All timeout values are in the format hh:mm:ss , with a maximum duration of 1193:0:0 in most cases. Use the no timeout command to reset all timeouts to their default values. If you want to simply reset one timer to the default, enter the timeout command for that setting with the default value.

Use 0 for the value to disable a timer.

You can configure the following global timeouts.

• timeout conn hh:mm:ss —The idle time after which a connection closes, between 0:5:0 and 1193:0:0. The default is 1 hour (1:0:0).

• timeout half-closed hh:mm:ss —The idle time until a TCP half-closed connection closes. A connection is considered half-closed if both the FIN and FIN-ACK have been seen. If only the FIN has been seen, the regular conn timeout applies. The minimum is 30 seconds. The default is 10 minutes.

• timeout udp hh:mm:ss —The idle time until a UDP connection closes. This duration must be at least 1 minute. The default is 2 minutes.

• timeout icmp hh:mm:ss —The idle time for ICMP, between 0:0:2 and 1193:0:0. The default is 2 seconds (0:0:2).

• timeout sunrpc hh:mm:ss —The idle time until a SunRPC slot is freed. This duration must be at least 1 minute. The default is 10 minutes.

• timeout H323 hh:mm:ss —The idle time after which H.245 (TCP) and H.323 (UDP) media connections close, between 0:0:0 and 1193:0:0. The default is 5 minutes (0:5:0). Because the same connection flag is set on both H.245 and H.323 media connections, the H.245 (TCP) connection shares the idle timeout with the H.323 (RTP and RTCP) media connection.

• timeout h225 hh:mm:ss —The idle time until an H.225 signaling connection closes. The H.225 default timeout is 1 hour (1:0:0). To close a connection immediately after all calls are cleared, a value of 1 second (0:0:1) is recommended.

• timeout mgcp hh:mm:ss —The idle time after which an MGCP media connection is removed, between 0:0:0 and 1193:0:0. The default is 5 minutes (0:5:0)

• timeout mgcp-pat hh:mm:ss —The absolute interval after which an MGCP PAT translation is removed, between 0:0:0 and 1193:0:0. The default is 5 minutes (0:5:0). The minimum time is 30 seconds.

• timeout sip hh:mm:ss —The idle time until a SIP signaling port connection closes, between 0:5:0 and 1193:0:0. The default is 30 minutes (0:30:0).

• timeout sip_media hh:mm:ss —The idle time until an SIP media port connection closes. This duration must be at least 1 minute. The default is 2 minutes. The SIP media timer is used for SIP RTP/RTCP with SIP UDP media packets, instead of the UDP inactivity timeout.

• timeout sip-provisional-media hh:mm:ss —The timeout value for SIP provisional media connections, between 0:1:0 and 0:30:0. The default is 2 minutes.

• timeout sip-invite hh:mm:ss —The idle time after which pinholes for PROVISIONAL responses and media xlates will be closed, between 0:1:0 and 00:30:0. The default is 3 minutes (0:3:0).

• timeout sip-disconnect hh:mm:ss —The idle time after which a SIP session is deleted if the 200 OK is not received for a CANCEL or a BYE message, between 0:0:1 and 00:10:0. The default is 2 minutes (0:2:0).

• timeout uath hh:mm:ss {absolute | inactivity} —The duration before the authentication and authorization cache times out and the user has to reauthenticate the next connection, between 0:0:0 and 1193:0:0. The default is 5 minutes (0:5:0). The default timer is absolute ; you can set the timeout to occur after a period of inactivity by entering the inactivity keyword. The uauth duration must be shorter than the xlate duration. Set to 0 to disable caching. Do not use 0 if passive FTP is used for the connection or if the virtual http command is used for web authentication.

• timeout xlate hh:mm:ss —The idle time until a translation slot is freed. This duration must be at least 1 minute. The default is 3 hours.

• timeout pat-xlate hh:mm:ss —The idle time until a PAT translation slot is freed, between 0:0:30 and 0:5:0. The default is 30 seconds. You may want to increase the timeout if upstream routers reject new connections using a freed PAT port because the previous connection might still be open on the upstream device.

• timeout tcp-proxy-reassembly hh:mm:ss —The idle timeout after which buffered packets waiting for reassembly are dropped, between 0:0:10 and 1193:0:0. The default is 1 minute (0:1:0).

• timeout floating-conn hh:mm:ss —When multiple routes exist to a network with different metrics, the ASA uses the one with the best metric at the time of connection creation. If a better route becomes available, then this timeout lets connections be closed so a connection can be reestablished to use the better route. The default is 0 (the connection never times out). To make it possible to use better routes, set the timeout to a value between 0:0:30 and 1193:0:0.

• timeout conn-holddown hh:mm:ss —How long the system should maintain a connection when the route used by the connection no longer exists or is inactive. If the route does not become active within this holddown period, the connection is freed. The purpose of the connection holddown timer is to reduce the effect of route flapping, where routes might come up and go down quickly. You can reduce the holddown timer to make route convergence happen more quickly. The default is 15 seconds, the range is 00:00:00 to 00:00:15.

class-map name
match parameter

policy-map name
class name

In the default configuration, the global_policy policy map is assigned globally to all interfaces. If you want to edit the global_policy, enter global_policy as the policy name. For the class map, specify the class you created earlier in this procedure.

• set connection embryonic-conn-max n —The maximum number of simultaneous embryonic TCP connections allowed, from 0 and 2000000. The default is 0, which allows unlimited connections.

• set connection per-client-embryonic-max n —The maximum number of simultaneous embryonic TCP connections allowed per client, from 0 and 2000000. The default is 0, which allows unlimited connections.

service-policy policymap_name {global | interface interface_name}

threat-detection statistics tcp-intercept [rate-interval minutes] [burst-rate attacks_per_sec] [average-rate attacks_per_sec]

Where:

• rate-interval minutes sets the size of the history monitoring window, between 1 and 1440 minutes. The default is 30 minutes. During this interval, the ASA samples the number of attacks 30 times.

• burst-rate attacks_per_sec sets the threshold for syslog message generation, between 25 and 2147483647. The default is 400 per second. When the burst rate is exceeded, syslog message 733104 is generated.

• average-rate attacks_per_sec sets the average rate threshold for syslog message generation, between 25 and 2147483647. The default is 200 per second. When the average rate is exceeded, syslog message 733105 is generated.

• show threat-detection statistics top tcp-intercept [all | detail]—View the top 10 protected servers under attack. The all keyword shows the history data of all the traced servers. The detail keyword shows history sampling data. The ASA samples the number of attacks 30 times during the rate interval, so for the default 30 minute period, statistics are collected every 60 seconds.

• clear threat-detection statistics tcp-intercept—Erases TCP Intercept statistics.

• check-retransmission—Prevent inconsistent TCP retransmission. This command is disabled by default.

• checksum-verification—Verify the TCP checksum, dropping packets that fail verification. This command is disabled by default.

• exceed-mss {allow | drop}—Allow or drop packets whose data length exceeds the TCP maximum segment size. The default is to allow the packets.

• invalid-ack {allow | drop}—Allow or drop packets with an invalid ACK. The default is to drop the packet, with the exception of WAAS connections, where they are allowed. You might see invalid ACKs in the following instances:

  • In the TCP connection SYN-ACK-received status, if the ACK number of a received TCP packet is not exactly the same as the sequence number of the next TCP packet sending out, it is an invalid ACK.

  • Whenever the ACK number of a received TCP packet is greater than the sequence number of the next TCP packet sending out, it is an invalid ACK.

• queue-limit pkt_num [timeout seconds]—Set the maximum number of out-of-order packets that can be buffered and put in order for a TCP connection, between 1 and 250 packets. The default is 0, which means this setting is disabled and the default system queue limit is used depending on the type of traffic:

  • Connections for application inspection (the inspect command), and TCP check-retransmission (the TCP map check-retransmission command) have a queue limit of 3 packets. If the ASA receives a TCP packet with a different window size, then the queue limit is dynamically changed to match the advertised setting.

  • For other TCP connections, out-of-order packets are passed through untouched.

  If you set the queue-limit command to be 1 or above, then the number of out-of-order packets allowed for all TCP traffic matches this setting. For example, for application inspection and TCP check-retransmission traffic, any advertised settings from TCP packets are ignored in favor of the queue-limit setting. For other TCP traffic, out-of-order packets are now buffered and put in order instead of passed through untouched.

  The timeout seconds argument sets the maximum amount of time that out-of-order packets can remain in the buffer, between 1 and 20 seconds; if they are not put in order and passed on within the timeout period, then they are dropped. The default is 4 seconds. You cannot change the timeout for any traffic if the pkt_num argument is set to 0; you need to set the limit to be 1 or above for the timeout keyword to take effect.

• reserved-bits {allow | clear | drop}—Set the action for reserved bits in the TCP header. You can allow the packet (without changing the bits), clear the bits and allow the packet, or drop the packet.

• seq-past-window {allow | drop}—Set the action for packets that have past-window sequence numbers, namely the sequence number of a received TCP packet is greater than the right edge of the TCP receiving window. You can allow the packets only if the queue-limit command is set to 0 (disabled). The default is to drop the packets.

• synack-data {allow | drop}—Allow or drop TCP SYNACK packets that contain data. The default is to drop the packet.

• syn-data {allow | drop}—Allow or drop SYN packets with data. The default is to allow the packet.

• tcp-options {selective-ack | timestamp | window-scale | range lower upper} {allow | clear}—Set the action for packets with TCP options. Three options are named: selective-ack (selective acknowledgment mechanism), timestamp, and window-scale (window scale mechanism). For other options, you specify them by number on the range keyword, where the range limits are 6-7, 9-255. You can enter the command multiple times in a map to define your complete policy. Note that if a TCP connection is inspected, all options are cleared except the MSS and selective-acknowledgment (SACK) options, regardless of your configuration.

  You can allow the packet (without changing the options), clear the options and allow the packet, or drop the packet. The default for the three named options is to allow them; the default for all other options is to clear them. Note that clearing the timestamp option disables PAWS and RTT.

• ttl-evasion-protection—Have the maximum TTL for a connection be determined by the TTL in the initial packet. The TTL for subsequent packets can decrease, but it cannot increase. The system will reset the TTL to the lowest previously-seen TTL for that connection. This protects against TTL evasion attacks. TTL evasion protection is enabled by default, so you would only need to enter the no form of this command.

  For example, an attacker can send a packet that passes policy with a very short TTL. When the TTL goes to zero, a router between the ASA and the endpoint drops the packet. It is at this point that the attacker can send a malicious packet with a long TTL that appears to the ASA to be a retransmission and is passed. To the endpoint host, however, it is the first packet that has been received by the attacker. In this case, an attacker is able to succeed without security preventing the attack.

• urgent-flag {allow | clear}—Set the action for packets with the URG flag. You can allow the packet, or clear the flag and allow the packet. The default is to clear the flag.

  The URG flag is used to indicate that the packet contains information that is of higher priority than other data within the stream. The TCP RFC is vague about the exact interpretation of the URG flag, therefore end systems handle urgent offsets in different ways, which may make the end system vulnerable to attacks.

• window-variation {allow | drop}—Allow or drop a connection that has changed its window size unexpectedly. The default is to allow the connection.

  The window size mechanism allows TCP to advertise a large window and to subsequently advertise a much smaller window without having accepted too much data. From the TCP specification, “shrinking the window” is strongly discouraged. When this condition is detected, the connection can be dropped.

class-map name
match parameter

For information on matching statements, see Create a Layer 3/4 Class Map for Through Traffic.

policy-map name
class name

In the default configuration, the global_policy policy map is assigned globally to all interfaces. If you want to edit the global_policy, enter global_policy as the policy name. For the class map, specify the class you created earlier in this procedure.

By default, there are no connection limits. If you implement limits, the system must start tracking them, which can increase CPU and memory usage and result in operational problems for systems under heavy load, especially in a cluster.

• set connection conn-max n—(TCP, UDP.) The maximum number of simultaneous connections that are allowed, between 0 and 2000000, for the entire class. The default is 0, which allows unlimited connections. For TCP connections, this applies to established connections only.

  • If two servers are configured to allow simultaneous connections, the connection limit is applied to each configured server separately.

  • Because the limit is applied to a class, one attack host can consume all the connections and leave none for the rest of the hosts that are matched to the class.

• set connection per-client-max n—(TCP, UDP.) The maximum number of simultaneous connections allowed per client, between 0 and 2000000. The default is 0, which allows unlimited connections. This argument restricts the maximum number of simultaneous connections that are allowed for each host that is matched to the class. For TCP connections, this includes established, half-open, and half-closed connections.

• set connection embryonic-conn-max n—The maximum number of simultaneous embryonic TCP connections allowed, between 0 and 2000000. The default is 0, which allows unlimited connections. By setting a non-zero limit, you enable TCP Intercept, which protects inside systems from a DoS attack perpetrated by flooding an interface with TCP SYN packets. Also set the per-client options to protect against SYN flooding.

• set connection per-client-embryonic-max n—The maximum number of simultaneous embryonic TCP connections allowed per client, between 0 and 2000000. The default is 0, which allows unlimited connections.

• set connection random-sequence-number {enable | disable}—Whether to enable or disable TCP sequence number randomization. Randomization is enabled by default.

The defaults described below assume you have not changed the global defaults for these behaviors using the timeout command; the global defaults override the ones described here. Enter 0 to disable the timer, so that a connection never times out.

• set connection timeout embryonic hh:mm:ss—The timeout period until a TCP embryonic (half-open) connection is closed, between 0:0:5 and 1193:00:00. The default is 0:0:30.

• set connection timeout idle hh:mm:ss [reset]—The idle timeout period after which an established connection of any protocol closes, between 0:0:1 and 1193:0:0. The default is 1:0:0. For TCP traffic, the reset keyword sends a reset to TCP endpoints when the connection times out.

  The default udp idle timeout is 2 minutes. The default icmp idle timeout is 2 seconds. The default esp and ha idle timeout is 30 seconds. For all other protocols, the default idle timeout is 2 minutes.

• set connection timeout half-closed hh:mm:ss—The idle timeout period until a half-closed connection is closed, between 0:5:0 (for 9.1(1) and earlier) or 0:0:30 (for 9.1(2) and later) and 1193:0:0. The default is 0:10:0. Half-closed connections are not affected by DCD. Also, the ASA does not send a reset when taking down half-closed connections.

• set connection timeout dcd [retry-interval [max_retries]]—Enable Dead Connection Detection (DCD). Before expiring an idle connection, the ASA probes the end hosts to determine if the connection is valid. If both hosts respond, the connection is preserved, otherwise the connection is freed. When operating in transparent firewall mode, you must configure static routes for the endpoints. You cannot use DCD in a cluster.

  The retry-interval sets the time duration in hh:mm:ss format to wait after each unresponsive DCD probe before sending another probe, between 0:0:1 and 24:0:0. The default is 0:0:15. The max-retries sets the number of consecutive failed retries for DCD before declaring the connection as dead. The minimum value is 1 and the maximum value is 255. The default is 5.

  For systems that are operating in a high-availability configuration, we recommend that you do not set the interval to less than one minute (0:1:0). If the connection needs to be moved between systems, the changes required take longer than 30 seconds, and the connection might be deleted before the change is accomplished.

This command, along with the icmp unreachable command, is required to allow a traceroute through the ASA that shows the ASA as one of the hops.

Advanced options are special purpose configurations that are not needed under normal circumstances. You configure them with the set connection advanced-options command.

• set connection advanced-options tcp_map_name —Customize TCP Normalizer behavior by applying a TCP map. For detailed information, see Customize Abnormal TCP Packet Handling (TCP Maps, TCP Normalizer).

• set connection advanced-options tcp-state-bypass —Implement TCP State Bypass. For detailed information, see Bypass TCP State Checks for Asynchronous Routing (TCP State Bypass).

service-policy policymap_name {global | interface interface_name}