1 /proc/sys/net/ipv4/vs/* Variables:
6 It sets the always mode drop rate, which is used in the mode 3
7 of the drop_rate defense.
12 It sets the available memory threshold (in pages), which is
13 used in the automatic modes of defense. When there is no
14 enough available memory, the respective strategy will be
15 enabled and the variable is automatically set to 2, otherwise
16 the strategy is disabled and the variable is set to 1.
19 0 - disabled (default)
22 If set, disable the director function while the server is
23 in backup mode to avoid packet loops for DR/TUN methods.
26 0 - disabled (default)
29 If set, maintain connection tracking entries for
30 connections handled by IPVS.
32 This should be enabled if connections handled by IPVS are to be
33 also handled by stateful firewall rules. That is, iptables rules
34 that make use of connection tracking. It is a performance
35 optimisation to disable this setting otherwise.
37 Connections handled by the IPVS FTP application module
38 will have connection tracking entries regardless of this setting.
40 Only available when IPVS is compiled with CONFIG_IP_VS_NFCT enabled.
42 cache_bypass - BOOLEAN
43 0 - disabled (default)
46 If it is enabled, forward packets to the original destination
47 directly when no cache server is available and destination
48 address is not local (iph->daddr is RTN_UNICAST). It is mostly
49 used in transparent web cache cluster.
52 0 - transmission error messages (default)
53 1 - non-fatal error messages
59 7 - connection new/expire, lookup and synchronization
61 9 - binding destination, template checks and applications
62 10 - IPVS packet transmission
63 11 - IPVS packet handling (ip_vs_in/ip_vs_out)
64 12 or more - packet traversal
66 Only available when IPVS is compiled with CONFIG_IP_VS_DEBUG enabled.
68 Higher debugging levels include the messages for lower debugging
69 levels, so setting debug level 2, includes level 0, 1 and 2
70 messages. Thus, logging becomes more and more verbose the higher
74 0 - disabled (default)
76 The drop_entry defense is to randomly drop entries in the
77 connection hash table, just in order to collect back some
78 memory for new connections. In the current code, the
79 drop_entry procedure can be activated every second, then it
80 randomly scans 1/32 of the whole and drops entries that are in
81 the SYN-RECV/SYNACK state, which should be effective against
84 The valid values of drop_entry are from 0 to 3, where 0 means
85 that this strategy is always disabled, 1 and 2 mean automatic
86 modes (when there is no enough available memory, the strategy
87 is enabled and the variable is automatically set to 2,
88 otherwise the strategy is disabled and the variable is set to
89 1), and 3 means that that the strategy is always enabled.
92 0 - disabled (default)
94 The drop_packet defense is designed to drop 1/rate packets
95 before forwarding them to real servers. If the rate is 1, then
96 drop all the incoming packets.
98 The value definition is the same as that of the drop_entry. In
99 the automatic mode, the rate is determined by the follow
100 formula: rate = amemthresh / (amemthresh - available_memory)
101 when available memory is less than the available memory
102 threshold. When the mode 3 is set, the always mode drop rate
103 is controlled by the /proc/sys/net/ipv4/vs/am_droprate.
105 expire_nodest_conn - BOOLEAN
106 0 - disabled (default)
109 The default value is 0, the load balancer will silently drop
110 packets when its destination server is not available. It may
111 be useful, when user-space monitoring program deletes the
112 destination server (because of server overload or wrong
113 detection) and add back the server later, and the connections
114 to the server can continue.
116 If this feature is enabled, the load balancer will expire the
117 connection immediately when a packet arrives and its
118 destination server is not available, then the client program
119 will be notified that the connection is closed. This is
120 equivalent to the feature some people requires to flush
121 connections when its destination is not available.
123 expire_quiescent_template - BOOLEAN
124 0 - disabled (default)
127 When set to a non-zero value, the load balancer will expire
128 persistent templates when the destination server is quiescent.
129 This may be useful, when a user makes a destination server
130 quiescent by setting its weight to 0 and it is desired that
131 subsequent otherwise persistent connections are sent to a
132 different destination server. By default new persistent
133 connections are allowed to quiescent destination servers.
135 If this feature is enabled, the load balancer will expire the
136 persistence template if it is to be used to schedule a new
137 connection and the destination server is quiescent.
139 nat_icmp_send - BOOLEAN
140 0 - disabled (default)
143 It controls sending icmp error messages (ICMP_DEST_UNREACH)
144 for VS/NAT when the load balancer receives packets from real
145 servers but the connection entries don't exist.
148 0 - disabled (default)
150 The secure_tcp defense is to use a more complicated TCP state
151 transition table. For VS/NAT, it also delays entering the
152 TCP ESTABLISHED state until the three way handshake is completed.
154 The value definition is the same as that of drop_entry and
157 sync_threshold - INTEGER
160 It sets synchronization threshold, which is the minimum number
161 of incoming packets that a connection needs to receive before
162 the connection will be synchronized. A connection will be
163 synchronized, every time the number of its incoming packets
164 modulus 50 equals the threshold. The range of the threshold is
167 snat_reroute - BOOLEAN
169 not 0 - enabled (default)
171 If enabled, recalculate the route of SNATed packets from
172 realservers so that they are routed as if they originate from the
173 director. Otherwise they are routed as if they are forwarded by the
176 If policy routing is in effect then it is possible that the route
177 of a packet originating from a director is routed differently to a
178 packet being forwarded by the director.
180 If policy routing is not in effect then the recalculated route will
181 always be the same as the original route so it is an optimisation
182 to disable snat_reroute and avoid the recalculation.
184 sync_persist_mode - INTEGER
187 Controls the synchronisation of connections when using persistence
189 0: All types of connections are synchronised
190 1: Attempt to reduce the synchronisation traffic depending on
191 the connection type. For persistent services avoid synchronisation
192 for normal connections, do it only for persistence templates.
193 In such case, for TCP and SCTP it may need enabling sloppy_tcp and
194 sloppy_sctp flags on backup servers. For non-persistent services
195 such optimization is not applied, mode 0 is assumed.
197 sync_version - INTEGER
200 The version of the synchronisation protocol used when sending
201 synchronisation messages.
203 0 selects the original synchronisation protocol (version 0). This
204 should be used when sending synchronisation messages to a legacy
205 system that only understands the original synchronisation protocol.
207 1 selects the current synchronisation protocol (version 1). This
208 should be used where possible.
210 Kernels with this sync_version entry are able to receive messages
211 of both version 1 and version 2 of the synchronisation protocol.