Advanced Setup

From PFSenseDocs
Jump to: navigation, search

The advanced settings available under System > Advanced do not normally need adjusting on a typical setup. There are there for additional tweaking or for those who need the functionality given.

The options available are all described in detail on their individual pages, but are split into separate tabs.

Admin Access

Contains settings for the WebGUI such as the protocol (HTTP/HTTPS) and port, security settings for management, SSH daemon settings, Serial console settings, and console options.

Firewall/NAT

Contains settings that tweak the behavior of the firewall, such as fragmentation, optimization algorithms, and state table settings.

Also has the settings for NAT Reflection.

The FreeBSD documentation is a bit unspecific when it comes to the PF optimization algorithms. Here are the actual values which are chosen for each optimization algorithm (taken from the source code, first line is raw value, second line is human readable form):


Normal High Latency Conservative Aggressive
tcp.first
First TCP packet
60
1min
180
3min
3600
60min
30
30sec
tcp.opening
No response yet
30
30sec
35
35sec
900
15min
5
5sec
tcp.established
Established
86400
24h
86400
24h
432000
5days
18000
5h
tcp.closing
Half closed
900
15min
905
15min + 5sec
3600
1h
60
60sec
tcp.finwait
Got both FINs
45
45sec
50
50sec
600
10min
30
30sec
tcp.closed
Got an RST
90
90sec
95
95sec
180
3min
30
30sec
tcp.tsdiff
Allowed TS diff
30
30sec
60
60sec
60
60sec
10
10sec

The optimization algorithm satellite is an alias for high-latency; it is therefore not available in the pfSense GUI.

Firewall Maximum States controls the number of concurrent connections which can be tracked by the firewall. The rule of thumb is that one state table entry roughly consumes 1kB of kernel RAM. For an i386 version of pfSense, some special restrictions apply, as the kernel address space is limited to 1GB. Exceeding that value will cause a kernel panic and should therefore be avoided at all cost. While more than 1,200,000 state table entries have been achieved on a i386 machine with 2GB of physical RAM, a safe Firewall Maximum States setting would be 1,000,000 (one million states). There is no 1GB kernel address space limit on the AMD64 (x86-64) versions so higher values may be used on that architecture.

When the state table is full, no further connections will be accepted until existing connections are dropped from the state table. The firewall will not prematurely force active entries out of the table (no existing connection will be dropped in favor of new ones). However, the Firewall Adaptive Timeouts can be used to reduce timeouts and possibly time-out existing connections earlier when the state table is getting full.

Firewall Adaptive Timeouts speed up the expiration of state table entries as the state table gets fuller. The first parameter (a1 in the formulas) is the number of state table entries where adaptive timeouts start. The second parameter (a2 in the formulas) is the number of state table entries at which the timeouts would become zero. If the number of state table entries is between the two values, the timeouts are scaled linearly. The second parameter must be above the Firewall Maximum States limit - otherwise the firewall would drop all connections when it is reached.

The following formula calculates the adaptive timeout factor at the Firewall Maximum States limit:

Firewall Adaptive Timeouts solved for factor at limit.png

For example, if the limit is 100.000, the first parameter is 60.000 and the second parameter is 120.000, timeouts would be reduced to 1/3 (~33%) when the state table is full.

Solving the formula for the second parameter yields:

Firewall Adaptive Timeouts solved for a2.png

For example, if the state table limit is set to 100.000, adaptive timeouts should start at 60.000 and the desired timeout factor for a full state table is 0.25 (25%), the second parameter is calculated as 113.333.

If either parameter is zero, adaptive timeouts are disabled.

Networking

Contains settings for IPv6, and various network interface settings such as hardware checksums, device polling, and ARP message suppression.

Hardware Large Receive Offloading (LRO)

LRO works by aggregating multiple incoming packets from a single stream into a larger buffer before they are passed higher up the networking stack, thus reducing the number of packets to be processed. LRO should not be used on machines acting as routers as it breaks the end-to-end principle and can significantly impact performance. pfSense is most frequently used as a router or an equivalent.

Hardware TCP Segmentation Offloading (TSO)

TSO is similar to LRO, but for sending. It works by queuing large buffers and letting the network interface card (NIC) split them into separate packets just before transmit.

Implications

Both LRO and TSO can help when pfSense is an endpoint and not a router. If pfSense is being used as an appliance (e.g. for DNS), it is possible the options could enhance performance.

Hardware Checksum Offloading

The Ethernet hardware calculates the Ethernet CRC32 checksum and the receive engine validates this checksum. If the received checksum is wrong pfSense normally won't even see the packet, as the Ethernet hardware internally throws away the packet (though there are exceptions, such as when the interface is in promiscuous mode).

Higher-level checksums are "traditionally" calculated by the protocol implementation and the completed packet is then handed over to the hardware. Recent network hardware can perform the IP checksum calculation, also known as checksum offloading. The network driver won't calculate the checksum itself but will simply hand over an empty (zero or garbage-filled) checksum field to the hardware.

Some cards will additionally process TCP and UDP checksums, as above, but this isn't of value on a router.

It's possible, when everything else is right, that IP checksum offloading can provide a modest performance improvement, but this is unlikely to be more than "noticeable" with the traffic processed by most pfSense installations. However, at 10Gbps or above, such offloading can become quite useful. Support for these is an important component of the pfSense "3.0" effort.

Defaults

pfSense 2.x defaults both the LRO and TSO settings to disabled and the Hardware Checksum Offloading settings to enabled.

Miscellaneous

Contains settings that do not fit into the other categories:

  • Load Balancing (sticky connections)
  • PowerD
  • Temperature sensors
  • Hardware cryptographic acceleration
  • IPsec Advanced settings (until pfSense 2.2)
  • Disable clearing of states with scheduled rules

System Tunables

Contains an interface to manage setting various FreeBSD sysctl values that tweak various system behaviors.

Notifications

Controls how the system will notify administrators when an alert happens. Current options include SMTP (e-mail) notifications and growl notifications.

  • Growl
    • IP Address: This is the IP address which will receive growl notifications
    • Password: The password of the remote growl notification device
  • SMTP E-Mail
    • IP Address of E-Mail server: IP address or FQDN of the SMTP E-Mail server through which notifications will be sent
    • From e-mail address: E-mail address that will appear in the from field. (Ex. hostname@domain.com)
    • Notification E-Mail address: E-mail address which will receive email notifications
    • Notification E-Mail auth username (optional): Username for SMTP authentication
    • Notification E-Mail auth password (optional): Password for SMTP authentication

Click Test SMTP to send a test message after configuring and saving e-mail notifications to verify that the settings are correct.