1 1 00:00:00,630 --> 00:00:02,670 Security Data Collection. 2 2 00:00:02,670 --> 00:00:05,160 In this lesson, we're going to talk about all that data 3 3 00:00:05,160 --> 00:00:07,770 that you're collecting inside your SIEM. 4 4 00:00:07,770 --> 00:00:09,720 Now, a lot of this can become intelligence, 5 5 00:00:09,720 --> 00:00:13,110 but intelligence loses its value over time. 6 6 00:00:13,110 --> 00:00:15,030 So when you're dealing with this, you need to make sure 7 7 00:00:15,030 --> 00:00:17,310 that you're capturing and analyzing the information 8 8 00:00:17,310 --> 00:00:20,610 in real time or as close to real time as possible. 9 9 00:00:20,610 --> 00:00:21,870 The sooner you can find out 10 10 00:00:21,870 --> 00:00:23,820 about a bad guy intruding into your network, 11 11 00:00:23,820 --> 00:00:25,440 the quicker you can get them out, right? 12 12 00:00:25,440 --> 00:00:28,710 And so intelligence needs to be current and relevant. 13 13 00:00:28,710 --> 00:00:31,590 Now, as we talked about the intelligence process earlier, 14 14 00:00:31,590 --> 00:00:34,470 we talked about the fact that we have five different stages. 15 15 00:00:34,470 --> 00:00:35,970 We start out with Requirements, 16 16 00:00:35,970 --> 00:00:37,800 we move into Collection & Processing, 17 17 00:00:37,800 --> 00:00:40,980 then Analysis, then Dissemination, and then Feedback. 18 18 00:00:40,980 --> 00:00:42,120 Here on the screen, 19 19 00:00:42,120 --> 00:00:43,560 you can see the fact that I've highlighted 20 20 00:00:43,560 --> 00:00:46,740 Collection & Processing, Analysis, and Dissemination. 21 21 00:00:46,740 --> 00:00:47,760 The reason for that 22 22 00:00:47,760 --> 00:00:50,130 is this is what we're focusing on right now. 23 23 00:00:50,130 --> 00:00:51,810 When we're talking about a SIEM, 24 24 00:00:51,810 --> 00:00:53,520 this is where a SIEM operates. 25 25 00:00:53,520 --> 00:00:55,080 It helps you collect the information. 26 26 00:00:55,080 --> 00:00:57,750 It helps you process that information and normalize it. 27 27 00:00:57,750 --> 00:00:59,430 It helps you analyze that information. 28 28 00:00:59,430 --> 00:01:01,380 And then you can even run reports 29 29 00:01:01,380 --> 00:01:04,560 or send information out to others as part of dissemination. 30 30 00:01:04,560 --> 00:01:05,850 And so a SIEM really does fit 31 31 00:01:05,850 --> 00:01:08,910 into these three phases of the intelligence lifecycle. 32 32 00:01:08,910 --> 00:01:11,100 Now, one of the things your SIEMs can do for you is, 33 33 00:01:11,100 --> 00:01:12,210 you can actually configure them 34 34 00:01:12,210 --> 00:01:15,270 to automate much of the security intelligence lifecycle, 35 35 00:01:15,270 --> 00:01:16,530 especially when it comes 36 36 00:01:16,530 --> 00:01:19,110 to gathering and collecting that data. 37 37 00:01:19,110 --> 00:01:21,000 I don't want to have to go out to all these different systems 38 38 00:01:21,000 --> 00:01:22,860 across my network and grab their logs 39 39 00:01:22,860 --> 00:01:25,110 and start analyzing them, but using a SIEM, 40 40 00:01:25,110 --> 00:01:27,870 I can make all those systems feed me that data, 41 41 00:01:27,870 --> 00:01:29,730 and it can go into that central repository 42 42 00:01:29,730 --> 00:01:31,650 that we can later analyze. 43 43 00:01:31,650 --> 00:01:33,540 Now, one of the big things you have to consider 44 44 00:01:33,540 --> 00:01:35,160 when you're configuring your SIEMs is, 45 45 00:01:35,160 --> 00:01:36,690 what do you want to collect? 46 46 00:01:36,690 --> 00:01:38,190 Because some people have a tendency 47 47 00:01:38,190 --> 00:01:39,780 to just try to collect everything, 48 48 00:01:39,780 --> 00:01:41,940 they dump all the data into the system. 49 49 00:01:41,940 --> 00:01:43,140 But the problem with that 50 50 00:01:43,140 --> 00:01:45,570 is it can end up overloading your system. 51 51 00:01:45,570 --> 00:01:47,220 All this data has to be stored, 52 52 00:01:47,220 --> 00:01:49,440 it has to be processed, and it has to be normalized. 53 53 00:01:49,440 --> 00:01:51,870 And if I'm sending it in with millions of endpoints, 54 54 00:01:51,870 --> 00:01:54,120 that can really quickly overwhelm my systems. 55 55 00:01:54,120 --> 00:01:56,460 So instead, you should spend some time upfront 56 56 00:01:56,460 --> 00:01:58,950 doing your planning based on the requirements 57 57 00:01:58,950 --> 00:02:01,170 and determining exactly what you need to collect. 58 58 00:02:01,170 --> 00:02:04,080 Remember, while your SIEM could collect all the logs 59 59 00:02:04,080 --> 00:02:06,900 across all of your systems, this isn't a good idea. 60 60 00:02:06,900 --> 00:02:09,270 Instead, you need to configure your SIEM to focus 61 61 00:02:09,270 --> 00:02:12,510 on the events related to the things that you need to know. 62 62 00:02:12,510 --> 00:02:13,830 Not everything is important, 63 63 00:02:13,830 --> 00:02:16,980 so you need to identify what is and collect on that. 64 64 00:02:16,980 --> 00:02:18,930 Now, one of the biggest features of a SIEM 65 65 00:02:18,930 --> 00:02:20,670 is the ability to process data 66 66 00:02:20,670 --> 00:02:23,250 and then look for different trends and alert on those. 67 67 00:02:23,250 --> 00:02:24,990 Now, just like all alerting systems, 68 68 00:02:24,990 --> 00:02:26,430 it does suffer from the problems 69 69 00:02:26,430 --> 00:02:28,710 of false positives and false negatives. 70 70 00:02:28,710 --> 00:02:30,870 When we talk about the problem with false negatives, 71 71 00:02:30,870 --> 00:02:32,730 this is when security administrators are exposed 72 72 00:02:32,730 --> 00:02:34,560 to a threat without being aware of them 73 73 00:02:34,560 --> 00:02:37,710 because your system falsely categorized it as negative 74 74 00:02:37,710 --> 00:02:40,200 instead of alerting that there was something bad there. 75 75 00:02:40,200 --> 00:02:41,910 Now, on the other hand, we also have issues 76 76 00:02:41,910 --> 00:02:43,380 when we have false positives. 77 77 00:02:43,380 --> 00:02:45,300 If our system starts having a lot of positives 78 78 00:02:45,300 --> 00:02:46,440 but they aren't real, 79 79 00:02:46,440 --> 00:02:48,930 we're going to overwhelm our analysis and response resources 80 80 00:02:48,930 --> 00:02:50,880 because some person has to look at that 81 81 00:02:50,880 --> 00:02:52,200 and analyze it and determine, 82 82 00:02:52,200 --> 00:02:54,030 was there really an event that happened? 83 83 00:02:54,030 --> 00:02:55,800 So we want to make sure we are tuning our systems 84 84 00:02:55,800 --> 00:02:56,970 to make sure that we don't have 85 85 00:02:56,970 --> 00:02:59,520 a lot of false negatives or false positives. 86 86 00:02:59,520 --> 00:03:02,610 To help us do that, we develop what's called a use case. 87 87 00:03:02,610 --> 00:03:03,960 By developing use cases, 88 88 00:03:03,960 --> 00:03:07,050 we can mitigate the risk of these false indicators. 89 89 00:03:07,050 --> 00:03:08,520 Now, when I talk about a use case, 90 90 00:03:08,520 --> 00:03:11,250 this is a specific condition that should be reported, 91 91 00:03:11,250 --> 00:03:13,050 such as a suspicious log on, 92 92 00:03:13,050 --> 00:03:15,270 or a process executing from a temporary directory, 93 93 00:03:15,270 --> 00:03:16,740 or something like that. 94 94 00:03:16,740 --> 00:03:18,990 Essentially, we want to think about what is this bad thing 95 95 00:03:18,990 --> 00:03:20,280 that we want to collect on? 96 96 00:03:20,280 --> 00:03:22,470 And then we develop this use case around it. 97 97 00:03:22,470 --> 00:03:23,580 Based on that use case, 98 98 00:03:23,580 --> 00:03:26,970 we can then configure our SIEM to collect the relevant data. 99 99 00:03:26,970 --> 00:03:28,470 Now, what we want to do here is essentially 100 100 00:03:28,470 --> 00:03:30,960 develop a template for each of these use cases. 101 101 00:03:30,960 --> 00:03:32,160 And as we do that, 102 102 00:03:32,160 --> 00:03:34,470 they're going to contain a couple of different things. 103 103 00:03:34,470 --> 00:03:36,360 We're going to contain things like the data sources 104 104 00:03:36,360 --> 00:03:38,580 with the indicators that we want to collect on. 105 105 00:03:38,580 --> 00:03:40,080 It's going to contain the query strings 106 106 00:03:40,080 --> 00:03:42,120 that we're going to use to correlate those different indicators 107 107 00:03:42,120 --> 00:03:43,620 across different systems. 108 108 00:03:43,620 --> 00:03:45,060 We want to make sure we have the actions 109 109 00:03:45,060 --> 00:03:47,040 that are going to occur when the event is triggered. 110 110 00:03:47,040 --> 00:03:49,080 Essentially, what are we going to do to respond 111 111 00:03:49,080 --> 00:03:50,790 when we see this bad thing happen? 112 112 00:03:50,790 --> 00:03:52,260 And then by having those three things 113 113 00:03:52,260 --> 00:03:53,940 as part of our use case template, 114 114 00:03:53,940 --> 00:03:55,800 that tells us that this set 115 115 00:03:55,800 --> 00:03:57,930 is what is known as this bad thing. 116 116 00:03:57,930 --> 00:03:59,850 And so eventually, we would write a rule 117 117 00:03:59,850 --> 00:04:01,530 or a query to be able to identify 118 118 00:04:01,530 --> 00:04:03,960 all of those things across our systems. 119 119 00:04:03,960 --> 00:04:06,390 In addition to providing those three things in the use case, 120 120 00:04:06,390 --> 00:04:08,520 we also need to make sure that each use case 121 121 00:04:08,520 --> 00:04:11,280 captures the 5Ws when we're dealing with an event. 122 122 00:04:11,280 --> 00:04:14,250 This would be things like when, when did this event start? 123 123 00:04:14,250 --> 00:04:17,040 And when did this event end, if it's already ended? 124 124 00:04:17,040 --> 00:04:18,810 We also want to figure out who, 125 125 00:04:18,810 --> 00:04:22,500 who was involved in this event, which user or which system? 126 126 00:04:22,500 --> 00:04:24,960 And then we want to figure out what, what happened 127 127 00:04:24,960 --> 00:04:27,750 and what is the specific details of this event? 128 128 00:04:27,750 --> 00:04:29,610 Essentially, did somebody try to run a program 129 129 00:04:29,610 --> 00:04:30,810 and there was malware in it? 130 130 00:04:30,810 --> 00:04:32,880 Did somebody try to attack our network from the outside? 131 131 00:04:32,880 --> 00:04:35,790 What happened? We need those specific details. 132 132 00:04:35,790 --> 00:04:39,030 Then we go and figure out where, where did the event happen? 133 133 00:04:39,030 --> 00:04:43,020 Was it on a host, a server, a file system, the network? 134 134 00:04:43,020 --> 00:04:44,730 Where is this issue? 135 135 00:04:44,730 --> 00:04:46,650 And then we want to also figure out where, 136 136 00:04:46,650 --> 00:04:48,660 where did the event originate from? 137 137 00:04:48,660 --> 00:04:51,360 Did it come from the inside 'cause it was an insider threat? 138 138 00:04:51,360 --> 00:04:52,350 Did it come from the outside 139 139 00:04:52,350 --> 00:04:54,030 because there was an external hacker? 140 140 00:04:54,030 --> 00:04:55,410 This is an important piece of information 141 141 00:04:55,410 --> 00:04:56,460 for us to know too. 142 142 00:04:56,460 --> 00:04:58,110 So by knowing those 5Ws, 143 143 00:04:58,110 --> 00:04:59,580 we're going to be able to better understand 144 144 00:04:59,580 --> 00:05:01,980 what happened and then how we can respond to it.