Based on kernel version 3.9. Page generated on 2013-05-02 23:12 EST.
1 2 The sync patches work is based on initial patches from 3 Krisztian <hidden@balabit.hu> and others and additional patches 4 from Jamal <hadi@cyberus.ca>. 5 6 The end goal for syncing is to be able to insert attributes + generate 7 events so that the an SA can be safely moved from one machine to another 8 for HA purposes. 9 The idea is to synchronize the SA so that the takeover machine can do 10 the processing of the SA as accurate as possible if it has access to it. 11 12 We already have the ability to generate SA add/del/upd events. 13 These patches add ability to sync and have accurate lifetime byte (to 14 ensure proper decay of SAs) and replay counters to avoid replay attacks 15 with as minimal loss at failover time. 16 This way a backup stays as closely uptodate as an active member. 17 18 Because the above items change for every packet the SA receives, 19 it is possible for a lot of the events to be generated. 20 For this reason, we also add a nagle-like algorithm to restrict 21 the events. i.e we are going to set thresholds to say "let me 22 know if the replay sequence threshold is reached or 10 secs have passed" 23 These thresholds are set system-wide via sysctls or can be updated 24 per SA. 25 26 The identified items that need to be synchronized are: 27 - the lifetime byte counter 28 note that: lifetime time limit is not important if you assume the failover 29 machine is known ahead of time since the decay of the time countdown 30 is not driven by packet arrival. 31 - the replay sequence for both inbound and outbound 32 33 1) Message Structure 34 ---------------------- 35 36 nlmsghdr:aevent_id:optional-TLVs. 37 38 The netlink message types are: 39 40 XFRM_MSG_NEWAE and XFRM_MSG_GETAE. 41 42 A XFRM_MSG_GETAE does not have TLVs. 43 A XFRM_MSG_NEWAE will have at least two TLVs (as is 44 discussed further below). 45 46 aevent_id structure looks like: 47 48 struct xfrm_aevent_id { 49 struct xfrm_usersa_id sa_id; 50 xfrm_address_t saddr; 51 __u32 flags; 52 __u32 reqid; 53 }; 54 55 The unique SA is identified by the combination of xfrm_usersa_id, 56 reqid and saddr. 57 58 flags are used to indicate different things. The possible 59 flags are: 60 XFRM_AE_RTHR=1, /* replay threshold*/ 61 XFRM_AE_RVAL=2, /* replay value */ 62 XFRM_AE_LVAL=4, /* lifetime value */ 63 XFRM_AE_ETHR=8, /* expiry timer threshold */ 64 XFRM_AE_CR=16, /* Event cause is replay update */ 65 XFRM_AE_CE=32, /* Event cause is timer expiry */ 66 XFRM_AE_CU=64, /* Event cause is policy update */ 67 68 How these flags are used is dependent on the direction of the 69 message (kernel<->user) as well the cause (config, query or event). 70 This is described below in the different messages. 71 72 The pid will be set appropriately in netlink to recognize direction 73 (0 to the kernel and pid = processid that created the event 74 when going from kernel to user space) 75 76 A program needs to subscribe to multicast group XFRMNLGRP_AEVENTS 77 to get notified of these events. 78 79 2) TLVS reflect the different parameters: 80 ----------------------------------------- 81 82 a) byte value (XFRMA_LTIME_VAL) 83 This TLV carries the running/current counter for byte lifetime since 84 last event. 85 86 b)replay value (XFRMA_REPLAY_VAL) 87 This TLV carries the running/current counter for replay sequence since 88 last event. 89 90 c)replay threshold (XFRMA_REPLAY_THRESH) 91 This TLV carries the threshold being used by the kernel to trigger events 92 when the replay sequence is exceeded. 93 94 d) expiry timer (XFRMA_ETIMER_THRESH) 95 This is a timer value in milliseconds which is used as the nagle 96 value to rate limit the events. 97 98 3) Default configurations for the parameters: 99 ---------------------------------------------- 100 101 By default these events should be turned off unless there is 102 at least one listener registered to listen to the multicast 103 group XFRMNLGRP_AEVENTS. 104 105 Programs installing SAs will need to specify the two thresholds, however, 106 in order to not change existing applications such as racoon 107 we also provide default threshold values for these different parameters 108 in case they are not specified. 109 110 the two sysctls/proc entries are: 111 a) /proc/sys/net/core/sysctl_xfrm_aevent_etime 112 used to provide default values for the XFRMA_ETIMER_THRESH in incremental 113 units of time of 100ms. The default is 10 (1 second) 114 115 b) /proc/sys/net/core/sysctl_xfrm_aevent_rseqth 116 used to provide default values for XFRMA_REPLAY_THRESH parameter 117 in incremental packet count. The default is two packets. 118 119 4) Message types 120 ---------------- 121 122 a) XFRM_MSG_GETAE issued by user-->kernel. 123 XFRM_MSG_GETAE does not carry any TLVs. 124 The response is a XFRM_MSG_NEWAE which is formatted based on what 125 XFRM_MSG_GETAE queried for. 126 The response will always have XFRMA_LTIME_VAL and XFRMA_REPLAY_VAL TLVs. 127 *if XFRM_AE_RTHR flag is set, then XFRMA_REPLAY_THRESH is also retrieved 128 *if XFRM_AE_ETHR flag is set, then XFRMA_ETIMER_THRESH is also retrieved 129 130 b) XFRM_MSG_NEWAE is issued by either user space to configure 131 or kernel to announce events or respond to a XFRM_MSG_GETAE. 132 133 i) user --> kernel to configure a specific SA. 134 any of the values or threshold parameters can be updated by passing the 135 appropriate TLV. 136 A response is issued back to the sender in user space to indicate success 137 or failure. 138 In the case of success, additionally an event with 139 XFRM_MSG_NEWAE is also issued to any listeners as described in iii). 140 141 ii) kernel->user direction as a response to XFRM_MSG_GETAE 142 The response will always have XFRMA_LTIME_VAL and XFRMA_REPLAY_VAL TLVs. 143 The threshold TLVs will be included if explicitly requested in 144 the XFRM_MSG_GETAE message. 145 146 iii) kernel->user to report as event if someone sets any values or 147 thresholds for an SA using XFRM_MSG_NEWAE (as described in #i above). 148 In such a case XFRM_AE_CU flag is set to inform the user that 149 the change happened as a result of an update. 150 The message will always have XFRMA_LTIME_VAL and XFRMA_REPLAY_VAL TLVs. 151 152 iv) kernel->user to report event when replay threshold or a timeout 153 is exceeded. 154 In such a case either XFRM_AE_CR (replay exceeded) or XFRM_AE_CE (timeout 155 happened) is set to inform the user what happened. 156 Note the two flags are mutually exclusive. 157 The message will always have XFRMA_LTIME_VAL and XFRMA_REPLAY_VAL TLVs. 158 159 Exceptions to threshold settings 160 -------------------------------- 161 162 If you have an SA that is getting hit by traffic in bursts such that 163 there is a period where the timer threshold expires with no packets 164 seen, then an odd behavior is seen as follows: 165 The first packet arrival after a timer expiry will trigger a timeout 166 aevent; i.e we dont wait for a timeout period or a packet threshold 167 to be reached. This is done for simplicity and efficiency reasons. 168 169 -JHS