Truth in IT
    • Sign In
    • Register
        • Videos
        • Channels
        • Pages
        • Galleries
        • News
        • Events
        • All
Truth in IT Truth in IT
  • Data Management ▼
    • Converged Infrastructure
    • DevOps
    • Networking
    • Storage
    • Virtualization
  • Cybersecurity ▼
    • Application Security
    • Backup & Recovery
    • Data Security
    • Identity & Access Management (IAM)
    • Zero Trust
    • Compliance & GRC
    • Endpoint Security
  • Cloud ▼
    • Hybrid Cloud
    • Private Cloud
    • Public Cloud
  • Webinar Library
  • TiPs
  • DRAW
  • CPE

Active Directory Logon Auditing Deep Dive

Manage Engine
11/16/2026
0 (0%)
Share
  • Comments
  • Download
  • Transcript
Report Like Favorite
  • Share/Embed
  • Email
Link
Embed

Transcript


A very warm welcome to all of you. This is going to be an interesting webinar. And this webinar is all about going for a deep dive on Active Directory Auditing. Well, in particular, we're going to have a clean exploration on doing a perfect audit on logon activities. Because trust me, when we start tweaking and troubleshooting logon scenarios, and that is where we would get stuck at some point, and maybe at the beginning or in the middle or maybe at the last before we get to a conclusion, all right? This is the ultimate point is doing a logon audit is not an easy job for administrators, security officers, or even any auditors, all right? So that is why we really thought of this could be a cool idea if we come up with a webinar that would showcase, that would really discuss very important points on logon auditing. And well, you're, okay, so let's jump on real quickly. This is going to be a webinar for Active Directory Logon Auditing. We will be discussing each and every attribute that's significantly going to benefit the administrators all around the world. And so my name is Bruce McLean, and I'm a Active Directory Consultant working with ManageEngine for more than 80 years, and I do have some extensive experience. And usually I do present in the seminars and conferences. And my forte is, yep, Active Directory and IT security. So without any further delay, let's jump on to today's agenda. Number one, we're going to discuss why do we need to audit the logon activity? You know what? This is really an essential cog in an auditing machine, especially in a user behavior analytics machine. It is an essential cog, really. So that's why we're probably going to need to do an audit, you know, to monitor the user logon activity, right? Okay, and then, you know, when we start digging deeper for the logon activity, okay, there are two important variables, confusing variables, really. You know what? Observing the logon activity and monitoring the logon activities for domain account, a user domain account, and local accounts on computer. When I say domain accounts, yeah, obviously, everyone knows that's an Active Directory account. And simultaneously, we're also going to need to observe and monitor the logon activities that happened for a local account. Let's think about in this scenario. We have, yep, workstations. We're going to have servers. And we may have servers with business-critical information, like web servers, application servers. And they have local accounts on them. And if anyone, well, yeah, maybe a perpetrator who's trying to sneak into and who somehow he found the information, maybe through footprinting or social engineering, he probably have the data of what are all the web servers you've got in your infrastructure, and he just got into it. Instead of using an Active Directory account, he got himself as a member of local administrator's account, or he probably have created, you know, a backdoor account into that machine. He completely owns that machine, right, in one way or the other. So that is why auditing the logon activity for both the domain accounts and the local accounts is of paramount importance. And how are we going to do that? All right, that is why the next bulletin point that we're going to observe, what is the difference between the authentication and logon? Yeah, they may sound similar, but they are different clinically, right? And also, we're going to find out which entity is going to be a great help in troubleshooting scenarios, logon scenarios to perform an efficient logon audit. And also, we're going to discuss how the logon events is really going to be helpful for detecting attacks, right? All right, let's jump on to another one. All right, so before we go further, I heard that there was a bit of problem with screen sharing. And yeah, well, let me show you the quick peek of the agenda that we're going to discuss today. We will be discussing on why do we need auditing? Why do we need to use a logon auditing in the first place? And we're also going to need to find out how we're going to administer or manage or perform the audit, the logon audit for domain accounts and the local accounts. And so this is going to be today's agenda. And let's discuss why do we need a logon auditing? Considering today's persistent attacks. Okay, the scenario is very simple, okay? Let's say I work for a company. Obviously, I do. So I work for my company. I have a laptop. I have a computer. I have a domain account. And I come in, I log in. I just type in my username and password, and I log into my computer. I do my job. I do my daily routine, all right? Of course, that is an event, right? A logon event. Maybe I can prepare my workstation, or my administrators would have prepared the domain controller in such a way that whatever I do when I come into the domain, okay, well, starting from the way that I get in, which means a logon. So that is a normal event, okay? All right, let's talk about another scenario with regards to the same event. So a rogue employee who's going to get out of the job, well, who's going to be terminated, or maybe an ex-employee, right, who still has the credentials for his account, and obviously, he's not in the job anymore, or maybe he's going to be out of the job very sooner, but he logs into his computer, his workstation, or maybe through a remote desktop services, through VPN. But the point is, he still has got a hold of his credentials, right? Okay, that is an event. Okay, his intentions may be malicious, maybe not. We never know that, right? We can't really come to a conclusion at this point. So it could be in between a normal event and a serious attack. And finally, okay, a perpetrator, maybe through phishing, through malware injection, when he gains a little bit of access to any of a compromised computer and has a hold of an entity, a user, he probably going to show his true colors, okay, by doing a nasty job, okay? But in order for him to do that, in order for an attacker to do that, he's probably have to have the hold of credentials. So he comes in, he logs in with the credentials. He may have gained access to the local hashes, okay? And obviously, Windows is still storing the last 10 hash values for an account. It could be a local account, it could be an Active Directory account. What if, what if in the computer which the attacker has the access to, in that computer, maybe some point before, the administrator, the domain administrator have logged into that computer, maybe for patching or to help the end user in a maybe service pack or some serious problem the user had before. The system administrator's password hashes would also be stored in the local SAM, right? The database. And that can be easily exfiltrated by the attacker. Well, there are a lot of tools like PowerSupply, right? Mimikatz, that's a powerful tool, actually. You know what? Attackers would use, obviously, those tools to get a hold of the hashed values of any user who have previously logged into that computer. You know, this is called pass the hash attack. And, but in order for the attacker to do that, he needs to get into that computer in the first place. He needs to have a valid domain credentials. By then, he can just keep on, he can keep on jump onto different computers. One, two, three, four. And finally, you know, he may, he would obviously gonna have the privilege of getting the domain admin's credentials at some point. This could be a slow and steady attack. This could be a very fast attack. We never know until we phase one. The severity may go very higher at any point. So a logon event, that is why we need to audit. We really need to monitor the logon activities of anybody in the Active Directory infrastructure. Okay, and then we also need a solution to comprehensively manipulate the logon events that are being captured from any endpoints, domain controllers, and that is why we are here, by the way. All right, and now, let's talk about what are all the possible variables that would come into the picture when it comes to logon auditing. All right, well, we're gonna discuss two kinds of accounts, as we know that. So we're gonna discuss, we know there's gonna be domain accounts in the Active Directory, and there are significant approaches, okay, to shadow the domain accounts and get into the domain, infiltrate, and data exfiltration can also happen here. So observing the domain accounts, how to perform, how to audit the logon activities of a domain account. And there's another variable, another type of account is local accounts on a computer. So, of course, Windows has an option to create local accounts besides the Active Directory domain account. So we're also gonna take a look at what those local accounts are and how we can perform the audit for the logon activities for those domain accounts. And not only those are the two important variables, but we also have what type of logon that could be, all right? So, for example, let's talk about a previous example that I told you before. Okay, I have a domain user account. I log into my computer. So when I try to log in, when I type in my password, yep, I'm physically in front of my computer. I'm gonna type in my username and password, and that would contact the nearby domain controller. So the domain controller is the authority for validating my credentials just to check, okay, if I was entering the right password or the wrong password, right? So that's the interactive logon activity. But not to the domain controller. Well, to my workstation, to my machine, or to the computer in which I'm logging in, I'm trying to enter in. So for that workstation, it's an interactive activity. All right, that's an example, perfect example. And there's also a way, not one way, there are more significant ways that logon can happen in non-interactive, what also known as the network logon. How about it? Okay, I'm, or after I logged into my computer, I'm trying to access a file share, and I'm trying to map the drive, all right? I have a service running on my computer, yeah, which does some good job in the domain for the productivity, right? And I have configured, or maybe my system administrator, okay, has configured a domain account to run the service. Well, that is a service logon, a non-interactive logon. So there are different ways an account can log in, or can gain access to the computers. So we're gonna discuss on those two variables, okay? And how we're gonna perform the audit, or how we're gonna administer, okay? How we're gonna administer or observe the behavior of, or how to find out the difference between logon activity for a domain account and local account. All right, I just have an illustration pictured just to denote what's the exact difference between the domain accounts or the local accounts. So as given right on the screen here, the domain accounts is gonna be obviously in the active directory. And yep, they have access to the domain resources. All resources in the domain, they will have access to. Based on the access control, based on the permissions held by a domain account, they're gonna have a, yeah, they're gonna have a pass-through on each resources, right? Okay, and local accounts, where do they exist? What is their boundary? Okay, obviously their boundary is gonna be the computer where the accounts exist. I have a computer, so we have a server. Yeah, it could be a server, it could be a workstation, but definitely not the domain controller. Yep, the domain controller, yeah, it just has the active directory and not any local accounts. So for my workstation or a server, a member server may have, obviously will have local accounts they defined on the computers themselves. And yep, for all those accounts, local accounts reside on a computer, their boundary is just on the computer themselves. It can't really have, practically have access to, technically, which is possible, but still the boundary is just within the computer itself. And that is why Microsoft has significant approach to turn on or to pave a way for auditing both the logon activity for accounts in the domain and also accounts that exist locally on a computer. All right, and now how are we gonna perform an audit? How Microsoft is gonna help us with. Has Microsoft been helpful at all times? I would say pretty much, yeah. But only with a couple of, you know what, confusions, I can say. In terms of turning on the auditing, yeah, we will still have a catch to take care of that, right? And now let's talk about the logon activity that happen for both the domain accounts and the local accounts. And yep, I'm gonna take the same example. I log into my computer with my username and password. That is going to be validated by one of the DCs in the domain, all right? Okay, so according to my domain controller, it knows me as an Active Directory user because I try to log in with my domain account and I type in the password. All the job of the DC at this point of time is to verify whether it can allow me into the network or not. All right, if my password is right, if the password that I entered was right, okay, I will be in. If not, I'll be out, right? Okay, that is the authentication that is going to happen on the domain controller. It's an authentication event that will be triggered on the domain controller. Not just one, all the domain controllers. That's why the contextually distinct, you know what, both actions. However, the same time, you know, the machine that I use to enter my username and password, though I use my domain account, where I'm entering my username and password, on my workstation or maybe on a member server. So when I type in my username and password, according to that machine, in that computer standpoint, that's being referred as a logon, a local activity that's happening on that computer. So this is what exact difference between when we say an authentication activity happened and a logon activity happened. So we probably gonna need to audit and monitor both user authentication and also user logon. All right, so when it comes to authentication and logon, there are, yep, known famous protocols that are being used, Kerberos and NTLM. And obviously we're not gonna go for a deep dive on how the protocol works. And we know for a fact, NTLM, it's kind of pretty old. Yep, not pretty kind, it's pretty old. So is a Kerberos, well, according to me. So, but Kerberos, a bit safer than NTLM. So because NTLM may pave a way for past the hash attack, it doesn't mean Kerberos can be, you know what, attacked, of course. For Kerberos, they're passed a ticket attack. We can intercept the tickets transaction that's happening between the client and the server. And we can do, an attacker can gain access at any point. So if NTLM has a challenge, negotiate response mechanism, and obviously Kerberos is gonna have a five step, five step authentication mechanism to allow a domain user in. Okay, and obviously for a local account to log on, to observe the log on activity for a local account, how they are going to get in, how they are going to log in. Let's talk about it. Okay, I have an illustration right in front of my computer, a picture which describes how a log on actually happens and in a differential view, right? Okay, and obviously, you know, nowadays, Windows 8, 8.1, 10, 2012. Now Microsoft, they probably removed the option to list the number of domains that can be contacted. All right, but right now, what we can do in order for us to enter the domain name and credentials, all right, I just have to type in my domain name slash my username and my password. That refers to domain log on. Okay, what if I wanted to log into that computer with the local account? All right, I just got to type in the computer name in front of my username and then type in my password that would allow me just within that computer and the local log on. What's the protocol being used for the local log on? Yeah, it's, anyway, all my hashes, my passwords information, hashed information is going to be available in the local SAM, Security Accounts Manager database on that machine. So the challenge response negotiation, it happens within that computer itself. That's the NTLM protocol. And for my domain authentication, Kerberos would come into the picture. The point is there are significant attack vectors that can be derived by a perpetrator for both authentication and log on activity, pretty much on log on activity. And the way we look out for an information or a credibility to troubleshoot or to stop the possible attack, we're going to be using majorly the log on events. Yeah, I'm sure the authentication even has some significance for the audit trail, for the accumulation, consolidation of events from all the DCs and pretty much, yep. But still in an attack standpoint, we're going to be utilizing, leveraging a lot of attributes that come out of the log on event. So in other words, we will be, or we should be in a position to capture the log on events from each and every computer in the network. I know that's going to be a tiresome job. We need a perfect solution for that. And that's why we have AD audit plus a solution, a perfect solution for that actually. Yeah, I'm sure we will. We will be looking at the options, available options, how to do an efficient audit through AD audit plus as well. All right, and now this is another way. You know, the previous illustration, the picture, it illustrated the way to log on interactively on any computer. And this illustration, it explains how to perform a network log on. Well, this is not just the only way, all right? I just took an example to map a drive. Yeah, I'm trying to map a drive from my workstation. I just got into my workstation, but I'm still trying to connect to my file server and map a drive. All right, I just, when I type in my domain username and the password, I get in, okay. That will be stored in the memory, right? If any perpetrator just got intercepted into the transaction, he can still gain access to my credentials, right? Okay, so this is a way, is another way to have a non-interactive log on to happen. All right, and then how to perform or enable audit for both authentication and log on activity. All right, you know, when it comes to log on, we found that there are two different kinds of log on, right, interactive or remote interactive or the network log on. Here, the credentials entered at just one time, but separate log on for each computer access, right? In other words, I log into my computer and even in the case of the mapped drive, all right, I'm trying to connect to a remote computer, maybe a file share, and I'm trying to map the drive on that file server or any server to my computer. For that, I need to enter my credentials and assuming that I do have some permission over that drive, so it's being mapped right now to my computer. So according to the remote computer, right? According to my remote computer, it is a local log on attempt, right? Okay, so that is what it is being illustrated right about. So we're gonna be able to monitor both authentication activity and log on activity. So I've got some event IDs in the picture that would really help us with doing an efficient audit on authentication on the DMAIC controller. All right, you know what? I'm gonna probably show how to enable audit on the DMAIC controllers, how to prepare the DCs to track, to monitor the authentication activities, right? Okay, let's jump on quickly to my DMAIC controller I've got in my environment right here. Just to save time, I've already opened the group policy management console. Right, okay, I'm trying to enable the policies on the default domain controllers policy GPL, which is linked to the domain controllers OU in my Active Directory domain. You know what, Microsoft has named poorly that this is where confusion comes in. Remember, I was telling you about what kind of confusion that we would get, right? Okay, here I got a category, an audit category called account log on. Well, underneath that category, I have subcategories, audit subcategory. So maybe I can understand right now, this can give us a more granular way of doing an audit, right? Okay, if I were to perform only the credential validation, okay, not with the default authentication service set forth by Microsoft, it could be any way of authentication, right? I've got the credential validation setting that I can be turned it on, but I don't want that anyway. I've got one setting that caught my eye, it's a Kerberos authentication service that should be monitored on the DCs. I've enabled to trigger events, right? For both successful and failed attempt of authentication of any domain accounts. And however, then maybe there's another question will come in. What about the local account? Since that has more significance in terms of tracking an attack. Okay, this is where log on, log off. So the account log on events, if I enable any setting under that category, the domain controllers would trigger, would get that event triggered in the security log that can be captured and that can be manipulated. And yeah, that can be made very useful, right? So on a local machine, maybe a server workstation or all the computers exist in the domain. So the log on log off category is the one that we need to enable the settings in. And especially I've got the log on audit log off, audit log on, audit account lockout, or any special log on that happened with any other negotiation or any other authentication package comes in the middle. Okay, so we can enable the respective audit subcategory setting underneath the log on log off category. On, just make sure we need to do this. We need to enable log on log offs audit setting on all the computers in the domain. But that shouldn't be a problem, right? We can do it through a group policy object, right? Okay, now getting back to the presentation and here, what are all the account log on events from the domain controllers and why do we need them? Okay, so for right now, we would understand the need for the authentication event monitoring. Okay, I have a couple of domain controllers in my domain and I've turned on auditing centralized way to track the authentication events in my domain. So any domain account, we're trying to authenticate, it's being taken care of by one of the DCs. So if I could turn on the auditing on all the domain controllers, which I should, because the domain controller security log events will not be replicated from one domain controller to another, where it is really mandatory for us to enable auditing, enable the audit policy setting on all the domain controllers and this should be centralized. All right, all right, we'll answer whether it's really a challenging one or not, maybe in the later point of time, of course it is, right? Okay, and why and at what point do these authentication events would help us? Well, we can make use of the authentication events for centralized audit trail or for further correlation if we were planning to forward events from the domain controller to a centralized SIEM solution. In fact, AD AutoPlus is one of the major components of our SIEM offering anyway, right? So for those scenarios, we need to have or we need to leverage the log on, the authentication events that happen on the domain controllers. And now log on, okay, since I already, maybe I didn't give any surprise before, right? But I just opened it up already. Log on, log off category events are very important. In fact, of paramount importance in detecting attacks very quickly, very quickly than authentication event fields because there's another difference between a log on and authentication here. When it comes to authentication, it doesn't have a validity, right? It doesn't really have a validity, but it's not the case for the log on event. Log on has a validity and the significant field represents the validities to log on ID, which is gonna be the same. From the time a user logged into a computer, whatever he does after he logged into the computer till he logs off. Log off is eventually a local process on a computer. It has nothing to do with the domain controller unless that's a local log off on a DC, right? The log off is essentially a local event that happened on a workstation. And the boundary of the log off is just the machine itself. And another point why we need or why the log on log off events are more significant because it would reveal the log on type. We will be able to differentiate, you know, while doing an investigation of an attack or a scenario, right? We will be able to differentiate between the log on type. Microsoft has the code words, numbers, that will be one of the fields in the event, local log on event, right? So that can be leveraged as well to identify what type of log on it was. Whether it's a interactive log on, whether it's a network log on, a user might have logged the screen, go away, come back again, unlock it. So there are different ways an account can log into a computer. Of course, whenever we type in a username and password, it's somewhere in the domain, whether it's on a local computer, and obviously that's gonna contact the nearby DC. So that will be considered as an authentication event on the DC. However, do they gonna help us with? Okay, absolutely not. Maybe not to a greater extent, right? That is why we have to rely on local log on events that should be captured from all the computers in a troubleshooting standpoint. Right? Okay, now let's take a look. Oh, I think I covered this already. It's the only way to see when a user log off. If we want to get some information more productively, right, okay, I'm an administrator and I have users in the domain. I need to be able to track their log on time, log off time, and I need to be able to do some math here and I need to be able to generate it as a perfect report to calculate the production time. And in fact, in a different, okay, this is just based on a capacity planning standpoint. In a security standpoint, when an attackers gain access to a malicious, I mean, through a malicious source, when he gains access to a computer, he's probably gonna be there for a while until he's detected, right? Okay, if he found that he just got detected and obviously he will leave, he may or may not leave, but still, even if he leaves, he's probably gonna leave the back door opened up, right? Okay, so the point is, if we could calculate the log on session time when a user log in, I mean, when a local, I mean, whether it's a domain account or the local account, when that account logs in, when that account logs off, we need to be able to calculate the time between the session. All right, and now we'll find out that what are the significant attributes in the log on event that is gonna help us with doing a job? Okay, I got all the numbers given for the log on type, which is way too important for us to track, to monitor, to detect an attack, a log on attack significantly, right? Okay, when I, all right, you know what? Maybe I can show you exactly what I'm talking about by getting into a computer and show you the event viewer. Got my computer, opened up, got my event viewer. When I filter, all right, 4624, that is the log on success event, but I also gonna need 4625, that's the failure log on event. I just take it up, all right, let's take a look at what's this. Okay, what are all the fields that are gonna help us with? What we were talking about, the log on type, right? Okay, I've got log on type five, which represents this is a service log on. And what is that? It's a local system account, by the way. No, either local account or a domain account is not involved here. All right, where to do to show exactly, okay, yep, it's a services. What process which initiated this log on type? A log on type five is, that's gonna be obviously services process, right? Okay, and do I have, oh, the local, log on type seven, explicit credentials log on. All right, this is another good example. Another way that a log on can happen locally, right? Okay, I've logged into my computer and I need to open a program that running on my computer, which requires elevated privileges. What I do, I immediately go ahead, right click on the icon of that application, click on run as administrator or just run as, okay? If I just enter a different credentials that I got a hold of, maybe it could be an admin credentials. That type of log on is the explicit credentials log on. Maybe I use a bad job when I try to execute the bad job that requires validation of my credentials. So I can have the right parameters passed in my bad job to authenticate or to log on, to have a log on for that account, right? So this type of log on is being referred as explicit credentials log on, which is being denoted in the event log on. So there are different log on types as we discussed. All right, let's get back to the presentation. So I've got some numbers, service, bad job, network, cached, interactive, okay. And then, all right, besides the log on type, what are all the other fields? Hmm, account name, which we will be finding in the event, log on event, and log on ID. Yep, that is another unique field. It has more significance in terms of tracking an attack because we can calculate how long the user was in the session. And using this log on ID, it's a hexadecimal, some characters, that log on ID will help us in filtering out the right and, yep, rightful events here because we would still have the legitimate events happening on that computer. In order for us to differentiate between a legitimate event and an attack, we probably could use the log on ID and the process name. How about we create a baseline, which is very important in detecting an attack because we never know exactly what is the difference between an attack and a normal event, or in between, just want to rephrase the previous slide that we had, right? Log on event could be anything. In order for us to differentiate that, the process name could do a greater help. If it was a svchost.exe, word.exe, notepad, a PowerShell, so what process has initiated that log on activity on that computer? So recognizing the process and correlated with the username and the log on ID because, for example, we know there is a concept called privileged access workstation, P-A-W. Obviously, that's still in the role, which is being used by administrators or any authoritative users, okay, any privileged users in the domain if they want to use, if they want to administer, if they want to perform remote administration of any entity in the Active Directory. They're probably going to use, you know, that's a very good practice, even recommended by Microsoft. So have a practice of having privileged access workstation. Yep, isolate them very quickly and apply a lot of restrictions to them, to those machines because they're going to be using only for administrative purposes. So in those machines, we know we can create a baseline by having a list of, by getting a list of what are all the users who are going to be accessing those machines. All right, only the admins, that's okay. What if you get an audit trail? In the audit trail, you get an event seeing that logon activity happen on that privileged access workstation by a normal user. Well, that's not supposed to happen. It's not supposed to happen, well, but it happened. How are we going to differentiate that? Through the process name, identify the process name, identify the network address too, and the source port, because this field refers what the original source for that initiation. For example, when I, all right, let's take another example. To access a file share from my computer, I log into my computer, I just connect to my file share, file server, file share, right? So according to the file server, a remote logon attempt is happening to access a resource within it, within the file server. So according to that file server, that's a network logon just happening. In that event, obviously the source and network address, which is going to be my machine's IP address and the port that could help us in a way as well when things occur, right? And common authentication packages, some of the fields that I can show you in the security log event, all right? Let's get into the machine very quickly. I just opened it very randomly. Got a source network address and the port number, what authentication process which is involved right here, what type of logon, yep. And then, all right, which account is responsible that got the logon successful attempt happen on this machine? It's a network logon type two. You know what? I never know this user. I just probably have to think about it and tweak and investigate further. How? Okay, I got the source network address. So I can do an NSLOOKUP. I can try to find out which machine has this IP address and in that machine, someone is trying to, someone has logged into that computer. From that computer, he had initiated a logon, a network logon to this machine, MAPC1. Well, I just got it investigated pretty quickly, right? With the different fields that can be leveraged from a local logon event 4624. All right, let's get back to the presentation. Okay, well, when we try to, so far we've been discussing about a logon, successful attempt, how to track and what are the significant fields. When it comes to detect the failed attempt of logon, which is way more important than detecting a successful attempt because tracking, monitoring the logon failure attempt is the possible way to prevent, to know the baseline of where things go wrong, really. And the different ways a failure can happen, whether it's, whether for authentication or for the logon event, it doesn't matter, all right? And when it comes to logon event, if it was a logon failure, these are all the status codes that we can find in the logon failure event 4625 that are gonna help us with. All right, I've got status code 64, nswit64. That represents logon failure having a bad username. Well, you know, what are the possible way that an attack can happen here? Brute force. Try to shotgun administrator account attack. Someone may try to, everyone knows there is a user called A-D-M-I-N-I-S-T-R-A-T-O-R, administrator, right? They may try to brute force the credential for the administrator. And that is why we usually suggest the clients to never ever use the built-in administrator account, whether for Active Directory or for the local servers and machines, never ever use the local built-in administrator account. Rename the account with some significant naming convention that we can easily understand, right? Not the attacker. And there are different, and why do we need these status codes to differentiate between the legitimate activity and an attack? The first two status codes, if we get more number of logon failure for one or more accounts with these top two status codes, they're definitely an attack. We need a baseline, right? In order for us to build a baseline, these status code would help us a lot. And now let's talk about various attack scenarios and how the logon events are gonna help us with. The first one, brute force. Very old and still effective, even now. And number two, kind of pretty low sophisticated attack it is. Performing a shotgun attack against the administrator, built-in administrator account and honeypot accounts. And yep, finally, I've got another one to discuss here is the usage of harvested credentials. And obviously we know the attack vectors, right? For the pass the hash or pass the ticket, well, pass the hash for NTLM package and pass the ticket for Kerberos. And why do we need? And what's the very important here? To detect the failure due to bad username. Because if an attacker get a hold of a compromised account, where exactly is using this? He had already harvested the credentials and where is using the credential? Just to avoid a consequence, a dreadful consequence for the admins. We need to really know where those harvested credentials are being used right now. We probably have to come up with a different combinations. Well, I'll talk about it, right? And then, well, brute force and how the brute force can be detected. Pretty much in a fast and furious way, when more number of 4625 log on failure events happen on local computer with the same username and the status code is 6A, that's bad username. That's pretty fast and furious. Someone is trying to really, you know what? Guess the possible combination of a user account and a password. And in another way to detect the brute force in a slow and steady way, we're gonna need to create a baseline. We know, just get a list of administrators and local accounts on any computer, have different naming conventions for privileged accounts so that we can easily detect and identify, even create custom profiles, right? To isolate the monitoring capability when things go ugly only for the privileged account with the help of naming conventions, with the help of a baseline, right? And another variable for creating a baseline is, let's take a week, all right? In a week, seven days, how many failures are happening legitimately for an account? Or if there was a privileged account in particular, how many weeks and how many hours and how many minutes? So let's create a baseline in terms of time period. So determine a threshold right here. And then finally, create a profile to get a notification, to get an alert if all these variable have been met. And that's definitely an attack. That's definitely a brute force attack, right? Okay, and then another one is attack against the default administrator and honeypots. We're gonna need to be able to investigate the occurrences of 4625 with the word administrator. I'm pretty sure we can find that inside the event log, a security log, right? Okay, I can do a find, I can add a filter. You can do a find here. I'm gonna type in my, all right, I just got one myself, Bruce McClain. If we could detect, if we could get ahold of 4624 and 4625 with the word administrator, or we may have honeypot accounts in the domain that exists. That's a very wise way to have honeypot accounts created. Because when we try, when an attacker try to guess the username or a password, when they try to harvest the credentials, they probably gonna keep on trying. They probably gonna attempt, keep on attempting. But having the honeypot account served, they never end, okay? But even if they end, within the time, we will be able to identify and detect the behavior, the pattern, right? Okay, and that is how. And the number of event IDs that are gonna help us with, 4625 and 4624. And with the release of Windows 2016, Server 2016 and Windows 10, there's another way Microsoft has introduced a couple of event IDs, 4798 and 4799, for local object enumeration. Let's say there's a Windows 10 machine, which is being, well, controlled by an attacker. It's trying to find out who are the administrative privileged users that exist on the domain. So local account enumeration, and even Microsoft has right now an event that can be triggered on the machine whenever or whoever try to enumerate the user account, or a group, try to find out which member belongs to this group, or which group this user belongs to, all right? So 4798 and 4799, all right? Group membership was enumerated by Bruce McClain. So which account? All right, this is the local account. You see, because machine name is in front, not the domain name. Which process is involved right here? I know I did this, right? Okay, so let's getting back to 4798 and 4799 is very important. And finally, the Harvester's Credential Usage Detection. Please get a hold of large number of occurrences of failures due to bad username. And we have, or we should be having, logon restrictions applied for at least privileged users so that no privileged user account should be used on a normal user computer, except for emergency purposes, right? We should be having a different account, Honeypot account, for the administrators so that they can be used for emergency administration purposes. And various combination, how about it? There are different variables, right, when it comes to attacks. What are they? Baseline, if we have one. Number two, to determine the threshold. The time period, because Fast and Furious and Sloan study different ways to detecting an attack, right, so we have to go with various combination and just add all the baseline variable. And we need a centralized solution to do all these, to cover up all the areas that we talked about, and that is why we have AD Audit Plus, right? Okay, I'm gonna take a quick, I'm gonna show you very quickly about the AD Audit, how we can achieve all these very quickly. You know what? All right. I've got AD Audit Plus here. I've got my domain configured, well, which means I'm collecting authentication events from the domain controllers, and I have configured my member servers that I'm collecting events from. I do also have my workstation configured, and that I'm monitoring for the local logon events. All right, okay. Let's take a look at some of the data, because the event collection, which is a key factor, is happening seamlessly. All we have to do, we just turn on auditing. Even AD Audit can help you with that. Just collect events, put it in a database, centralized location, which AD Audit has the package within it, right? Database package, a proprietary one. And now, in terms of security, find information, or even in terms of compliance, we need to be able to run a report, get the data out of the software in a way that we want it to get them displayed, right? Okay, we have two different categories. User logon reports, local logon logoff. And I'm pretty sure, right now, we would be in a situation to find the exact difference between those two report category. Local logon, logoff, I mean, user logon reports, all these reports, underneath this category, they are based on user authentication events. When I try to run a report right here for the logon failure, okay, check out the fields. I will be getting username, password, and let's go with last 24 hours. Got the username, password, the source, all right. And what's the reason, whether it's a bad username or password? Check out the user-friendliness of the software, which I can leverage, I can save a lot of time. What's the package, the logon service, which is happening? All other relevant fields, the data can be displayed. But here, but still, a lot of significant fields are missing, right, in the authentication events. Maybe we can use this for the consolidated audit trail. But we still more to detect and attack, local logon, logoff. About logon activity, by default, this report will be run for all the machines, configured machines in ID AutoPlus. I got away here. Which username is logged in from which IP address? The source computer from where the login happened and the logon type, whether it's a interactive network. Remember, we saw different logon types, numbers, right? I got the numbers, I got the type given, and record number, a brief message, the source port, the authentication package, the logon ID. We've got all the fields covered and just got the result displayed in more friendly manner. How about, we already have a lot of reports, right? Logon for detecting attacks or even that's gonna replay attack, of course. Even to calculate, to generate report for productivity, we have reports called user work hours to calculate the actual working hours. But that's not the major point to be discussed right now. But in terms of detecting attack, we need the data to get displayed in a more friendly manner with rows and columns. That is why we have, I mean, ID Auto has a way to get the result displayed in a way that we want it. And in fact, besides that, I can even go for customization. I create a report, how the view should be. In fact, I already have one. Whether I just need to be, I just need the result to be displayed in a graphical view or a summary view, how? And what kind of data to be displayed right here? I just need only a bad username. Or logon failure happened due to password expiry. I just save it and next. But you know what? My goal, in order for me to detect the attack, whether it's a brute force, yeah, there are different variables that would come for the brute force. Whether if I have to create a baseline and detect the shotgun attacks on the administrator, I'm gonna need to add more variables, right? With the help of a lot of fields, significant attributes that come out of the logon event, local logon event. All right, I'm gonna add a filter. How about I just need to add only few variables, say few users. I've got them grouped, okay? Administrators. Any logon failure that happened for a privileged user. I think I can select only specific groups right here. And one more variables. I can even add the where, come on. IP address. Since I have baselined, you know, what are all the machines where usually administrator have login. So if I just go ahead, try to not equals and type in the IP addresses with a comma, more number of IP addresses with a comma, all those IP addresses could refer the IP addresses of the privileged access workstations in where the legitimate activity for administrators would usually happen. Because my goal or our goal is to display a report, generate a report for logon failure when it happened for administrators, which is not on the privileged access workstations, right? Any workstations other than the chosen ones, right? So let's even, I can configure an alert with all the variables that I was talking about before. I have just create a name, description. Okay, then add the report profile. The one that I need is under local logon logoff category. Local logon failure for computers. And then filters, yep, I can apply a threshold in order for me to detect the brute force attack. All right, there's gonna be say 100 failures that happen within two minutes. Yeah, it's pretty fast, right? With the same username, the caller username field which denotes who, which account is involved in the activity, in the event. All right, I got that. And filter, I'm gonna add the required filter. With the help of a lot of fields and relational operator, I just enable the notification, type in my email address and save it. At the same time, I can also execute a script which can take an action here. All right, let's get into presentation. We are at the verge of finishing the webinar. And just to summarize what we discussed so far, we started discussing about the logon auditing and the importance of logon auditing. And the different attributes that help us with detecting possible attack scenarios and how we can leverage AD Autoplus, components of AD Autoplus to make use of the logon and authentication events to detect possible attack scenarios. So there are a lot of webinars, you know, crispy topic based webinars lined up for us. All right, in fact, we do have a webinar on March 7th. All right, which has a topic of discussing on various attack scenarios for the entire Active Directory. Well, today we focused on the logon attacks, right? So we will be soon focusing on different attack scenarios besides the logon here. So we have a webinar on March 7th. And also, in the US, most awaited seminar series with the Direct Melbourne, our technical evangelist, Vivian Satien, our technical evangelist, who will be there to interact with the administrators with you all around the United States. We're gonna have seminars on March next month coming up, all right, on three dates, okay? So the registrations are happening right now. And I'm sure, please, please make use of, please leverage the content, the discussions, interactions, and overall, the knowledge of the evangelist who will be coming up during the seminar, okay? So I'm sure I can have my colleague who is assisting you in the chat who will have some information pasted in the chat window regarding the seminar since, well, we are having some early bird registrations going on right here. I'm sure this could be very potential fit for your needs and infrastructure. So thank you so much for your time today. I really appreciate everyone. And until the next webinar, so this is Bruce McLean signing off right now on behalf of Manage Engine. I thank you, everyone. And you all have a very good day. 笙ェ Ooh, ooh, ooh, ooh, ooh, ooh, ooh, ooh, ooh, ooh, ooh 笙ェ

TL;DR

  • Active Directory logon auditing is essential for detecting attacks because adversaries must authenticate to move laterally, escalate privileges, and maintain persistence—making logon events a critical early warning signal.
  • Authentication events (domain controller) and logon events (local endpoints) serve different purposes: authentication provides centralized audit trails, while logon events reveal session context, logon types, and source details needed for attack detection.
  • Comprehensive monitoring requires auditing both domain accounts and local accounts on servers/workstations, as attackers often exploit local administrator credentials and backdoor accounts to evade Active Directory visibility.
  • Logon type codes (interactive, network, service, unlock, etc.) and logon IDs enable security teams to differentiate legitimate activity from suspicious patterns like pass-the-hash attacks and unauthorized remote access.
  • ManageEngine ADAudit Plus automates logon auditing by collecting events from all endpoints, providing customizable reports and threshold-based alerts to detect brute force attempts, privilege escalation, and anomalous access patterns.
  • Effective logon auditing requires baseline establishment (e.g., privileged access workstations) and filtering to reduce noise, enabling security teams to focus on deviations that indicate compromise or policy violations.

The Critical Importance of Logon Auditing

This technical webinar explores why Active Directory logon auditing is essential for enterprise security, particularly in detecting persistent attacks. The session distinguishes between normal user logon events, suspicious activity from disgruntled employees with retained credentials, and malicious actors using techniques like pass-the-hash attacks to escalate privileges. The presenter emphasizes that attackers often gain initial access through compromised credentials and then laterally move across systems by extracting cached administrator password hashes from local SAM databases. Understanding logon patterns is positioned as a fundamental component of user behavior analytics and early attack detection, making comprehensive logon monitoring a non-negotiable security control for Active Directory environments.

Authentication vs. Logon Events

A key technical distinction covered is the difference between authentication events (captured on domain controllers) and logon events (captured on local workstations and servers). Authentication events occur when credentials are validated against Active Directory but lack context about session validity and logon type. Logon events, by contrast, provide critical forensic details including logon type codes (interactive, network, service, unlock, etc.), logon ID for session tracking, source IP addresses, and authentication packages used. The webinar demonstrates that while authentication events are useful for centralized audit trails, logon events are far more valuable for attack detection because they reveal how access occurred, from where, and for how long—essential data points for identifying brute force attempts, lateral movement, and unauthorized access patterns.

Auditing Domain and Local Accounts

The session addresses the often-overlooked challenge of monitoring both domain account logons and local account activity on servers and workstations. Local accounts on business-critical systems like web servers and application servers represent significant attack vectors, as adversaries may create backdoor accounts or compromise existing local administrator credentials to maintain persistence outside of Active Directory visibility. The presenter explains that comprehensive security requires enabling audit policies across all endpoints—not just domain controllers—to capture logon events wherever they occur. This dual-focus approach ensures that attackers cannot evade detection by operating through local accounts that bypass centralized Active Directory authentication.

Leveraging ManageEngine ADAudit Plus for Detection

The demonstration showcases how ManageEngine's ADAudit Plus solution simplifies the complex task of logon auditing by collecting, correlating, and presenting logon data in actionable formats. The tool provides pre-built reports for authentication events, local logon/logoff activity, user work hours, and logon failures, with customizable filters to baseline normal behavior and detect anomalies. Key capabilities include threshold-based alerting for brute force detection (e.g., 100 failures within two minutes), filtering by privileged user groups, excluding known legitimate source IPs (like privileged access workstations), and automated response through script execution. The solution addresses the challenge of manually parsing Windows Event IDs 4624 (successful logon) and 4625 (failed logon) across distributed endpoints, presenting security teams with user-friendly dashboards and reports that translate raw event data into intelligence for incident response and compliance reporting.

Chapters

0:00 - Introduction and Webinar Overview
2:00 - Why Logon Auditing Matters
5:49 - Normal vs. Malicious Logon Scenarios
8:00 - Pass-the-Hash Attack Mechanics
27:34 - Enabling Audit Policies via Group Policy
28:10 - Authentication Events on Domain Controllers
30:00 - Logon Events vs. Authentication Events
34:39 - Logon Type Codes and Event Viewer Demo
53:45 - ADAudit Plus Reporting Capabilities
56:00 - Detecting Attacks with Custom Reports
59:00 - Configuring Alerts and Automated Response
59:51 - Summary and Upcoming Events

Key Quotes

0:54 "Doing a logon audit is not an easy job for administrators, security officers, or even any auditors."
2:12 "Observing the logon activity and monitoring the logon activities for domain account, a user domain account, and local accounts on computer."
8:18 "Windows is still storing the last 10 hash values for an account. It could be a local account, it could be an Active Directory account."
9:19 "This is called pass the hash attack. And, but in order for the attacker to do that, he needs to get into that computer in the first place. He needs to have a valid domain credentials."
30:40 "Log on has a validity and the significant field represents the validities to log on ID, which is gonna be the same. From the time a user logged into a computer, whatever he does after he logged into the computer till he logs off."
31:19 "Another point why we need or why the log on log off events are more significant because it would reveal the log on type."

FAQ

What is the difference between authentication events and logon events in Active Directory?

Authentication events occur on domain controllers when credentials are validated against Active Directory, providing a centralized audit trail but lacking context about how access was used. Logon events occur on local workstations and servers, capturing detailed session information including logon type, source IP, authentication package, and logon ID—making them far more valuable for detecting attacks and understanding user behavior.

Why is it important to audit local account logons in addition to domain accounts?

Attackers often create backdoor local accounts or compromise existing local administrator credentials on servers and workstations to maintain persistence outside of Active Directory visibility. Business-critical systems like web servers and application servers with local accounts represent significant attack vectors that require dedicated monitoring to detect unauthorized access that bypasses centralized authentication.

How can logon auditing help detect pass-the-hash attacks?

Pass-the-hash attacks require adversaries to authenticate using stolen password hashes to move laterally across systems. By monitoring logon events for unusual patterns—such as privileged accounts authenticating from non-standard workstations, multiple rapid logons across different systems, or logons using NTLM authentication from unexpected sources—security teams can identify lateral movement attempts before attackers escalate to domain administrator privileges.


Categories:
  • » Data Protection
Channels:
News:
Events:
Tags:
  • Identity & Access
  • Security Operations
  • Threat Intelligence
  • Technical Deep Dive
  • Webinar
  • Active Directory Security
  • Logon Auditing
  • Pass-the-Hash Attacks
  • Authentication vs. Logon Events
  • Local Account Monitoring
  • Logon Type Codes
  • Brute Force Detection
Show more Show less

Browse videos

  • Related
  • Featured
  • By date
  • Most viewed
  • Top rated
  •  

              Video's comments: Active Directory Logon Auditing Deep Dive

              How secure is your Microsoft enterprise? Under 5 minutes, earn $10 in TruthPoints.

              View CPE Certificates

              Upcoming 360 View Events

              • Nov
                19

                Securing the Microsoft Enterprise: Strengthening Configuration Management, Identity Security, and Data Protection

                11/19/202601:00 PM ET
                More events

                XStreaminars (watch here)

                • Dec
                  09

                  Beyond Continuous Compliance: Getting Ready to Detect, Recalculate, and Adapt

                  12/09/202601:00 PM ET
                  More events

                  Industry Events (Sponsor Hosted)

                  • Nov
                    18

                    Hands-On Workshop: Connecting On-Prem Data to Databricks via MinIO AIStor (International)

                    11/18/202608:00 AM ET
                    • Nov
                      19

                      Zendesk Deep Dive: Get the work done faster with Employee Service AI Agents

                      11/19/202601:00 PM ET
                      • Nov
                        19

                        AI Agent Showdown: Discover What Muse, Dots, and Duet Bring to Your Website

                        11/19/202601:00 PM ET
                        • Dec
                          02

                          Skylight en acción: salvaguardando la IA más allá del usuario.

                          12/02/202604:00 AM ET
                          • Dec
                            03

                            CISO Insights: Steering Cybersecurity Strategies in an AI-Driven Era

                            12/03/202601:00 PM ET
                            More events

                            Upcoming Webinar Calendar

                            • 11/18/2026
                              08:00 AM
                              11/18/2026
                              Hands-On Workshop: Connecting On-Prem Data to Databricks via MinIO AIStor (International)
                              https://www.truthinit.com/index.php/channel/2194/hands-on-workshop-connecting-on-prem-data-to-databricks-via-minio-aistor-international/
                            • 11/19/2026
                              01:00 PM
                              11/19/2026
                              Securing the Microsoft Enterprise: Strengthening Configuration Management, Identity Security, and Data Protection
                              https://www.truthinit.com/index.php/channel/2076/securing-the-microsoft-enterprise-strengthening-configuration-management-identity-security-and-data-protection/
                            • 11/19/2026
                              01:00 PM
                              11/19/2026
                              Zendesk Deep Dive: Get the work done faster with Employee Service AI Agents
                              https://www.truthinit.com/index.php/channel/2181/zendesk-deep-dive-get-the-work-done-faster-with-employee-service-ai-agents/
                            • 11/19/2026
                              01:00 PM
                              11/19/2026
                              AI Agent Showdown: Discover What Muse, Dots, and Duet Bring to Your Website
                              https://www.truthinit.com/index.php/channel/2228/ai-agent-showdown-discover-what-muse-dots-and-duet-bring-to-your-website/
                            • 12/02/2026
                              04:00 AM
                              12/02/2026
                              Skylight en acción: salvaguardando la IA más allá del usuario.
                              https://www.truthinit.com/index.php/channel/2147/skylight-en-acción-salvaguardando-la-ia-más-allá-del-usuario/
                            • 12/03/2026
                              01:00 PM
                              12/03/2026
                              CISO Insights: Steering Cybersecurity Strategies in an AI-Driven Era
                              https://www.truthinit.com/index.php/channel/2231/ciso-insights-steering-cybersecurity-strategies-in-an-ai-driven-era/
                            • 12/09/2026
                              01:00 PM
                              12/09/2026
                              Beyond Continuous Compliance: Getting Ready to Detect, Recalculate, and Adapt
                              https://www.truthinit.com/index.php/channel/2250/beyond-continuous-compliance-getting-ready-to-detect-recalculate-and-adapt/
                            • 12/09/2026
                              02:00 PM
                              12/09/2026
                              Strategies for Safeguarding Endpoints in the Agentic Era with Cyera
                              https://www.truthinit.com/index.php/channel/2256/strategies-for-safeguarding-endpoints-in-the-agentic-era-with-cyera/
                            • 01/19/2027
                              04:00 AM
                              01/19/2027
                              Skylight in Aktion: Gewährleisten Sie KI-Schutz über den Benutzer hinaus
                              https://www.truthinit.com/index.php/channel/2176/skylight-in-aktion-gewährleisten-sie-ki-schutz-über-den-benutzer-hinaus/
                            • 01/19/2027
                              02:00 PM
                              01/19/2027
                              HUMAN Dialogue: Embracing the Rise of the Empowered AI Consumer.
                              https://www.truthinit.com/index.php/channel/2160/human-dialogue-embracing-the-rise-of-the-empowered-ai-consumer/
                            • 03/18/2027
                              01:00 PM
                              03/18/2027
                              Securing Access for People, Partners and AI Agents
                              https://www.truthinit.com/index.php/channel/2251/securing-access-for-people-partners-and-ai-agents/
                            • 06/17/2027
                              01:00 PM
                              06/17/2027
                              Stop the Leak: Data Loss Prevention for the AI Era
                              https://www.truthinit.com/index.php/channel/955/360view-stop-the-leak-data-loss-prevention-for-the-ai-era/
                            • 08/19/2027
                              01:00 PM
                              08/19/2027
                              Who Has Access to What...and Should They?
                              https://www.truthinit.com/index.php/channel/2252/who-has-access-to-what-and-should-they/
                            • 10/21/2027
                              01:00 PM
                              10/21/2027
                              After VMware: Building the Private Cloud You Actually Want
                              https://www.truthinit.com/index.php/channel/2253/after-vmware-building-the-private-cloud-you-actually-want/
                            • 11/18/2027
                              01:00 PM
                              11/18/2027
                              Assume Breach, Prove Recovery
                              https://www.truthinit.com/index.php/channel/2254/assume-breach-prove-recovery/
                            • 12/16/2027
                              01:00 PM
                              12/16/2027
                              The Self-Running Service Desk
                              https://www.truthinit.com/index.php/channel/2255/the-self-running-service-desk/
                            Truth in IT
                            • Sponsor
                            • About Us
                            • Terms of Service
                            • Privacy Policy
                            • Contact Us
                            • Preference Management
                            Desktop version
                            Standard version