keyport Rd Holmdel, NJ 07733 USA EMail: firstname.lastname@example.org Eric Noel AT&T Labs EMail: email@example.com Charles Shen Columbia University EMail: firstname.lastname@example.org Ahmed Abdelal Sonus Networks EMail: email@example.com