What needs improvement with Carbon Black CB Defense?

Performing a malware scan usually takes a lot of time, more than 24 hours.

The product's reporting capabilities are an area of concern where improvements are required. From an improvement perspective, the price of the product needs to be lowered.

Getting the right technical support is a challenge.

It is challenging to reach the product’s technical support team. This particular area needs improvement. The device control feature could also be compatible with the user’s profile as well.

In my company, we face issues sometimes when there is a need to write custom rules or we want to write for some rules that are different from the standard rules provided by the solution. A person needs to set up some rules for end-user machines, during which the person needs to be completely aware of the tool and the user interface. Without proper knowledge, a person can't write custom rules. In general, a person without proper knowledge cannot set up the rules in the UI. The challenge is that if I write one custom rule and put it on all users, then sometimes it may not work for some of the users, while it may work for others. Some developers may work with some files that are mandatory for them to run regularly, and if my team wants to block such files, then we can mention it in our custom file name, but that also blocks the file for the developers. If the files used by developers get blocked, then the developers can raise an issue and state that they need an exemption for those particular programs since they need to run them regularly. Writing custom rules, stability, and pricing are areas of concern in the solution that need improvement.

There is room for improvement in the proxy servers. The implementation and management of those servers are difficult. The proxy servers have proxy servers in place to not connect directly to the Internet, and the implementation and management of those servers are difficult. Moreover, some customers request disabling Bluetooth in endpoints, but Carbon Black doesn't do that. So, there should be some flexibility for customization.

VMware Carbon Black Endpoint takes a step back when compared to other solutions in the market. Cortex XDR is a better solution compared to VMware Carbon Black Endpoint. In our company, we also wanted to have network detection, like a host-based IDS on VMware Carbon Black Endpoint, but we did not get it. The aforementioned reasons have forced our company to look for an upgrade or another solution altogether. In the future, I would like to see VMware Carbon Black Endpoint offering a host-based intrusion detection system with a better incident response within the platform where you can raise an incident, assign it, and have some response functionality in it, like triaging the incident and other stuff.

The product's stability could be improved.

The maturity of the Kubernetes security is absent in Carbon Black CB Defense. The solution has to mature on container security and a lot of cloud environment security. Security is available only for Windows, while security for Linux and Mac is not very strong. The deadlock issue causes me to put more effort into installing an upgrade. The numerous issues with the environment of the product solution should be addressed. Work orders are taking more than two months to get resolved. There's been one issue open for two months, and the solution they gave is being implemented step by step. Still, it is not meeting the requirements and breaking the system. Hence, our business is completely disturbed.

It is challenging to extract a report on the status of ongoing scans. They should work on this particular area of the solution.

The compatibility of Carbon Black CB Defense with operating systems is the only issue. Certain OS are not supported, resulting in an inability to install PDC. The deployment of sensors requires extensive fine-tuning, which should be a simple process. To streamline this process, they should create deployment packages with customized options based on policies and other factors. Creating these packages ourselves is time-consuming, which can impede our productivity. There is also a bypass issue that needs to be considered. Improvements are needed to address the compatibility issues between operating systems and Carbon Black CB Defense. Sometimes, the sensor enters a block state for unknown reasons. To prevent this, it would be helpful if they added a feature to ensure that it does not cause any problems. Additionally, there are issues with collecting events from machines due to sensor problems. We are working with Gateway to connect to all PCI or DMZ environments, and it would be beneficial to have a simpler configuration at the architecture levels. In reality, the deployment process is more complicated. We must add a script to customize the deployment process and deploy it on Mission C. Afterward, we install the sensor, which requires a company code, policy name, and other essential details. Furthermore, we are experiencing other issues, such as VMs pausing applications due to CBC. Troubleshooting these problems is time-consuming, and we usually must report the problem to the vendor, whose analysis can take an hour or longer. By that time, critical business functions may have already been impacted. Container protection is still in the initial stage, where they have integration in the market, but there's a lot of room for improvement, and there are a lot of changes required.

I would say that the technical support team should be improved since it takes them a lot of time to provide us with support. In the next release, I would like to see a host-based firewall.

CB Defense could be more compatible with Linux, and its cloud provision could be improved.

I would like to see the user credentials feature improved. I would also like to see more reporting features and better ways to roll the reports out.

When you're investigating an alert, you will get a graph and will see the details related to the process that triggered the alert. Below the graph, there are network connections, file modifications, industry modifications, and multiple other activities. If you want to specifically find which additional modification has been performed, you will have to find the log you're searching for. There isn't a search bar to check for file modifications or network connections. In that case, you don't have a search bar, so you have to check each and every event, which could be more than 1,000. You would have to check 1,000 events manually, or you would have to export sheets to view what you are searching for. If they added a search bar, it would reduce the time it takes to do investigations. If you want to log into a device, there's a process named winlogon.exe, which is supposed to be initiated. If I'm using Carbon Black, I will have to check where winlogon.exe is being observed or at what time it was being observed. Because there's no search bar, I will have to check for the event in all the device events. A search bar in the investigation page and some AI-related tasks like outgoing alerts, or recent tactics that are being used in the market, must be embedded in the tool so that it's easier to find alerts. The AI must be stronger so it can identify activity that is actually malicious.

Currently, it's hard to comment on areas for improvement, because I haven't used Carbon Black CB Defense long enough. What was rolled out to my company are mixed versions of Carbon Black CB Defense, so what I'd like to see in the next release is more synchronization, where it can detect the endpoint that's running an old version and suggest updates. That's the only thing I can think of right now.

There is no option for the solution to block automatically based on behavior. First, the solution needs a lot of time to record all the behaviors. Then, we manually have to create a behavior analysis rule to detect any malicious activity. The solution would be improved and be more effective if there was a way for this process to be done automatically.

It is still evolving, as we see. We started using the version 3.0. We've been migrating and upgrading as well, laterally, until version 3.2. So, we have been seeing a lot of improvements in general in terms of bug fixes and in terms of what are the things that we had encountered. I think they can probably bring in because there is a little bit of a gap between the native Antivirus solutions like Symantec or McAfee. So, you really can't say whether an end user will not be able to judge whether it's a Malware-free software that they are downloading or not. In those cases, if you have an application and a device control feature, I think it would be of great help.

There could be more knowledge. I think they made a mistake when they took away the Check Point integration, because it provides more automation and also more threat intelligence. Maybe you didn't see something within Carbon Black's sphere of what it knows, within their product line or their threat cloud or whatever they use for their intelligence. Maybe it didn't see anything of the files that it knows about, but what about somebody else's? And what about kicking into another product that does those kinds of things like sandboxing? I don't know why they would take that away. That doesn't make sense to me because they need to expand on that. The more they expand on that, the more confidence you have as a security guy. You have more confidence that that file is clean, and there's nothing bad about it. Bringing back the integration with Check Point would be a good start. This product is being used extensively in our organization. I'm actually looking for a replacement because of the fact that we lost that integration. That's really crucial, honestly. Otherwise, it becomes much more manpower-intensive. I need to spend more man-hours going through it instead of using automations. I prefer to set up things so my team doesn't have to spend a huge amount of time running down rabbit trails all the time. The more we can automate and still be secure about it, that is what we try to do. There are no additional features I would like to see added. I know they already have a cloud offering as well. You can manage things through their cloud for people that are always on-site. We mostly just use it for our own managed devices. We didn't really put it on. We never planned and don't plan to put it on or make it available to a BYOD kind of thing. This is all company-managed devices. It just made more sense for us to do it internally than putting it in the cloud. But we could have done either one, I suppose. But since we started out inside, we just kept it that way. It was just easier.

This product should be cheaper.

The node management could be much better. The one thing that they cannot do very easily is change the tenant from a backend. As an example, assets were sold from a company called Rayonier Advanced Materials and went to GreenFirst, which became GreenFirst as a startup. We had a tenant where all the machines were registered to the cloud. That is the tenant that was there for Rayonier. It is very hard for them to make changes to the tenants, such as rename or anything like that. What they really would push you to do is, "Your tenant is going to be under your company name. You have to uninstall all the agents and reinstall them again." Making changes at a tenant-level would be a welcome feature to allow divestitures and things like that. They can do some of these things, but they're not very user friendly or easily done. They basically tell you to do the hard lifting yourself. For example, they basically kept pushing me and saying, "Uninstall your antivirus on about 500 machines and reinstall it with the new tenant information." I would say "No, everything is a tenant. Rename me the tenant." I would like to see the GUI improved and easier troubleshooting. One thing they did that makes it easier in troubleshooting versus the older versions of the software is that now you can actually drill down to see the parent process and go all the way down. In CrowdStrike, they have a timeline where they actually build the whole scenario as to what happened. It's like a playback. It's almost like a movie. You play back and it says, "Okay, this process ran," and then it shows what it caused and everything. You can see all that and if there are any screen outputs it puts it on because CrowdStrike actually maintains some of those things. A playback feature would be very valuable.

I haven't run into anything that needs improvement. The website interface can be a little bit better, but it's still good as compared to most others.

I believe they could improve the new intelligence solution to monitor activity, in the network. They will most likely need to create or include a feature that checks the network.

Integration is difficult, but CB Defense is more powerful than others. It is difficult to implement but easy to pick up many detections.

Carbon Black does not have a big market in Pakistan right now. They are actually trying to penetrate the region right now. They don't have many customers. Even we are new to the Carbon Black as well, in that we knew about Carbon Black for a long time, however, as far as implementing it and giving it to our customers, we are still new to it. The pricing could be more reasonable.

The on-prem one was very problematic, especially version 7.2, which did not play nice with Symantec at all. The last upgrade of the client actually triggered a block to the networking, to our active directory domain controllers. There was a bug that we found was in Macs. It was triggering false positives as it wasn't able to figure out the right parent upon login. With the Carbon Black Cloud, we just got it two to three weeks ago. So far, I haven't seen any false positives. The cloud seems to be a much better product. With the on-prem one, the bug has been reported by the community in early January or February, something like that, at the beginning of the year, and it's still not addressed. They have released two versions since then, and yet neither of them addresses this specific issue. I need more time to explore the cloud deployment, as we've only had it for three weeks at this point.

Sometimes the solution blocks items that were previously approved and we don't know why. It is sometimes hard when I attempt to investigate, to know the commands. It's not easy to do that. You need to upload the right information. Occasionally, when we get alerts, we don't get all the information we need, such as the computer's serial number. If I reveal an alert in a new window, I need to go back to the main link as it doesn't work. Sometimes we need to close the solution and then open it up again. Occasionally, we'll have issues with the latest version and they'll basically tell us that they will improve it in the next iteration. They need to work on their version release quality. It would be good to have more information about the devices. If you get an alert that a malicious file is on your computer, Carbon Black really doesn't give you the full picture. We also need to wait for the user who owns the computer to be online before we can investigate everything. It's hard when you are working across time zones.

This solution could have greater granular control on how certain applications work. You are able to do the operation of allowing or disallow, or you can block unusual usage of an application, but they do not define it well. The PowerShell is being called in any way that the threat actor might use it versus an administrator. You are in a way taking this solutions' best guess at it or their understanding of it. They do not clearly tell you in technical terms how they make that determination. They should be more forthright about it, or if they can not tell us, they should just give us the control to make those selections. We are choosing it because at least we have that control where we do not have that same amount of control with other solutions like Cylance. However, they are still not telling us precisely what constitutes suspicious behavior, what actions, or what calls. It is a check box to say, lock if we have inappropriate use, or block if we have suspicious behavior. It would be helpful to tell us what that actually meant. In the future, I would like to see more granular control of PowerShell and more administrative tools.

In the next release, it would help if we can get better control over containers. This will help secure the containers in multiple environments. For example, we need to secure the Kubernetes containers. Apart from admin user login to see containers processes running, developers & operate team users also should be seeing the container's processes running.

It could be a bit complicated. You have to be very familiar with Carbon Black to understand what it is doing and why it is doing. I would like to have more explanations and simplification in the user interface. It would be good to get help and see more explanations. It should tell us that a software is blocked and the reason for it. It would be good to be able to build chains in terms of what caused what, what worked, and what caused an issue. We are now moving from Carbon Black to Cortex XDR. While choosing antivirus software, we were also looking at Carbon Black because it also has an antivirus package, and it is next-generation, but we were told that Carbon Black doesn't support firewalls. We have Palo Alto firewalls. We would have chosen this solution if it supported firewalls, in particular next-generation firewalls, but unfortunately, it doesn't. Therefore, we decided on Cortex XDR because it integrates with Palo Alto firewalls.

I can't think of any feature that needs to be enhanced or reviewed at this time. Some of the features that I see as an end-user, unfortunately, I haven't been able to see from a project management standpoint. I'm not sure if we're actually taking advantage of all the available features. I don't know if it's because we haven't configured it yet, or we are not using it. I'm not sure as to the logic of how we've decided to customize it. We've only really used it since February and therefore there may be more to do on that front. That's why it's hard to say if something is missing or if we just aren't utilizing it.

The whitelisting system, and the concept of it, overall, is pretty decent. The problem with the whitelisting capability is that it's pretty archaic. Based on all the security roles and the release privilege, it could take time for an application to be whitelisted and approved for use. The Mac support needs improvement, as it had next to none. The biggest problem we had was the Mac support. It had very little, and my C-suite is almost exclusively Mac, as is my marketing and development department.

The application control can be improved. It should also have an automatic update of the agents.

The solution needs better overall compatibility with other products.

Its compatibility can be improved. It did crash a server during deployment, which is not something that I want to happen. Its deployment should also be easier. The whole deployment cycle needs to be simplified. It is an enterprise solution, and to set it up right now, you have to be an expert.

To improve the ability to connect also feeds of third resources (communities).

The feature set for the firewall needs improvement. I am looking forward to learning more about the integration with VMware at the hypervisor layer.

The EDR portion could be better. I'm not a big fan, but it works. The End Point Detection Response and the way it lays our processes with our endpoint and its detection engine, in the way that it detects the admin or alerts we based on a threat. I feel that they're a little behind on the market from my perspective. Overall, areas of improvement would be the EDR part, the detection, also the cloud console. If you're trying to write queries or something, it's very slow, just not robust. It's a cloud console so it should be fast. If I run a query and I press enter, if it took two seconds, it wouldn't give me a nice loading interface, because it's stuck. I would see an operating system most of the time. I feel like it should be faster. But as far as the price and everything, I think it's a good product.

When you view the triage, it will show you everything within a given time frame, and not only the attack that caused the alert, which is what I want to see. It shows you all the events during that time, and that can be quite confusing. If they could focus on the alert and the event that the user wants to see, that would be better. There is also room for improvement on the reporting side, because it doesn't have reports. Many of our customers would prefer some kind of exportable report, like a summary. Carbon Black should have this feature.

As far as I know, Carbon Defense has nothing that can be installed on mobile devices. It lacks a defense solution for mobile devices, especially mobile tablets. I would like to see support for mobile devices and the pricing should be less than the pricing for a normal workstation. Also, there is not much education for customers about Defense versus its other products. They promote Defense as enough, but then they say if you need more protection you can go for CB Response. I don't know whether it's a technology issue or a marketing issue, but they should teach the customer more. They tell you you are secure with Carbon Defense but then they recommend Carbon Protect. There is not a lot of education on this. I don't want to have an incident in the future and their answer will be, "Sorry, you did not buy Protect." Security is a continuous process. I can accept that it has more features, but don't tell me, "You are not protected because you did not buy the more expensive product." In addition, these other products should be add-ons, not separate products. And the cost for them should be much less for adding on because you are already a customer. Finally, we receive a lot of high alerts. There is no priority system, from one to 10, where 10 is very dangerous and one is something easy. There is no way for us to tell why this alert is similar to that one.

This solution works well but needs lots of tuning and optimization.

The endpoint machines need improvement. The solution needs to be more effective for the end-user. It would be helpful to understand how to do some queries, but we’re still testing the solution right now, so everything is very new and we’re still learning the system.

Symantec needs more investigative features out-of-the-box. Though, they are using the Advanced Threat Protection add-on to correct some of this. It is also not quite as feature-rich as some of the more advanced MDR platforms out there. Carbon Black needs to do a better job of proving their platform in the industry, and providing a bit more access to do industry testing with real world examples to help prove their platform. In additional, they have been actively porting over a lot of features from some of their other products, and they should continue to expand on that. Going forward, this will be extremely helpful.

The UI interface needs improvement. The management needs further work in future versions.

In some areas one of the big issues for me is responsiveness to issues that arise with the solution. There are some components that leave a bit to be desired and/or that are bugs, or that even if it's a feature update request. These kinds of things are not the fastest company to respond to those. We did have a bug that was persistent for it's now going on two months and it hasn't been fixed. That is one of the drawbacks. This is really impacting what we need to do with it. But, the bigger issue is the organizational responsiveness to clients. In addition, I think there should be a cloud gateway. It needs to move into a transitory space between our On-Premise and external where it does not have to be in two separate instances. It should marry the two. Also, it would be good to have them working in the containerization space, as well. To have a mechanism for securing cloud modules a bit better. This would be ideal. It would help encompass more of the broad range security so we do not have to couple this with other outside solutions.