1snmp_index(3) Erlang Module Definition snmp_index(3)
2
6 snmp_index - Abstract Data Type for SNMP Indexing
7
9 The module snmp_index implements an Abstract Data Type (ADT) for an
10 SNMP index structure for SNMP tables. It is implemented as an ets table
11 of the ordered_set data-type, which means that all operations are O(log
12 n). In the table, the key is an ASN.1 OBJECT IDENTIFIER.
13 This index is used to separate the implementation of the SNMP ordering
15 from the actual implementation of the table. The SNMP ordering, that is
16 implementation of GET NEXT, is implemented in this module.
17 For example, suppose there is an SNMP table, which is best implemented
19 in Erlang as one process per SNMP table row. Suppose further that the
20 INDEX in the SNMP table is an OCTET STRING. The index structure would
21 be created as follows:
22 snmp_index:new(string)
24 For each new process we create, we insert an item in an snmp_index
27 structure:
28 new_process(Name, SnmpIndex) ->
30 Pid = start_process(),
31 NewSnmpIndex =
32 snmp_index:insert(SnmpIndex, Name, Pid),
33 <...>
34 With this structure, we can now map an OBJECT IDENTIFIER in e.g. a GET
37 NEXT request, to the correct process:
38 get_next_pid(Oid, SnmpIndex) ->
40 {ok, {_, Pid}} = snmp_index:get_next(SnmpIndex, Oid),
41 Pid.
42
45 The following data types are used in the functions below:
46 * index()
48 * oid() = [byte()]
50 * key_types = type_spec() | {type_spec(), type_spec(), ...}
52 * type_spec() = fix_string | string | integer
54 * key() = key_spec() | {key_spec(), key_spec(), ...}
56 * key_spec() = string() | integer()
58 The index() type denotes an snmp index structure.
60 The oid() type is used to represent an ASN.1 OBJECT IDENTIFIER.
62 The key_types() type is used when creating the index structure, and the
64 key() type is used when inserting and deleting items from the struc‐
65 ture.
66 The key_types() type defines the types of the SNMP INDEX columns for
68 the table. If the table has one single INDEX column, this type should
69 be a single atom, but if the table has multiple INDEX columns, it
70 should be a tuple with atoms.
71 If the INDEX column is of type INTEGER, or derived from INTEGER, the
73 corresponding type should be integer. If it is a variable length type
74 (e.g. OBJECT IDENTIFIER, OCTET STRING), the corresponding type should
75 be string. Finally, if the type is of variable length, but with a fixed
76 size restriction (e.g. IpAddress), the corresponding type should be
77 fix_string.
78 For example, if the SNMP table has two INDEX columns, the first one an
80 OCTET STRING with size 2, and the second one an OBJECT IDENTIFER, the
81 corresponding key_types parameter would be {fix_string, string}.
82 The key() type correlates to the key_types() type. If the key_types()
84 is a single atom, the corresponding key() is a single type as well, but
85 if the key_types() is a tuple, key must be a tuple of the same size.
86 In the example above, valid keys could be {"hi", "mom"} and {"no",
88 "thanks"}, whereas "hi", {"hi", 42} and {"hello", "there"} would be
89 invalid.
90 Warning:
92 All API functions that update the index return a NewIndex term. This is
94 for backward compatibility with a previous implementation that used a
95 B+ tree written purely in Erlang for the index. The NewIndex return
96 value can now be ignored. The return value is now the unchanged table
97 identifier for the ets table.
98 The implementation using ets tables introduces a semantic incompatibil‐
100 ity with older implementations. In those older implementations, using
101 pure Erlang terms, the index was garbage collected like any other
102 Erlang term and did not have to be deleted when discarded. An ets table
103 is deleted only when the process creating it explicitly deletes it or
104 when the creating process terminates.
105 A new interface delete/1 is now added to handle the case when a process
107 wants to discard an index table (i.e. to build a completely new). Any
108 application using transient snmp indexes has to be modified to handle
109 this.
110 As an snmp adaption usually keeps the index for the whole of the sys‐
112 tems lifetime, this is rarely a problem.
113
116 delete(Index) -> true
117 Types:
119 Index = NewIndex = index()
121 Key = key()
122 Deletes a complete index structure (i.e. the ets table holding
124 the index). The index can no longer be referenced after this
125 call. See the warning note above.
126 delete(Index, Key) -> NewIndex
128 Types:
130 Index = NewIndex = index()
132 Key = key()
133 Deletes a key and its value from the index structure. Returns a
135 new structure.
136 get(Index, KeyOid) -> {ok, {KeyOid, Value}} | undefined
138 Types:
140 Index = index()
142 KeyOid = oid()
143 Value = term()
144 Gets the item with key KeyOid. Could be used from within an SNMP
146 instrumentation function.
147 get_last(Index) -> {ok, {KeyOid, Value}} | undefined
149 Types:
151 Index = index()
153 KeyOid = oid()
154 Value = term()
155 Gets the last item in the index structure.
157 get_next(Index, KeyOid) -> {ok, {NextKeyOid, Value}} | undefined
159 Types:
161 Index = index()
163 KeyOid = NextKeyOid = oid()
164 Value = term()
165 Gets the next item in the SNMP lexicographic ordering, after
167 KeyOid in the index structure. KeyOid does not have to refer to
168 an existing item in the index.
169 insert(Index, Key, Value) -> NewIndex
171 Types:
173 Index = NewIndex = index()
175 Key = key()
176 Value = term()
177 Inserts a new key value tuple into the index structure. If an
179 item with the same key already exists, the new Value overwrites
180 the old value.
181 key_to_oid(Index, Key) -> KeyOid
183 Types:
185 Index = index()
187 Key = key()
188 KeyOid = NextKeyOid = oid()
189 Converts Key to an OBJECT IDENTIFIER.
191 new(KeyTypes) -> Index
193 Types:
195 KeyTypes = key_types()
197 Index = index()
198 Creates a new snmp index structure. The key_types() type is
200 described above.
201Ericsson AB snmp 5.5 snmp_index(3)