MFC r300667:

[FreeBSD/stable/8.git] / contrib / bsnmp / snmpd / snmpmod.3

.\"
.\" Copyright (c) 2004-2005
.\"     Hartmut Brandt.
.\"     All rights reserved.
.\" Copyright (c) 2001-2003
.\"     Fraunhofer Institute for Open Communication Systems (FhG Fokus).
.\"     All rights reserved.
.\"
.\" Author: Harti Brandt <harti@FreeBSD.org>
.\" 
.\" Redistribution and use in source and binary forms, with or without
.\" modification, are permitted provided that the following conditions
.\" are met:
.\" 1. Redistributions of source code must retain the above copyright
.\"    notice, this list of conditions and the following disclaimer.
.\" 2. Redistributions in binary form must reproduce the above copyright
.\"    notice, this list of conditions and the following disclaimer in the
.\"    documentation and/or other materials provided with the distribution.
.\" 
.\" THIS SOFTWARE IS PROVIDED BY AUTHOR AND CONTRIBUTORS ``AS IS'' AND
.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
.\" ARE DISCLAIMED.  IN NO EVENT SHALL AUTHOR OR CONTRIBUTORS BE LIABLE
.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
.\" SUCH DAMAGE.
.\"
.\" $Begemot: bsnmp/snmpd/snmpmod.3,v 1.14 2005/10/04 13:30:35 brandt_h Exp $
.\"
.Dd February 27, 2006
.Dt SNMPMOD 3
.Os
.Sh NAME
.Nm INSERT_OBJECT_OID_LINK_INDEX ,
.Nm INSERT_OBJECT_INT_LINK_INDEX ,
.Nm FIND_OBJECT_OID_LINK_INDEX ,
.Nm NEXT_OBJECT_OID_LINK_INDEX ,
.Nm FIND_OBJECT_INT_LINK_INDEX ,
.Nm NEXT_OBJECT_INT_LINK_INDEX ,
.Nm INSERT_OBJECT_OID_LINK ,
.Nm INSERT_OBJECT_INT_LINK ,
.Nm FIND_OBJECT_OID_LINK ,
.Nm NEXT_OBJECT_OID_LINK ,
.Nm FIND_OBJECT_INT_LINK ,
.Nm NEXT_OBJECT_INT_LINK ,
.Nm INSERT_OBJECT_OID ,
.Nm INSERT_OBJECT_INT ,
.Nm FIND_OBJECT_OID ,
.Nm FIND_OBJECT_INT ,
.Nm NEXT_OBJECT_OID ,
.Nm NEXT_OBJECT_INT ,
.Nm this_tick ,
.Nm start_tick ,
.Nm get_ticks ,
.Nm systemg ,
.Nm comm_define ,
.Nm community ,
.Nm oid_zeroDotZero ,
.Nm reqid_allocate ,
.Nm reqid_next ,
.Nm reqid_base ,
.Nm reqid_istype ,
.Nm reqid_type ,
.Nm timer_start ,
.Nm timer_start_repeat ,
.Nm timer_stop ,
.Nm fd_select ,
.Nm fd_deselect ,
.Nm fd_suspend ,
.Nm fd_resume ,
.Nm or_register ,
.Nm or_unregister ,
.Nm buf_alloc ,
.Nm buf_size ,
.Nm snmp_input_start ,
.Nm snmp_input_finish ,
.Nm snmp_output ,
.Nm snmp_send_port ,
.Nm snmp_send_trap ,
.Nm string_save ,
.Nm string_commit ,
.Nm string_rollback ,
.Nm string_get ,
.Nm string_get_max ,
.Nm string_free ,
.Nm ip_save ,
.Nm ip_rollback ,
.Nm ip_commit ,
.Nm ip_get ,
.Nm oid_save ,
.Nm oid_rollback ,
.Nm oid_commit ,
.Nm oid_get ,
.Nm index_decode ,
.Nm index_compare ,
.Nm index_compare_off ,
.Nm index_append ,
.Nm index_append_off
.Nd "SNMP daemon loadable module interface"
.Sh LIBRARY
Begemot SNMP library
.Pq libbsnmp, -lbsnmp
.Sh SYNOPSIS
.In bsnmp/snmpmod.h
.Fn INSERT_OBJECT_OID_LINK_INDEX "PTR" "LIST" "LINK" "INDEX"
.Fn INSERT_OBJECT_INT_LINK_INDEX "PTR" "LIST" "LINK" "INDEX"
.Fn FIND_OBJECT_OID_LINK_INDEX "LIST" "OID" "SUB" "LINK" "INDEX"
.Fn FIND_OBJECT_INT_LINK_INDEX "LIST" "OID" "SUB" "LINK" "INDEX"
.Fn NEXT_OBJECT_OID_LINK_INDEX "LIST" "OID" "SUB" "LINK" "INDEX"
.Fn NEXT_OBJECT_INT_LINK_INDEX "LIST" "OID" "SUB" "LINK" "INDEX"
.Fn INSERT_OBJECT_OID_LINK "PTR" "LIST" "LINK"
.Fn INSERT_OBJECT_INT_LINK "PTR" "LIST" "LINK"
.Fn FIND_OBJECT_OID_LINK "LIST" "OID" "SUB" "LINK"
.Fn FIND_OBJECT_INT_LINK "LIST" "OID" "SUB" "LINK"
.Fn NEXT_OBJECT_OID_LINK "LIST" "OID" "SUB" "LINK"
.Fn NEXT_OBJECT_INT_LINK "LIST" "OID" "SUB" "LINK"
.Fn INSERT_OBJECT_OID "PTR" "LIST"
.Fn INSERT_OBJECT_INT "PTR" "LIST"
.Fn FIND_OBJECT_OID "LIST" "OID" "SUB"
.Fn FIND_OBJECT_INT "LIST" "OID" "SUB"
.Fn NEXT_OBJECT_OID "LIST" "OID" "SUB"
.Fn NEXT_OBJECT_INT "LIST" "OID" "SUB"
.Vt extern uint64_t this_tick ;
.Vt extern uint64_t start_tick ;
.Ft uint64_t
.Fn get_ticks "void"
.Vt extern struct systemg systemg ;
.Ft u_int
.Fn comm_define "u_int priv" "const char *descr" "struct lmodule *mod" "const char *str"
.Ft const char *
.Fn comm_string "u_int comm"
.Vt extern u_int community ;
.Vt extern const struct asn_oid oid_zeroDotZero ;
.Ft u_int
.Fn reqid_allocate "int size" "struct lmodule *mod"
.Ft int32_t
.Fn reqid_next "u_int type"
.Ft int32_t
.Fn reqid_base "u_int type"
.Ft int
.Fn reqid_istype "int32_t reqid" "u_int type"
.Ft u_int
.Fn reqid_type "int32_t reqid"
.Ft void *
.Fn timer_start "u_int ticks" "void (*func)(void *)" "void *uarg" "struct lmodule *mod"
.Ft void *
.Fn timer_start_repeat "u_int ticks" "u_int repeat_ticks" "void (*func)(void *)" "void *uarg" "struct lmodule *mod"
.Ft void
.Fn timer_stop "void *timer_id"
.Ft void *
.Fn fd_select "int fd" "void (*func)(int, void *)" "void *uarg" "struct lmodule *mod"
.Ft void
.Fn fd_deselect "void *fd_id"
.Ft void
.Fn fd_suspend "void *fd_id"
.Ft int
.Fn fd_resume "void *fd_id"
.Ft u_int
.Fn or_register "const struct asn_oid *oid" "const char *descr" "struct lmodule *mod"
.Ft void
.Fn or_unregister "u_int or_id"
.Ft void *
.Fn buf_alloc "int tx"
.Ft size_t
.Fn buf_size "int tx"
.Ft enum snmpd_input_err
.Fo snmp_input_start
.Fa "const u_char *buf" "size_t len" "const char *source"
.Fa "struct snmp_pdu *pdu" "int32_t *ip" "size_t *pdulen"
.Fc
.Ft enum snmpd_input_err
.Fo snmp_input_finish
.Fa "struct snmp_pdu *pdu" "const u_char *rcvbuf"
.Fa "size_t rcvlen" "u_char *sndbuf" "size_t *sndlen" "const char *source"
.Fa "enum snmpd_input_err ierr" "int32_t ip" "void *data"
.Fc
.Ft void
.Fo snmp_output
.Fa "struct snmp_pdu *pdu" "u_char *sndbuf" "size_t *sndlen"
.Fa "const char *dest"
.Fc
.Ft void
.Fo snmp_send_port
.Fa "void *trans" "const struct asn_oid *port"
.Fa "struct snmp_pdu *pdu" "const struct sockaddr *addr" "socklen_t addrlen"
.Fc
.Ft void
.Fn snmp_send_trap "const struct asn_oid *oid" "..."
.Ft int
.Fn string_save "struct snmp_value *val" "struct snmp_context *ctx" "ssize_t req_size" "u_char **strp"
.Ft void
.Fn string_commit "struct snmp_context *ctx"
.Ft void
.Fn string_rollback "struct snmp_context *ctx" "u_char **strp"
.Ft int
.Fn string_get "struct snmp_value *val" "const u_char *str" "ssize_t len"
.Ft int
.Fn string_get_max "struct snmp_value *val" "const u_char *str" "ssize_t len" "size_t maxlen"
.Ft void
.Fn string_free "struct snmp_context *ctx"
.Ft int
.Fn ip_save "struct snmp_value *val" "struct snmp_context *ctx" "u_char *ipa"
.Ft void
.Fn ip_rollback "struct snmp_context *ctx" "u_char *ipa"
.Ft void
.Fn ip_commit "struct snmp_context *ctx"
.Ft int
.Fn ip_get "struct snmp_value *val" "u_char *ipa"
.Ft int
.Fn oid_save "struct snmp_value *val" "struct snmp_context *ctx" "struct asn_oid *oid"
.Ft void
.Fn oid_rollback "struct snmp_context *ctx" "struct asn_oid *oid"
.Ft void
.Fn oid_commit "struct snmp_context *ctx"
.Ft int
.Fn oid_get "struct snmp_value *val" "const struct asn_oid *oid"
.Ft int
.Fn index_decode "const struct asn_oid *oid" "u_int sub" "u_int code" "..."
.Ft int
.Fn index_compare "const struct asn_oid *oid1" "u_int sub" "const struct asn_oid *oid2"
.Ft int
.Fn index_compare_off "const struct asn_oid *oid1" "u_int sub" "const struct asn_oid *oid2" "u_int off"
.Ft void
.Fn index_append "struct asn_oid *dst" "u_int sub" "const struct asn_oid *src"
.Ft void
.Fn index_append_off "struct asn_oid *dst" "u_int sub" "const struct asn_oid *src" "u_int off"
.Sh DESCRIPTION
The
.Xr bsnmpd 1
SNMP daemon implements a minimal MIB which consists of the system group, part
of the SNMP MIB, a private configuration MIB, a trap destination table, a
UDP port table, a community table, a module table, a statistics group and
a debugging group.
All other MIBs are support through loadable modules.
This allows
.Xr bsnmpd 1
to use for task, that are not the classical SNMP task.
.Ss MODULE LOADING AND UNLOADING
Modules are loaded by writing to the module table.
This table is indexed by a string, that identifies the module to the daemon.
This identifier is used
to select the correct configuration section from the configuration files and
to identify resources allocated to this module.
A row in the module table is
created by writing a string of non-zero length to the
.Va begemotSnmpdModulePath
column.
This string must be the complete path to the file containing the module.
A module can be unloaded by writing a zero length string to the path column
of an existing row.
.Pp
Modules may depend on each other an hence must be loaded in the correct order.
The dependencies are listed in the corresponding manual pages.
.Pp
Upon loading a module the SNMP daemon expects the module file to a export
a global symbol
.Va config .
This symbol should be a variable of type
.Vt struct snmp_module :
.Bd -literal -offset indent
typedef enum snmpd_proxy_err (*proxy_err_f)(struct snmp_pdu *, void *,
    const struct asn_oid *, const struct sockaddr *, socklen_t,
    enum snmpd_input_err, int32_t);


struct snmp_module {
        const char *comment;
        int (*init)(struct lmodule *, int argc, char *argv[]);
        int (*fini)(void);
        void (*idle)(void);
        void (*dump)(void);
        void (*config)(void);
        void (*start)(void);
        proxy_err_f proxy;
        const struct snmp_node *tree;
        u_int tree_size;
        void (*loading)(const struct lmodule *, int);
};
.Ed
.Pp
This structure must be statically initialized and its fields have the
following functions:
.Bl -tag -width ".It Va tree_size"
.It Va comment
This is a string that will be visible in the module table.
It should give some hint about the function of this module.
.It Va init
This function is called upon loading the module.
The module pointer should
be stored by the module because it is needed in other calls and the
argument vector will contain the arguments to this module from the daemons
command line.
This function should return 0 if everything is ok or an UNIX error code (see
.Xr errno 3 ) .
Once the function returns 0, the
.Va fini
function is called when the module is unloaded.
.It Va fini
The module is unloaded.
This gives the module a chance to free resources that
are not automatically freed.
Be sure to free all memory, because daemons tend to run very long.
This function pointer may be
.Li NULL
if it is not needed.
.It Va idle
If this function pointer is not
.Li NULL ,
the function pointed to by it is called whenever the daemon is going
to wait for an event.
Try to avoid using this feature.
.It Va dump
Whenever the daemon receives a
.Li SIGUSR1
it dumps it internal state via
.Xr syslog 3 .
If the
.Va dump
field is not
.Li NULL
it is called by the daemon to dump the state of the module.
.It Va config
Whenever the daemon receives a
.Li SIGHUP
signal it re-reads its configuration file.
If the
.Va config
field is not
.Li NULL
it is called after reading the configuration file to give the module a chance
to adapt to the new configuration.
.It Va start
If not
.Li NULL
this function is called after successful loading and initializing the module
to start its actual operation.
.It Va proxy
If the daemon receives a PDU and that PDU has a community string whose
community was registered by this module and
.Va proxy
is not
.Li NULL
than this function is called to handle the PDU.
.It Va tree
This is a pointer to the node array for the MIB tree implemented by this module.
.It Va tree_size
This is the number of nodes in
.Va tree .
.It Va loading
If this pointer is not
.Li NULL
it is called whenever another module was loaded or unloaded.
It gets a
pointer to that module and a flag that is 0 for unloading and 1 for loading.
.El
.Pp
When everything is ok, the daemon merges the module's MIB tree into its current
global tree, calls the modules
.Fn init
function.
If this function returns an error, the modules MIB tree is removed from
the global one and the module is unloaded.
If initialization is successful, the modules
.Fn start
function is called.
After it returns the
.Fn loaded
functions of all modules (including the loaded one) are called.
.Pp
When the module is unloaded, its MIB tree is removed from the global one,
the communities, request id ranges, running timers and selected file
descriptors are released, the
.Fn fini
function is called, the module file is unloaded and the
.Fn loaded
functions of all other modules are called.
.Ss IMPLEMENTING TABLES
There are a number of macros designed to help implementing SNMP tables.
A problem while implementing a table is the support for the GETNEXT operator.
The GETNEXT operation has to find out whether, given an arbitrary OID, the
lessest table row, that has an OID higher than the given OID.
The easiest way
to do this is to keep the table as an ordered list of structures each one
of which contains an OID that is the index of the table row.
This allows easy removal, insertion and search.
.Pp
The helper macros assume, that the table is organized as a TAILQ (see
.Xr queue 3
and each structure contains a
.Vt struct asn_oid
that is used as index.
For simple tables with only a integer or unsigned index, an alternate form
of the macros is available, that presume the existence of an integer or
unsigned field as index field.
.Pp
The macros have name of the form
.Bd -literal -offset indent
{INSERT,FIND,NEXT}_OBJECT_{OID,INT}[_LINK[_INDEX]]
.Ed
.Pp
The
.Fn INSERT_*
macros are used in the SET operation to insert a new table row into the table.
The
.Fn FIND_*
macros are used in the GET operation to find a specific row in the table.
The
.Fn NEXT_*
macros are used in the GETNEXT operation to find the next row in the table.
The last two macros return a pointer to the row structure if a row is found,
.Li NULL
otherwise.
The macros
.Fn *_OBJECT_OID_*
assume the existence of a
.Vt struct asn_oid
that is used as index, the macros
.Fn *_OBJECT_INT_*
assume the existence of an unsigned integer field that is used as index.
.Pp
The macros
.Fn *_INDEX
allow the explicit naming of the index field in the parameter
.Fa INDEX ,
whereas the other macros assume that this field is named
.Va index .
The macros
.Fn *_LINK_*
allow the explicit naming of the link field of the tail queues, the others
assume that the link field is named
.Va link .
Explicitly naming the link field may be necessary if the same structures
are held in two or more different tables.
.Pp
The arguments to the macros are as follows:
.Bl -tag -width "INDEX"
.It Fa PTR
A pointer to the new structure to be inserted into the table.
.It Fa LIST
A pointer to the tail queue head.
.It Fa LINK
The name of the link field in the row structure.
.It Fa INDEX
The name of the index field in the row structure.
.It Fa OID
Must point to the
.Va var
field of the
.Fa value
argument to the node operation callback.
This is the OID to search for.
.It Fa SUB
This is the index of the start of the table index in the OID pointed to
by
.Fa OID .
This is usually the same as the
.Fa sub
argument to the node operation callback.
.El
.Ss DAEMON TIMESTAMPS
The variable
.Va this_tick
contains the tick (there are 100 SNMP ticks in a second) when
the current PDU processing was started.
The variable
.Va start_tick
contains the tick when the daemon was started.
The function
.Fn get_ticks
returns the current tick.
The number of ticks since the daemon was started
is
.Bd -literal -offset indent
get_ticks() - start_tick
.Ed
.Ss THE SYSTEM GROUP
The scalar fields of the system group are held in the global variable
.Va systemg :
.Bd -literal -offset indent
struct systemg {
        u_char          *descr;
        struct asn_oid  object_id;
        u_char          *contact;
        u_char          *name;
        u_char          *location;
        uint32_t        services;
        uint32_t        or_last_change;
};
.Ed
.Ss COMMUNITIES
The SNMP daemon implements a community table.
On recipte of a request message
the community string in that message is compared to each of the community
strings in that table, if a match is found, the global variable
.Va community
is set to the community identifier for that community.
Community identifiers are unsigned integers.
For the three standard communities there are three constants defined:
.Bd -literal -offset indent
#define COMM_INITIALIZE 0
#define COMM_READ       1
#define COMM_WRITE      2
.Ed
.Pp
.Va community
is set to
.Li COMM_INITIALIZE
while the assignments in the configuration file are processed.
To
.Li COMM_READ
or
.Li COMM_WRITE
when the community strings for the read-write or read-only community are found
in the incoming PDU.
.Pp
Modules can define additional communities.
This may be necessary to provide
transport proxying (a PDU received on one communication link is proxied to
another link) or to implement non-UDP access points to SNMP.
A new community is defined with the function
.Fn comm_define .
It takes the following parameters:
.Bl -tag -width ".It Fa descr"
.It Fa priv
This is an integer identifying the community to the module.
Each module has its own namespace with regard to this parameter.
The community table is indexed with the module name and this identifier.
.It Fa descr
This is a string providing a human readable description of the community.
It is visible in the community table.
.It Fa mod
This is the module defining the community.
.It Fa str
This is the initial community string.
.El
.Pp
The function returns a globally unique community identifier.
If a PDU is
received who's community string matches, this identifier is set into the global
.Va community .
.Pp
The function
.Fn comm_string
returns the current community string for the given community.
.Pp
All communities defined by a module are automatically released when the module
is unloaded.
.Ss WELL KNOWN OIDS
The global variable
.Va oid_zeroDotZero
contains the OID 0.0.
.Ss REQUEST ID RANGES
For modules that implement SNMP client functions besides SNMP agent functions
it may be necessary to identify SNMP requests by their identifier to allow
easier routing of responses to the correct sub-system.
Request id ranges
provide a way to acquire globally non-overlapping sub-ranges of the entire
31-bit id range.
.Pp
A request id range is allocated with
.Fn reqid_allocate .
The arguments are: the size of the range and the module allocating the range.
For example, the call
.Bd -literal -offset indent
id = reqid_allocate(1000, module);
.Ed
.Pp
allocates a range of 1000 request ids.
The function returns the request
id range identifier or 0 if there is not enough identifier space.
The function
.Fn reqid_base
returns the lowest request id in the given range.
.Pp
Request id are allocated starting at the lowest one linear throughout the range.
If the client application may have a lot of outstanding request the range
must be large enough so that an id is not reused until it is really expired.
.Fn reqid_next
returns the sequentially next id in the range.
.Pp
The function
.Fn reqid_istype
checks whether the request id
.Fa reqid
is within the range identified by
.Fa type .
The function
.Fn reqid_type
returns the range identifier for the given
.Fa reqid
or 0 if the request id is in none of the ranges.
.Ss TIMERS
The SNMP daemon supports an arbitrary number of timers with SNMP tick granularity.
The function
.Fn timer_start
arranges for the callback
.Fa func
to be called with the argument
.Fa uarg
after
.Fa ticks
SNMP ticks have expired.
.Fa mod
is the module that starts the timer.
These timers are one-shot, they are not restarted.
Repeatable timers are started with
.Fn timer_start_repeat
which takes an additional argument
.Fa repeat_ticks .
The argument
.Fa ticks
gives the number of ticks until the first execution of the callback, while
.Fa repeat_ticks
is the number of ticks between invocations of the callback.
Note, that currently the number of initial ticks silently may be set identical
to the number of ticks between callback invocations.
The function returns a timer identifier that can be used to stop the timer via
.Fn timer_stop .
If a module is unloaded all timers started by the module that have not expired
yet are stopped.
.Ss FILE DESCRIPTOR SUPPORT
A module may need to get input from socket file descriptors without blocking
the daemon (for example to implement alternative SNMP transports).
.Pp
The function
.Fn fd_select
causes the callback function
.Fa func
to be called with the file descriptor
.Fa fd
and the user argument
.Fa uarg
whenever the file descriptor
.Fa fd
can be read or has a close condition.
If the file descriptor is not in
non-blocking mode, it is set to non-blocking mode.
If the callback is not needed anymore,
.Fn fd_deselect
may be called with the value returned from
.Fn fd_select .
All file descriptors selected by a module are automatically deselected when
the module is unloaded.
.Pp
To temporarily suspend the file descriptor registration
.Fn fd_suspend
can be called.
This also causes the file descriptor to be switched back to
blocking mode if it was blocking prior the call to
.Fn fd_select .
This is necessary to do synchronous input on a selected socket.
The effect of
.Fn fd_suspend
can be undone with
.Fn fd_resume .
.Ss OBJECT RESOURCES
The system group contains an object resource table.
A module may create an entry in this table by calling
.Fn or_register
with the
.Fa oid
to be registered, a textual description in
.Fa str
and a pointer to the module
.Fa mod .
The registration can be removed with
.Fn or_unregister .
All registrations of a module are automatically removed if the module is
unloaded.
.Ss TRANSMIT AND RECEIVE BUFFERS
A buffer is allocated via
.Fn buf_alloc .
The argument must be 1 for transmit and 0 for receive buffers.
The function may return
.Li NULL
if there is no memory available.
The current buffersize can be obtained with
.Fn buf_size .
.Sh PROCESSING PDUS
For modules that need to do their own PDU processing (for example for proxying)
the following functions are available:
.Pp
Function
.Fn snmp_input_start
decodes the PDU, searches the community, and sets the global
.Va this_tick .
It returns one of the following error codes:
.Bl -tag -width ".It Er SNMPD_INPUT_VALBADLEN"
.It Er SNMPD_INPUT_OK
Everything ok, continue with processing.
.It Er SNMPD_INPUT_FAILED
The PDU could not be decoded, has a wrong version or an unknown
community string.
.It Er SNMPD_INPUT_VALBADLEN
A SET PDU had a value field in a binding with a wrong length field in an
ASN.1 header.
.It Er SNMPD_INPUT_VALRANGE
A SET PDU had a value field in a binding with a value that is out of range
for the given ASN.1 type.
.It Er SNMPD_INPUT_VALBADENC
A SET PDU had a value field in a binding with wrong ASN.1 encoding.
.It Er SNMPD_INPUT_TRUNC
The buffer appears to contain a valid begin of a PDU, but is too short.
For streaming transports this means that the caller must save what he
already has and trying to obtain more input and reissue this input to
the function.
For datagram transports this means that part of the
datagram was lost and the input should be ignored.
.El
.Pp
The function
.Fn snmp_input_finish
does the other half of processing: if
.Fn snmp_input_start
did not return OK, tries to construct an error response.
If the start was OK, it calls the correct function from
.Xr bsnmpagent 3
to execute the request and depending on the outcome constructs a response or
error response PDU or ignores the request PDU.
It returns either
.Er SNMPD_INPUT_OK
or
.Er SNMPD_INPUT_FAILED .
In the first case a response PDU was constructed and should be sent.
.Pp
The function
.Fn snmp_output
takes a PDU and encodes it.
.Pp
The function
.Fn snmp_send_port
takes a PDU, encodes it and sends it through the given port (identified by
the transport and the index in the port table) to the given address.
.Pp
The function
.Fn snmp_send_trap
sends a trap to all trap destinations.
The arguments are the
.Fa oid
identifying the trap and a NULL-terminated list of
.Vt struct snmp_value
pointers that are to be inserted into the trap binding list.
.Ss SIMPLE ACTION SUPPORT
For simple scalar variables that need no dependencies a number of support
functions is available to handle the set, commit, rollback and get.
.Pp
The following functions are used for OCTET STRING scalars, either NUL terminated
or not:
.Bl -tag -width "XXXXXXXXX"
.It Fn string_save
should be called for SNMP_OP_SET.
.Fa value
and
.Fa ctx
are the resp\&.\& arguments to the node callback.
.Fa valp
is a pointer to the pointer that holds the current value and
.Fa req_size
should be -1 if any size of the string is acceptable or a number larger or
equal zero if the string must have a specific size.
The function saves
the old value in the scratch area (note, that any initial value must have
been allocated by
.Xr malloc 3 ) ,
allocates a new string, copies over the new value, NUL-terminates it and
sets the new current value.
.It Fn string_commit
simply frees the saved old value in the scratch area.
.It Fn string_rollback
frees the new value, and puts back the old one.
.It Fn string_get
is used for GET or GETNEXT.
The function
.It Fn string_get_max
can be used instead of
.Fn string_get
to ensure that the returned string has a certain maximum length.
If
.Fa len
is -1, the length is computed via
.Xr strlen 3
from the current string value.
If the current value is NULL,
a OCTET STRING of zero length is returned.
.It Fn string_free
must be called if either rollback or commit fails to free the saved old value.
.El
.Pp
The following functions are used to process scalars of type IP-address:
.Bl -tag -width "XXXXXXXXX"
.It Fn ip_save
Saves the current value in the scratch area and sets the new value from
.Fa valp .
.It Fn ip_commit
Does nothing.
.It Fn ip_rollback
Restores the old IP address from the scratch area.
.It Fn ip_get
Retrieves the IP current address.
.El
.Pp
The following functions handle OID-typed variables:
.Bl -tag -width "XXXXXXXXX"
.It Fn oid_save
Saves the current value in the scratch area by allocating a
.Vt struct asn_oid
with
.Xr malloc 3
and sets the new value from
.Fa oid .
.It Fn oid_commit
Frees the old value in the scratch area.
.It Fn oid_rollback
Restores the old OID from the scratch area and frees the old OID.
.It Fn oid_get
Retrieves the OID
.El
.Ss TABLE INDEX HANDLING
The following functions help in handling table indexes:
.Bl -tag -width "XXXXXXXXX"
.It Fn index_decode
Decodes the index part of the OID.
The parameter
.Fa oid
must be a pointer to the
.Va var
field of the
.Fa value
argument of the node callback.
The
.Fa sub
argument must be the index of the start of the index in the OID (this is
the
.Fa sub
argument to the node callback).
.Fa code
is the index expression (parameter
.Fa idx
to the node callback).
These parameters are followed by parameters depending on the syntax of the index
elements as follows:
.Bl -tag -width ".It Li OCTET STRING"
.It Li INTEGER
.Vt int32_t *
expected as argument.
.It Li COUNTER64
.Vt uint64_t *
expected as argument.
Note, that this syntax is illegal for indexes.
.It Li OCTET STRING
A
.Vt u_char **
and a
.Vt size_t *
expected as arguments.
A buffer is allocated to hold the decoded string.
.It Li OID
A
.Vt struct asn_oid *
is expected as argument.
.It Li IP ADDRESS
A
.Vt u_int8_t *
expected as argument that points to a buffer of at least four byte.
.It Li COUNTER, GAUGE, TIMETICKS
A
.Vt u_int32_t
expected.
.It Li NULL
No argument expected.
.El
.It Fn index_compare
compares the current variable with an OID.
.Fa oid1
and
.Fa sub
come from the node callback arguments
.Fa value->var
and
.Fa sub
resp.
.Fa oid2
is the OID to compare to.
The function returns -1, 0, +1 when the
variable is lesser, equal, higher to the given OID.
.Fa oid2
must contain only the index part of the table column.
.It Fn index_compare_off
is equivalent to
.Fn index_compare
except that it takes an additional parameter
.Fa off
that causes it to ignore the first
.Fa off
components of both indexes.
.It Fn index_append
appends OID
.Fa src
beginning at position
.Fa sub
to
.Fa dst .
.It Fn index_append_off
appends OID
.Fa src
beginning at position
.Fa off
to
.Fa dst
beginning at position
.Fa sub
+
.Fa off .
.El
.Sh SEE ALSO
.Xr gensnmptree 1 ,
.Xr bsnmpd 1 ,
.Xr bsnmpagent 3 ,
.Xr bsnmpclient 3 ,
.Xr bsnmplib 3
.Sh STANDARDS
This implementation conforms to the applicable IETF RFCs and ITU-T
recommendations.
.Sh AUTHORS
.An Hartmut Brandt Aq harti@FreeBSD.org