The minimum period of time (in seconds) that the event broker service will tolerate inactivity on the client connection. This keepalive value is also enforced for MQTT and SMF (Solace Message Format) connections. The keepalive timeout value is calculated based on the client provided timeout interval (3 x the keepalive interval for SMF, 1.5 x the keepalive interval for MQTT). The default is 30 and valid ranges are 3–3600.

The number of seconds a Web client has to send a request before its session times out and be terminated for being inactive. The default value is 30 seconds.

The maximum transport payload size (in bytes) before fragmentation occurs for clients using the client profile. The size of the header is not included.

Solace Message Format (SMF) messages that are sent to a consuming Web client are contained within a Web transport message that the event broker sends in its HTTP response to that client. Each Web transport message that is sent can contain multiple SMF messages or partial SMF messages.
The maximum Web payload value sets the maximum number of bytes allowed in a single Web transport message (not including the header). This value determines the number of SMF messages that can be sent in a single HTTP response and the size of the Web transport message sent in the HTTP response. The value range is 300 to 10000000, in bytes. The default is 1000000 bytes.

SMF messages that are sent to a consuming Web client are contained within a Web transport message that the event broker sends in its HTTP response to that client. Each Web transport message that is sent can contain multiple SMF messages or partial SMF messages.

The maximum Web payload value sets the maximum number of bytes allowed in a single Web transport message (not including the header). This value determines the number of SMF messages that can be sent in a single HTTP response and the size of the Web transport message sent in the HTTP response. Note that large SMF messages can be fragmented across Web transport messages to respect the value set for the maximum possible Web payload.

The maximum permitted number of simultaneous Web transport client connections to the event broker service that can be made using the same client username account. The default is the maximum value supported by the platform.

The maximum permitted number of simultaneous Solace Message Format (SMF) client connections to the event broker that can be made using the same client username account. The default is the maximum value supported by the platform.

The maximum permitted number of simultaneous Web transport client connections to the event broker service that can be made using the same client username account. The default is the maximum value supported by the platform.

The maximum number of durable and non-durable queues and topic endpoints that can be owned by the clients using the same client username within a client profile. The default is 1000.

The maximum number of egress flows (that is, Guaranteed Message client receive flows or consumer flows) that can be created by a single client associated with this client profile. The default is 1000.

The maximum number of ingress flows (that is, Guaranteed Message client publish flows) that can be created by a single client associated with this client profile. The default is 1000.

The maximum number of subscriptions for a single client in the client profile. When you set this option, consider the total maximum number of permitted topic subscriptions and the total maximum number of client connections for the type of event broker that is used. That is, to ensure reliable system performance, you must find the right balance for your network, while staying within the system limits mentioned. The balance is generally between allowing the creation of many client applications and allowing each client to add a large number of topic subscriptions. The default varies by platform.

The maximum number of simultaneous transacted sessions and/or XA sessions allowed for a single client associated with the client profile. The default is 10.

The total maximum number of simultaneous transactions (both local transactions and transactions within the XA transaction branches) allowed for a single client associated with the client profile. The default varies by platform.

The egress queue maximum depth for Guaranteed Messages that represents the number of work units for the client priority queues. The valid range is 2 to 262144. The default is 20000.

The minimum number of messages that must be on the Guaranteed Message queue before the queue’s depth is checked against the maximum depth setting (thereby allowing the queue to absorb a burst of large messages that exceeds the number of allowed work units). A valid range is 0 to 262144 with the default of 255. The value of 255 is recommended for memory usage optimized configurations, such as message applications; a value of 66000 is for WAN optimized configurations.

The egress queue maximum depth for Direct Messages 1 [Class of Service (COS) 1] that represents the number of work units for the client priority queues. The valid range is 2 to 262144. The default is 20000.

The minimum number of messages that must be on the Direct 1 (COS 1) queue before the queue’s depth is checked against the maximum depth setting (thereby allowing the queue to absorb a burst of large messages that exceeds the number of allowed work units). A valid range is 0 to 262144 with the default of 4.

The egress queue maximum depth for Direct Messages 2 (COS 2) that represents the number of work units for the client priority queues. The valid range is 2 to 262144. The default is 20000.

The minimum number of messages that must be on the Direct 2 (COS 2) queue before the queue’s depth is checked against the maximum depth setting (thereby allowing the queue to absorb a burst of large messages that exceeds the number of allowed work units). A valid range is 0 to 262144 with the default of 4.

The egress queue maximum depth for Direct Messages 3 [Class of Service (COS) 3] that represents the number of work units for the client priority queues. The valid range is 2 to 262144. The default is 20000.

The minimum number of messages that must be on the Direct 3 (COS 3) queue before the queue’s depth is checked against the maximum depth setting (thereby allowing the queue to absorb a burst of large messages that exceeds the number of allowed work units). A valid range is 0 to 262144 with the default of 4.

The egress queue maximum depth for Control 1 that represents the number of work units for the client priority queues. The valid range is 2 to 262144. The default is 20000.

The minimum number of messages that must be on the Direct 1 (COS 1) queue before the queue’s depth is checked against the maximum depth setting (thereby allowing the queue to absorb a burst of large messages that exceeds the number of allowed work units). A valid range is 0 to 262144 with the default of 4.

The TCP initial congestion window size is the number of segments that TCP sends before waiting for an acknowledgment from the peer. The TCP initial congestion window size is used when starting up a TCP connection or on recovery from idle (that is, no traffic). Larger values of the initial window allows a connection to come up to speed more quickly. For further details, refer to RFC 2581.
Changing the TCP initial congestion window size from its default of 2 results in non-compliance with RFC 2581. Further, if this setting is set to a value too high, it may cause congestion in the network. Contact support before you attempt to change this TCP parameter.

The maximum number of keepalive probes (from 2 to 5 ) that TCP should send before dropping the connection. The default is 5.

The time (from 3 to 120 seconds) a connection must remain idle before TCP begins sending keepalive probes. The default is 3.

The time (from 1 to 30 seconds) to set as the interval between individual keepalive probes. The default is 1.

The TCP maximum segment size (MSS) used for client to the event broker service. The default is 1460.

The TCP window size between the event broker service and the client.
If the maximum window size is set to less than the bandwidth-delay product, then the TCP connection operates below its maximum potential throughput. If the maximum window is set to less than about twice the bandwidth-delay product, then occasional packet loss causes the TCP connection to operate below its maximum potential throughput as it handles the missing acknowledgments and retransmissions. The default is 256.

Alternately, if the TCP maximum window size is set too large, in the presence of a high offered load, TCP gradually increases its congestion window size until either the congestion window size reaches the maximum window size, or packet loss occurs in the network.

Initially, when the TCP congestion window size is small, the physical bandwidth-delay of the network acts as a memory buffer for packets in flight. But as the congestion window crosses the bandwidth-delay product, the buffering of in-flight packets moves to queues in event broker services and other equipment throughout the network. As the TCP congestion window continues to increase in size, these various equipment queues overflow, causing packet loss and TCP backoff.

The amount of time to delay the delivery of messages to a client after the initial message has been delivered. You can specify a value from 0-6000 milliseconds. The default is 0.

The maximum number of topics the event broker service that can track for performing the eliding function on each client connection. You can specify a value from 1-320000. The default is 256.