Advanced configurations

The Advanced configurations page provides access to the global settings for the Layer 4 and Layer 7 load balancing services. These settings apply to all Services of the corresponding type and rarely need to be changed from their defaults.

The settings are split across two tabs - Layer 4 and Layer 7.

To configure the global settings:

Using the WebUI, navigate to Cluster Configuration > Advanced configurations.
Select the Layer 4 or Layer 7 tab as required.
Configure the settings as described in the sections below.
Click Submit.

Layer 4

The Layer 4 tab configures the global settings for the Layer 4 (LVS/ldirectord) services.

Setting	Description
Check interval	The Layer 4 (ldirectord) health check interval, in seconds. If this value is too low, you may experience unexpected Real Server downtime.
Check timeout	The Layer 4 (ldirectord) health check timeout, in seconds. If this value is too low, you may induce unexpected Real Server downtime.
Negotiate timeout	The Layer 4 (ldirectord) negotiate health check timeout, in seconds. Negotiate checks may take longer to process as they involve more server-side processing than a simple TCP socket connect check. If this value is too low, you may induce unexpected Real Server downtime.
TCP FIN timeout	The time to remember a TCP session for after seeing a FIN packet.
UDP timeout	The time to remember a session for after seeing a UDP packet. The timeout is reset on every UDP packet received.
Failure count	The Layer 4 (ldirectord) number of times a check has to fail before taking a server offline. The time to detect a failure and take a server down will be (check interval + check timeout) x failure count.
Quiescent	Determines what happens when a Real Server fails its health check. When enabled (checked), the Real Server is not removed from the load balancing table on a health check failure - instead its weight is set to 0. Persistent connections continue to be routed to the failed server, but no new connections are accepted. When disabled (unchecked), the server is completely removed from the load balancing table on a health check failure; persistent connections are broken and sent to a different Real Server.

Setting

Description

Check interval

The Layer 4 (ldirectord) health check interval, in seconds. If this value is too low, you may experience unexpected Real Server downtime.

Check timeout

The Layer 4 (ldirectord) health check timeout, in seconds. If this value is too low, you may induce unexpected Real Server downtime.

Negotiate timeout

The Layer 4 (ldirectord) negotiate health check timeout, in seconds. Negotiate checks may take longer to process as they involve more server-side processing than a simple TCP socket connect check. If this value is too low, you may induce unexpected Real Server downtime.

TCP FIN timeout

The time to remember a TCP session for after seeing a FIN packet.

UDP timeout

The time to remember a session for after seeing a UDP packet. The timeout is reset on every UDP packet received.

Failure count

The Layer 4 (ldirectord) number of times a check has to fail before taking a server offline. The time to detect a failure and take a server down will be (check interval + check timeout) x failure count.

Quiescent

Determines what happens when a Real Server fails its health check. When enabled (checked), the Real Server is not removed from the load balancing table on a health check failure - instead its weight is set to 0. Persistent connections continue to be routed to the failed server, but no new connections are accepted. When disabled (unchecked), the server is completely removed from the load balancing table on a health check failure; persistent connections are broken and sent to a different Real Server.

Quiescent only applies to health checks - it has no effect on taking Real Servers offline in the System Overview. To manually force a Real Server to be removed from the table, disable Quiescent and arrange for the server to fail its health check. This can be done, for example, by shutting down the daemon or service, changing the negotiate check value, or shutting down the server.

Layer 7

The Layer 7 tab configures the global settings for the Layer 7 (HAProxy) services. The settings are grouped into a number of sections as described below.

Logging

Setting	Description
Log level	Sets the logging level for the Layer 7 (HAProxy) service. The options are Off, Health Check, Errors and Debug. Logs are written to /var/log/haproxy.log.

Setting

Description

Log level

Sets the logging level for the Layer 7 (HAProxy) service. The options are Off, Health Check, Errors and Debug. Logs are written to /var/log/haproxy.log.

By default, the Layer 7 (HAProxy) service logs service state, restarts, reloads and health checks only. Setting the log level to Errors or Debug logs errors or individual connections respectively. Logging every connection on a busy Service is likely to rapidly exhaust the available log space.

Connections

Setting	Description
Connection timeout	The maximum time HAProxy waits for a connection attempt to a Real Server to succeed. Enter a value and select the unit (seconds, minutes, hours or days). This setting should not normally require changing.
Client timeout	The global client inactivity timeout. Enter a value and select the unit. This setting should not normally require changing.
Real server response timeout	The global Real Server response timeout. Enter a value and select the unit. This setting should not normally require changing.
Maximum connections	The per-process maximum number of concurrent connections. This setting should not require changing unless you are running a high volume site.

Setting

Description

Connection timeout

The maximum time HAProxy waits for a connection attempt to a Real Server to succeed. Enter a value and select the unit (seconds, minutes, hours or days). This setting should not normally require changing.

Client timeout

The global client inactivity timeout. Enter a value and select the unit. This setting should not normally require changing.

Real server response timeout

The global Real Server response timeout. Enter a value and select the unit. This setting should not normally require changing.

Maximum connections

The per-process maximum number of concurrent connections. This setting should not require changing unless you are running a high volume site.

Email alerts

Setting	Description
Email alert to	The address that email alerts are sent to.
Email alert from	The "from" address used for email alerts.
Email server address	The email server address, entered as either an IP address or FQDN.
Email server port	The TCP port of the email server. The default is 25.

Setting

Description

Email alert to

The address that email alerts are sent to.

Email alert from

The "from" address used for email alerts.

Email server address

The email server address, entered as either an IP address or FQDN.

Email server port

The TCP port of the email server. The default is 25.

Statistics page

Enable the Enable stats page checkbox to activate the HAProxy web-based statistics report. When enabled, the following additional settings are displayed:

Setting	Description
Password	The password used to access the HAProxy statistics report.
Service IP	The listening IP for the HAProxy statistics report, selected from the configured Service IPs.
Port	The listening port for the HAProxy statistics report. The default is 7777.
Advanced stats	When enabled, additional actions (such as enabling/disabling servers) are available on the HAProxy stats page.
Enable SSL	When enabled, the statistics page is served over SSL/TLS.
SSL min version	The minimum SSL/TLS version accepted for the statistics page, e.g. TLSv1.1.
SSL max version	The maximum SSL/TLS version accepted for the statistics page, e.g. TLSv1.3.
SSL certificate	The certificate used to secure the statistics page, selected from the configured SSL Certificates.

Setting

Description

Password

The password used to access the HAProxy statistics report.

Service IP

The listening IP for the HAProxy statistics report, selected from the configured Service IPs.

Port

The listening port for the HAProxy statistics report. The default is 7777.

Advanced stats

When enabled, additional actions (such as enabling/disabling servers) are available on the HAProxy stats page.

Enable SSL

When enabled, the statistics page is served over SSL/TLS.

SSL min version

The minimum SSL/TLS version accepted for the statistics page, e.g. TLSv1.1.

SSL max version

The maximum SSL/TLS version accepted for the statistics page, e.g. TLSv1.3.

SSL certificate

The certificate used to secure the statistics page, selected from the configured SSL Certificates.

Buffer size

Setting	Description
Request buffer length	The request buffer length, in bytes. The default value is 16384 bytes.
Header buffer length	The header buffer length, in bytes. The header buffer is a section of the request buffer, reserved for the addition and rewriting of request headers. The default value is 1024 bytes. Most applications only require a small header buffer, as few headers are added or rewritten.

Setting

Description

Request buffer length

The request buffer length, in bytes. The default value is 16384 bytes.

Header buffer length

The header buffer length, in bytes. The header buffer is a section of the request buffer, reserved for the addition and rewriting of request headers. The default value is 1024 bytes. Most applications only require a small header buffer, as few headers are added or rewritten.

Changing the request buffer length affects the performance of HAProxy - do not make changes unless you know exactly what you are doing. Lower values allow more sessions to coexist in the same amount of RAM, and higher values allow some applications with very large cookies to work. It is strongly recommended not to change this from the default value, as very low values will break some services such as the statistics page, and values larger than the default will increase memory usage, possibly causing the system to run out of memory. Consider reducing the Maximum connections value if the request buffer is increased.

Persistence table replication

When configured, HAProxy’s persistence tables are replicated to the peer appliance in an HA pair.

Setting	Description
Persistence replication service IP	The Service IP used for persistence table replication.
Persistence replication port	The TCP port used for persistence table replication. The default is 7778.

Setting

Description

Persistence replication service IP

The Service IP used for persistence table replication.

Persistence replication port

The TCP port used for persistence table replication. The default is 7778.

Multi threading

Setting	Description
Enable Max threads	Enables or disables multi-threading for the Layer 7 (HAProxy) service. Multi-threading can improve performance if connection limits are being reached.
Number of threads	The maximum number of worker threads started per HAProxy process.

Setting

Description

Enable Max threads

Enables or disables multi-threading for the Layer 7 (HAProxy) service. Multi-threading can improve performance if connection limits are being reached.

Number of threads

The maximum number of worker threads started per HAProxy process.

Multi-threading is enabled by default and the number of threads is set automatically based on the number of detected CPUs / vCPUs.