Skip to content

Commit

Permalink
Update nsh.8
Browse files Browse the repository at this point in the history
Correct Spelling Mistakes througout the document
  • Loading branch information
ogmaconnect committed Sep 23, 2022
1 parent d10ca02 commit 23e2588
Showing 1 changed file with 41 additions and 41 deletions.
82 changes: 41 additions & 41 deletions nsh.8
Expand Up @@ -66,7 +66,7 @@ is a shell to configure
.Ox
kernel's networking functions such as routing
of packets, firewalling, network address translation, rate limiting,
bandwidth queueing, LAN bridging, IP tunneling, and encryption.
bandwidth queueing, LAN bridging, IP tunnelling, and encryption.
.Nm
provides simple wrappers around these functions to aid setting up a network.
The goals of this software are:
Expand Down Expand Up @@ -123,7 +123,7 @@ When run without any command line arguments,
presents an unprivileged shell to the user.
All
.Nm
interactive commandline modes allow basic command line editing features from
interactive command line modes allow basic command line editing features from
.Xr editline 7
library.
The command history of the current session is available through the up / down
Expand All @@ -142,7 +142,7 @@ e.g. ambiguous command entry.
nsh/show i
% Ambiguous argument i
.Ed
.Ss Commandline completion
.Ss Command line completion
.Nm
has double <Tab> command line completion for user convenience if the command
is not ambiguous double <Tab> completes the command.
Expand All @@ -151,7 +151,7 @@ with the available command line options that match what has been typed thus
far.
.Bl -dash
.It
E.g. commandline completion display
E.g. command line completion display
.Bd -literal -offset indent
nsh(p)/i
ifstate inet ip ipsec
Expand Down Expand Up @@ -196,7 +196,7 @@ nsh(p)/pf disable
.Ss Standard Prompt vs Privileged Prompt
.Nm
shell starts as an unprivileged prompt which displays as the text of the FQDN
(fully qualfied domain name) of the machine followed by a forward slash.
(fully qualified domain name) of the machine followed by a forward slash.
.Bl -dash
.It
e.g. standard prompt of the device firewall.machine.com
Expand Down Expand Up @@ -238,7 +238,7 @@ nsh(p)/
Display available commands and options that can used in the current
.Nm
mode.
The help or ? can be follwed by a
The help or ? can be followed by a
.Nm
command,
.Nm
Expand Down Expand Up @@ -380,7 +380,7 @@ kernel can accommodate 256 rtables.
They have a 1:1 relationship with routing domains, except that routing domain 0
can contain multiple routing tables.
In addition, routing tables initialized prior to their corresponding
routing domain, shall be inititalized with a routing domain of 0.
routing domain, shall be initialised with a routing domain of 0.
.Bl -dash
.It
e.g. Create a new routing table rdomain 3 create a loopback for rdomain 3.
Expand Down Expand Up @@ -728,7 +728,7 @@ The edited ruleset is automatically validated on saving and exiting the editor.
Note! firewall configuration changes DO NOT take effect until the "pf reload"
command is entered.
The editor used by nsh can be customised to your preferred editor using the
EDITOR and VISUAL enviornment variables.
EDITOR and VISUAL environment variables.
For packet filter configuration syntax, refer to
.Xr pf.conf 5 .
.Bd -literal -offset indent
Expand Down Expand Up @@ -758,8 +758,8 @@ The edited ruleset is automatically validated on saving and exiting the
editor.
Note ospfd configuration changes DO NOT take effect until the "ospf reload"
command is entered.
The editor used by nsh can be customised to your prefered editor using the
EDITOR and VISUAL enviornment variables.
The editor used by nsh can be customised to your preferred editor using the
EDITOR and VISUAL environment variables.
For OSPF configuration syntax, refer to
.Xr ospfd.conf 5 .
.Bd -literal -offset indent
Expand All @@ -784,7 +784,7 @@ The decouple feature is useful for monitoring OSPF networks without affecting
the routing table of the system.
OSPF decouple should only be done where there is only one link between the
system and the rest of the OSPF network.
The ospf fib reload command refetches and relearns the routes in the FIB and
The ospf fib reload command re fetches and relearns the routes in the FIB and
passes them to the ospfd daemon for processing.
.Bd -literal -offset indent
nsh(p)/ospf fib decouple
Expand All @@ -797,7 +797,7 @@ Configure the detail level of
.Xr ospfd 8
logging messages.
Set ospf log verbose to enable detailed debug log output from ospfd.
set ospf log brief to disable detalled debug log output from ospfd.
set ospf log brief to disable detailed debug log output from ospfd.
.Bd -literal -offset indent
nsh(p)/ospf log verbose
.Ed
Expand Down Expand Up @@ -976,7 +976,7 @@ The configuration of
.Ic dvmrp
daemon can be edited with
.Cm edit
commmand, the configuration syntax of
command, the configuration syntax of
.Ic dvmrp
daemon is documented in
.Xr dvmrpd.conf 5
Expand All @@ -1003,7 +1003,7 @@ command, the syntax is documented in
.Pp
Enable or disable or configure the
.Xr sasyncd 8
IPSec Security Associaton synchronisation daemon for failover gateways.
IPSec Security Association synchronisation daemon for failover gateways.
The configuration of
.Ic sasync
daemon can be edited with
Expand Down Expand Up @@ -1070,7 +1070,7 @@ to control the
.Xr ldapd 8
daemon in a similar manner to
.Xr ldapctl 8
e.g. to set log verbose vs brief or to compact / reindex the LDAP database
e.g. to set log verbose vs brief or to compact / re-index the LDAP database
are documented in
.Xr ldapctl 8 .
.Pp
Expand Down Expand Up @@ -1478,7 +1478,7 @@ nsh(p)/show hostname
.Pp
Display essential information about the system network interfaces including
any network bridges / switches.
show interface without any arguments diplays information about all
show interface without any arguments displays information about all
interfaces available on the system.
.Pp
show interface
Expand Down Expand Up @@ -1552,7 +1552,7 @@ nsh(p)/show interface lo0

.Ed
.It
With a bridge, verbose mode diplays spanning tree member states and bridge
With a bridge, verbose mode displays spanning tree member states and bridge
members.
.Bd -literal -offset indent
nsh/show int bridge0
Expand All @@ -1569,7 +1569,7 @@ nsh/show int bridge0
.Ed
.It
With an IEEE 802.11 wireless interface, verbose mode displays the network ID,
network key, and powersaving mode (if enabled).
network key, and power-saving mode (if enabled).
.Bd -literal -offset indent
nsh/show int athn0
% athn0
Expand Down Expand Up @@ -1860,7 +1860,7 @@ nsh(p)/no editing
.Pp
Invoke a shell or run an entered shell-command with arguments if required.
(requires privileged mode).
The active users login shell is the shell that is involked by this feature.
The active users login shell is the shell that is invoked by this feature.
This feature disabled to enhance security.
.Pp
E.g. list files in /root
Expand Down Expand Up @@ -2351,7 +2351,7 @@ Note that this command clears all existing ip configuration on the interface.
.Op Ar rtable-id
.Pp
TODO Set or remove the rtable id on an interface.
TODO better exlplanation needed!
TODO better explanation needed!
.Pp
.Op no
.Ic priority
Expand Down Expand Up @@ -2630,7 +2630,7 @@ Set or remove configured transmit priority of the headers of a tunnel interface.
Valid options are standard traffic priority values (0-7) or set the headers
according to encapsulated packet or payload priority.
.Pp
E.g. to set the priority of headsrs of the tunnel gre1 to match that of the
E.g. to set the priority of headers of the tunnel gre1 to match that of the
payload.
.Bd -literal -offset indent
nsh(interface-gre1)/txprio payload
Expand All @@ -2639,7 +2639,7 @@ nsh(interface-gre1)/txprio payload
.Op no
.Cm rxprio Op Ar 0-7 | packet | payload
.Pp
Set or remove configured recieve priority of the headers of a tunnel
Set or remove configured receive priority of the headers of a tunnel
interfaces are standard traffic priority values (0-7) or set the headers
according to encapsulated packet / payload priority).
.Pp
Expand All @@ -2654,8 +2654,8 @@ nsh(interface-gre1)/rxprio 7
.Pp
Set or remove the 24 bit virtual network identifier tag.
Virtual network identifier tags are typically used in large multi tenant VXLAN
multiple routing domain enviornments.
If vnetid involked inside a vlan interface the acceptable range is the
multiple routing domain environments.
If vnetid invoked inside a vlan interface the acceptable range is the
standard 12-bit vlan id 1-4094 of the IEEE 802.1Q VLAN tag.
.Pp
E.g. set gre1 vnetid to 8192.
Expand All @@ -2667,7 +2667,7 @@ nsh(interface-gre1)/vnetid 8192
.Cm vnetflowid
.Pp
Allow or disallow the interface to use a portion of the virtual network
identifiier space as a flow identifier.
identifier space as a flow identifier.
This allowOBs loadbalancing of the encapsulated traffic over multiple links.
.Pp
E.g. enable vnetflowid load balancing for gre1.
Expand All @@ -2690,7 +2690,7 @@ nsh(interface-vlan1024)/parent em0
.Ic patch
.Ar pair-interface-name
.Pp
Set or remove patch (layer1+ conection) between current interface and another
Set or remove patch (layer1+ connection) between current interface and another
pair(4) interface.
A patch is a CPU efficient way of forwarding packets between two
.Xr pair 4
Expand All @@ -2703,7 +2703,7 @@ Patch can only connect two
.Xr pair 4
interfaces, no other interface types are supported.
.Pp
E.g. To connect pair1 and pair2 intefaces with a virtual patch cable.
E.g. To connect pair1 and pair2 interfaces with a virtual patch cable.
.Bd -literal -offset indent -compact
nsh(p)/interface pair1

Expand Down Expand Up @@ -2795,7 +2795,7 @@ Note that the pfsync protocol currently includes no authentication method.
It is advisable to layer authentication, signing and (possibly encrypted
tunnels for the underlay interfaces.
For simplicity on co-located pfsynced firewalls a secure way to use pfsync,
is through a a direct (layer1 (i.e. no switches)) cableonnecting directly
is through a a direct (layer1 (i.e. no switches)) cable connecting directly
between two pfsync capable devices (i.e. a conenction made with an ethernet
patch cable).
This command is valid only for
Expand All @@ -2811,7 +2811,7 @@ nsh(interface-pfsync0)/syncdev em1
.Ic syncpeer
.Op Ar ipv4-peer-pfsync-address
.Pp
Set or remove a manually enetered ip address of the pfsync interface of a
Set or remove a manually entered ip address of the pfsync interface of a
peer pf sync firewall.
By default, state change messages are sent out on the synchronisation
interface using IP multicast packets to the 224.0.0.240 group address.
Expand All @@ -2832,7 +2832,7 @@ nsh(interface-pfsync0)/syncpeer 192.0.0.10
.Ar 0-255
.Op defer
.Pp
Configures or removes the maximum number of updates which are collapsable into
Configures or removes the maximum number of updates which are collapsible into
one for a single state.
The default value is 128.
The transmission a pfsync update packet shall be delayed by a maximum of 1
Expand Down Expand Up @@ -2926,7 +2926,7 @@ nsh(interface-carp0)/carppass 19CharPassphrase!!!
.Op Ar interface-name
.Pp
Set or remove the interface on which the selected carp interface's carp
advertisments are sent and received.
advertisements are sent and received.
The carpdev is the "real interface" over which the carp virtual IP is
accessible.
carpdev is valid for
Expand Down Expand Up @@ -3001,11 +3001,11 @@ ip-stealth carp wont send packets with its own virtual MAC
virtual MAC address, therefore the switch would
unicast flood traffic to all switch ports
(unless there is some swithc acls to prevent flooding
unneccessarily.
unnecessarily.
.It
ip-unicast Used in conjunction with a HUB or a switch that
can replicate packets (monitoring or mirror) or
other non-standard switch forwarding mechanisim.
other non-standard switch forwarding mechanism.
.El
Note: IP balancing is being used on a firewall, it is recommended to
configure the carpnodes in a symmetrical manner.
Expand Down Expand Up @@ -3099,7 +3099,7 @@ nsh(interface-em0)/wol
.Ic mpls
.Pp
Set or remove the MPLS flag on the selected interface,if mpls is set on the
interface, the interface can send and recieve mpls traffic.
interface, the interface can send and receive mpls traffic.
.Pp
E.g enable mpls on em0
.Bd -literal -offset indent
Expand All @@ -3123,8 +3123,8 @@ nsh(interface-em0)/rad
.Op no
.Ic autoconf6
.Pp
Enable or disable IPv6 auto configuation of Ipv6 address on the inteface.
If autoconf6 is used alone (without tempoarary or autoconfprivacy being set
Enable or disable IPv6 auto configuration of Ipv6 address on the inteface.
If autoconf6 is used alone (without temporary or autoconfprivacy being set
on the interface then the autoconfigured address assigned is repeatable based
on the MAC address of the interface (EUI64).
.Pp
Expand Down Expand Up @@ -3207,8 +3207,8 @@ The psk and preshared-key is optional but recommended as it supplements the
public key cryptography with symmetric key cryptography.
.Pp
.Ic aip Ar allowed-ip/prefix
Set the peer's allowed IPv4 or IPv6 addresses or prefixes for tunneled traffic.
The option be reppeated to set multiple allowed ip/ranges.
Set the peer's allowed IPv4 or IPv6 addresses or prefixes for tunnelled traffic.
The option be repeated to set multiple allowed ip/ranges.
No addresses are allowed by default.
.Bl -dash
.It
Expand Down Expand Up @@ -3247,7 +3247,7 @@ nsh(interface-wg0)/wgkey QComa+ca+mWih+Vl/5G/p+UwhYy17hw5vdwysZpIAn0=
.Op no
.Ic wgport Ar 0-65535
.Pp
Set or remove the configuraiton for the local UDP port to be used by the
Set or remove the configuration for the local UDP port to be used by the
current wireguard interface when exchanging traffic with its wireguard peers.
The interface binds to INADDR_ANY and IN6ADDR_ANY_INIT.
If
Expand All @@ -3273,7 +3273,7 @@ between 0 and 255 on a default
kernel.
The routing domain of the rtable does not need be in the same routing domain
to which the interface is attached.
wgrtable configures which rdomain the interface's tunneled traffic appears.
wgrtable configures which rdomain the interface's tunnelled traffic appears.
.Pp
E.g. set wireguard interface wg0 routing table to routing domain 5.
.Bd -literal -offset indent
Expand Down Expand Up @@ -3356,7 +3356,7 @@ These interface names start with 'enc'.
.Pp
Generic Tunnel: This interface is used to configure a network tunnel to
another host or router.
It follows the RFC1933 tunneling standard.
It follows the RFC1933 tunnelling standard.
These interface names start with 'gif'.
.Pp
Ethernet Bridge: This interface is used to configure layer 2 bridging
Expand Down

0 comments on commit 23e2588

Please sign in to comment.