1 # SPDX-License-Identifier: GPL-2.0-only
3 # IP Virtual Server configuration
6 tristate "IP virtual server support"
7 depends on NET && INET && NETFILTER
8 depends on (NF_CONNTRACK || NF_CONNTRACK=n)
10 IP Virtual Server support will let you build a high-performance
11 virtual server based on cluster of two or more real servers. This
12 option must be enabled for at least one of the clustered computers
13 that will take care of intercepting incoming connections to a
14 single IP address and scheduling them to real servers.
16 Three request dispatching techniques are implemented, they are
17 virtual server via NAT, virtual server via tunneling and virtual
18 server via direct routing. The several scheduling algorithms can
19 be used to choose which server the connection is directed to,
20 thus load balancing can be achieved among the servers. For more
21 information and its administration program, please visit the
22 following URL: <http://www.linuxvirtualserver.org/>.
24 If you want to compile it in kernel, say Y. To compile it as a
25 module, choose M here. If unsure, say N.
30 bool "IPv6 support for IPVS"
31 depends on IPV6 = y || IP_VS = IPV6
34 Add IPv6 support to IPVS.
39 bool "IP virtual server debugging"
41 Say Y here if you want to get additional messages useful in
42 debugging the IP virtual server code. You can change the debug
43 level in /proc/sys/net/ipv4/vs/debug_level
46 int "IPVS connection table size (the Nth power of 2)"
50 The IPVS connection hash table uses the chaining scheme to handle
51 hash collisions. Using a big IPVS connection hash table will greatly
52 reduce conflicts when there are hundreds of thousands of connections
55 Note the table size must be power of 2. The table size will be the
56 value of 2 to the your input number power. The number to choose is
57 from 8 to 20, the default number is 12, which means the table size
58 is 4096. Don't input the number too small, otherwise you will lose
59 performance on it. You can adapt the table size yourself, according
60 to your virtual server application. It is good to set the table size
61 not far less than the number of connections per second multiplying
62 average lasting time of connection in the table. For example, your
63 virtual server gets 200 connections per second, the connection lasts
64 for 200 seconds in average in the connection table, the table size
65 should be not far less than 200x200, it is good to set the table
68 Another note that each connection occupies 128 bytes effectively and
69 each hash entry uses 8 bytes, so you can estimate how much memory is
72 You can overwrite this number setting conn_tab_bits module parameter
73 or by appending ip_vs.conn_tab_bits=? to the kernel command line
74 if IP VS was compiled built-in.
76 comment "IPVS transport protocol load balancing support"
78 config IP_VS_PROTO_TCP
79 bool "TCP load balancing support"
81 This option enables support for load balancing TCP transport
82 protocol. Say Y if unsure.
84 config IP_VS_PROTO_UDP
85 bool "UDP load balancing support"
87 This option enables support for load balancing UDP transport
88 protocol. Say Y if unsure.
90 config IP_VS_PROTO_AH_ESP
91 def_bool IP_VS_PROTO_ESP || IP_VS_PROTO_AH
93 config IP_VS_PROTO_ESP
94 bool "ESP load balancing support"
96 This option enables support for load balancing ESP (Encapsulation
97 Security Payload) transport protocol. Say Y if unsure.
100 bool "AH load balancing support"
102 This option enables support for load balancing AH (Authentication
103 Header) transport protocol. Say Y if unsure.
105 config IP_VS_PROTO_SCTP
106 bool "SCTP load balancing support"
109 This option enables support for load balancing SCTP transport
110 protocol. Say Y if unsure.
112 comment "IPVS scheduler"
115 tristate "round-robin scheduling"
117 The robin-robin scheduling algorithm simply directs network
118 connections to different real servers in a round-robin manner.
120 If you want to compile it in kernel, say Y. To compile it as a
121 module, choose M here. If unsure, say N.
124 tristate "weighted round-robin scheduling"
126 The weighted robin-robin scheduling algorithm directs network
127 connections to different real servers based on server weights
128 in a round-robin manner. Servers with higher weights receive
129 new connections first than those with less weights, and servers
130 with higher weights get more connections than those with less
131 weights and servers with equal weights get equal connections.
133 If you want to compile it in kernel, say Y. To compile it as a
134 module, choose M here. If unsure, say N.
137 tristate "least-connection scheduling"
139 The least-connection scheduling algorithm directs network
140 connections to the server with the least number of active
143 If you want to compile it in kernel, say Y. To compile it as a
144 module, choose M here. If unsure, say N.
147 tristate "weighted least-connection scheduling"
149 The weighted least-connection scheduling algorithm directs network
150 connections to the server with the least active connections
151 normalized by the server weight.
153 If you want to compile it in kernel, say Y. To compile it as a
154 module, choose M here. If unsure, say N.
157 tristate "weighted failover scheduling"
159 The weighted failover scheduling algorithm directs network
160 connections to the server with the highest weight that is
163 If you want to compile it in kernel, say Y. To compile it as a
164 module, choose M here. If unsure, say N.
167 tristate "weighted overflow scheduling"
169 The weighted overflow scheduling algorithm directs network
170 connections to the server with the highest weight that is
171 currently available and overflows to the next when active
172 connections exceed the node's weight.
174 If you want to compile it in kernel, say Y. To compile it as a
175 module, choose M here. If unsure, say N.
178 tristate "locality-based least-connection scheduling"
180 The locality-based least-connection scheduling algorithm is for
181 destination IP load balancing. It is usually used in cache cluster.
182 This algorithm usually directs packet destined for an IP address to
183 its server if the server is alive and under load. If the server is
184 overloaded (its active connection numbers is larger than its weight)
185 and there is a server in its half load, then allocate the weighted
186 least-connection server to this IP address.
188 If you want to compile it in kernel, say Y. To compile it as a
189 module, choose M here. If unsure, say N.
192 tristate "locality-based least-connection with replication scheduling"
194 The locality-based least-connection with replication scheduling
195 algorithm is also for destination IP load balancing. It is
196 usually used in cache cluster. It differs from the LBLC scheduling
197 as follows: the load balancer maintains mappings from a target
198 to a set of server nodes that can serve the target. Requests for
199 a target are assigned to the least-connection node in the target's
200 server set. If all the node in the server set are over loaded,
201 it picks up a least-connection node in the cluster and adds it
202 in the sever set for the target. If the server set has not been
203 modified for the specified time, the most loaded node is removed
204 from the server set, in order to avoid high degree of replication.
206 If you want to compile it in kernel, say Y. To compile it as a
207 module, choose M here. If unsure, say N.
210 tristate "destination hashing scheduling"
212 The destination hashing scheduling algorithm assigns network
213 connections to the servers through looking up a statically assigned
214 hash table by their destination IP addresses.
216 If you want to compile it in kernel, say Y. To compile it as a
217 module, choose M here. If unsure, say N.
220 tristate "source hashing scheduling"
222 The source hashing scheduling algorithm assigns network
223 connections to the servers through looking up a statically assigned
224 hash table by their source IP addresses.
226 If you want to compile it in kernel, say Y. To compile it as a
227 module, choose M here. If unsure, say N.
230 tristate "maglev hashing scheduling"
232 The maglev consistent hashing scheduling algorithm provides the
233 Google's Maglev hashing algorithm as a IPVS scheduler. It assigns
234 network connections to the servers through looking up a statically
235 assigned special hash table called the lookup table. Maglev hashing
236 is to assign a preference list of all the lookup table positions
239 Through this operation, The maglev hashing gives an almost equal
240 share of the lookup table to each of the destinations and provides
241 minimal disruption by using the lookup table. When the set of
242 destinations changes, a connection will likely be sent to the same
243 destination as it was before.
245 If you want to compile it in kernel, say Y. To compile it as a
246 module, choose M here. If unsure, say N.
249 tristate "shortest expected delay scheduling"
251 The shortest expected delay scheduling algorithm assigns network
252 connections to the server with the shortest expected delay. The
253 expected delay that the job will experience is (Ci + 1) / Ui if
254 sent to the ith server, in which Ci is the number of connections
255 on the ith server and Ui is the fixed service rate (weight)
258 If you want to compile it in kernel, say Y. To compile it as a
259 module, choose M here. If unsure, say N.
262 tristate "never queue scheduling"
264 The never queue scheduling algorithm adopts a two-speed model.
265 When there is an idle server available, the job will be sent to
266 the idle server, instead of waiting for a fast one. When there
267 is no idle server available, the job will be sent to the server
268 that minimize its expected delay (The Shortest Expected Delay
269 scheduling algorithm).
271 If you want to compile it in kernel, say Y. To compile it as a
272 module, choose M here. If unsure, say N.
274 comment 'IPVS SH scheduler'
276 config IP_VS_SH_TAB_BITS
277 int "IPVS source hashing table size (the Nth power of 2)"
281 The source hashing scheduler maps source IPs to destinations
282 stored in a hash table. This table is tiled by each destination
283 until all slots in the table are filled. When using weights to
284 allow destinations to receive more connections, the table is
285 tiled an amount proportional to the weights specified. The table
286 needs to be large enough to effectively fit all the destinations
287 multiplied by their respective weights.
289 comment 'IPVS MH scheduler'
291 config IP_VS_MH_TAB_INDEX
292 int "IPVS maglev hashing table index of size (the prime numbers)"
296 The maglev hashing scheduler maps source IPs to destinations
297 stored in a hash table. This table is assigned by a preference
298 list of the positions to each destination until all slots in
299 the table are filled. The index determines the prime for size of
300 the table as 251, 509, 1021, 2039, 4093, 8191, 16381, 32749,
301 65521 or 131071. When using weights to allow destinations to
302 receive more connections, the table is assigned an amount
303 proportional to the weights specified. The table needs to be large
304 enough to effectively fit all the destinations multiplied by their
307 comment 'IPVS application helper'
310 tristate "FTP protocol helper"
311 depends on IP_VS_PROTO_TCP && NF_CONNTRACK && NF_NAT && \
315 FTP is a protocol that transfers IP address and/or port number in
316 the payload. In the virtual server via Network Address Translation,
317 the IP address and port number of real servers cannot be sent to
318 clients in ftp connections directly, so FTP protocol helper is
319 required for tracking the connection and mangling it back to that of
322 If you want to compile it in kernel, say Y. To compile it as a
323 module, choose M here. If unsure, say N.
326 bool "Netfilter connection tracking"
327 depends on NF_CONNTRACK
329 The Netfilter connection tracking support allows the IPVS
330 connection state to be exported to the Netfilter framework
331 for filtering purposes.
334 tristate "SIP persistence engine"
335 depends on IP_VS_PROTO_UDP
336 depends on NF_CONNTRACK_SIP
338 Allow persistence based on the SIP Call-ID