Sonatype Lifecycle Reviews

Our use case for Nexus is to monitor all of our dependencies and the main thing we're using it for is tracking vulnerabilities listed against those.

It gives alerts for new vulnerabilities before our clients do, so we have time to review them, audit them, and determine how we need to proceed with resolving the issues before we get any client communication.

Before we had this in place, we had a much more reactive approach to CVE listings.   Since integrating this, and as we've refined our process over the past eight months or a year, we have moved to a proactive approach allowing auditing and decisions on mitigation before any incoming client submissions.

In addition, it has brought open-source intelligence and policy enforcement across our software development lifecycle. As a component of the lifecycle, it gives us more controls in place. As far as bringing in dependencies goes, we're able to see what a dependency is introducing, from a security and licensing perspective, before we publish a release to the public. So within the build stage, if we pull in a new dependency, Nexus will very quickly tell us whether it has issues or not. And we catch it. It scans in the build stages; we have it checking our staging where we're doing our regression; and it's also monitoring our released branches and letting us know if issues are found in our releases. It really does hit all stages of that lifecycle.

I like the JIRA integration, as well as the email notifications. They allow me to see things more in real-time without having to monitor the application directly. So as new items come in, it will generate a JIRA task and it will send me an email, so I know to go in and have a look at what is being alerted.

The policy engine is really cool. It allows you to set different types of policy violations, things such as the age of the component and the quality: Is it something that's being maintained? Those are all really great in helping get ahead of problems before they arise. You might otherwise end up with a library that's end-of-life and is not going to get any more fixes. This can really help you to try to get ahead of things, before you end up in a situation where you're refactoring code to remove a library. The policy engine absolutely provides the flexibility we need. We are rolling with the default policy, for the most part. We use the default policy and added on and adjusted it a little bit. But, out-of-the-box, the default policy is pretty good.

The data quality is good. The vulnerabilities are very detailed and include links to get in and review the actual postings from the reporters. There have been relatively few that I would consider false positives, which is cool. I haven't played with the licensing aspect that much, so I don't have any comment on the licensing data. One of the cool things about the data that's available within the application is that you can choose your vulnerable library and you can pull up the component information and see which versions of that library are available, that don't have any listed vulnerabilities. I've found myself using that a lot this week as we are preparing for a new library upgrade push.

The data quality definitely helps us to solve problems faster. I can pull up a library and see, "Okay, these versions are non-vulnerable," and raise my upgrade task. The most valuable part of the data quality is that it really helps me fit this into our risk management or our vulnerability management policy. It helps me determine: 

• Are we affected by this and how bad is it? 
• How quickly do we need to fix this? Or are we not affected?
• Is there any way to leverage it? 

Using that data quality to perform targeted, manual testing in order to verify that something isn't a direct issue and that we can designate for upgrade for the next release means that we don't have to do any interim releases.

As for automating open-source governance and minimizing risk, it does so in the sense of auditing vulnerabilities, thus far. It's still something of a reactive approach within the tool itself, but it comes in early enough in the lifecycle that it does provide those aspects.

The biggest thing that I have run into, which there are ways around, is being able to easily access the auditing data from a third-party tool; being able to pull all of that into one place in a cohesive manner where you can report off of that. We've had a little bit of a challenge with that. There are a number of things available to work with, to help with that in the tool, but we just haven't explored them yet.

We're going on our second year using the solution.

I've never had any stability issues with the application. I haven't performed any of the upgrades, but we've never had any downtime and we've never had any issues with notifications or an inability to access the information we need.

The technical support is fantastic. I reached out with a suspected false negative and had a response within hours, and within the next day they had determined that, yes, it was a false negative and, that same day, the notification came in when they had resolved the issue. So within less than 48 hours of reporting a false negative, I had a full turnaround and the result returned in the tool.

Before IQ server we used an open-source solution called OWASP Dependency-Check. We wanted something a little more plug-and-play, something a little more intuitive to configure and automate.

For the initial deployment, it was in place within a couple of days of starting the trial.

We did have an implementation strategy sketched out as far as requirements for success during the PoC go. The requirements were that it would easily integrate into our pipeline, so that it was very automated and hands-off. Part of the implementation strategy was that we expected to use Jenkins, which is our main build-management tool.

In terms of the integrations of the solution into developer tooling like IDEs, Git repos, etc., I wasn't really part of the team that was doing the integration into the pipeline, but I did work with the team. We didn't have any problems integrating it. And from what I did see, it looks like a very simple integration, just adding it straight into Jenkins. It integrated quite quickly into the environment.

At this point we haven't configured it to do any blocking or build-blocking just yet. But that's something we'll be reviewing, now that we have a good process.

We have absolutely seen ROI with Sonatype. The more proactive approach is definitely a return on investment. It significantly lowers the turnaround for responding to incoming issues. It also empowered our support staff to be able to pass along audit results without having to loop in the security team directly. There is a much lower overhead involved when doing it that way.

Also, the ability to better manage our vulnerability management by getting the detailed information from the scan results or the listings, and being able to audit them thoroughly and test them really helps with development resources in our case. We do not have to cram in a bunch of upgrades just for the sake of upgrading if we're constrained elsewhere. It really helps prioritize dev resources.

I don't know if it has directly saved time in releasing secure apps to market. It has definitely made everything more efficient, but unless things are critical and can definitely be leveraged, we don't necessarily delay a release.

The upgrade processes are definitely a quicker turnaround because it allows us to actually target versions that are not vulnerable. But it is hard to quantify whether, in the grand scheme of things, our developers are more productive as developers.

We looked at things like Black Duck, White Source, and White Hat.

The biggest issue, and this is why we went with Nexus, is that there were more results and there were far fewer false positives than in the other tools.

Take some time configuring your notifications and your JIRA integration properly, along with the policy tweaks. As you integrate and as you first deploy the tool, don't block any builds until you start to catch up on any issues that may be there. Really spend some time with that policy review and make sure it encompasses and aligns with your vulnerability management policy appropriately.

It is incorporated in all of our software branches, and we keep our most recent end-of-life branch active in it just to monitor for critical issues, so we can notify the community to upgrade. We may also add our new mobile application to it.

Nexus Lifecycle is definitely a nine out of 10. I would say 10 if it were a little easier to get the audit information out. Again, there are ways around that so I am not taking off much for that. It's a solid nine. The results are amazing. The quality of the data coming back is great. The audit interface is easy to use.

We use Nexus Lifecycle to check our third-party libraries for vulnerabilities. 
There are also different application teams that use Nexus Lifecycle to configure our product. I'm one of those product users, so I can only talk about it from the perspective of my product. 

The team managing Nexus Lifecycle reported that their internal libraries were not being identified, so they have asked Sonatype's technical team to include that in the upcoming version.

We have been using Nexus Lifecycle for about a year and a half.

Nexus Lifecycle is stable. 

Nexus Lifecycle scales to the level we need. It's working fine.

Sonatype support is quite responsive. When we needed something, we could reach out and set up a meeting. They provide the best support possible.

Setting up Nexus Lifecycle is simple.

We evaluated Veracode, and we evaluated Black Duck, as well. The marketing team from Sonatype was more responsive and followed up on the progress during the proof of concept, so that was one reason we chose Lifecycle, but the features are almost exactly the same across products.

I rate Nexus Lifecycle eight out of 10. 

We have many use cases. Our main use case is focused on Nexus Repository and a little bit on Nexus IQ, including Lifecycle. The basic use case is storing Maven, Java, JavaScript, and other kinds of artifacts. For some years now we have implemented more complex solutions to manage releases and staging. Since Nexus Repository introduced that feature for free and natively, we moved to the feature provided for managing release staging.

It has only improved things very little because, for now, we use the reports from IQ to improve the libraries, but it doesn't yet have enough coverage.

It has helped to enforce open-source intelligence and policies across our software development lifecycle, mainly by automating controls that we had put in place before. It has helped to enforce things. It has blocked some open-source components or, more accurately, raised warnings about them. It's not a blocker in our system because, for now, it's only implemented as an informative system.

The solution has also improved the time it takes to release secure apps to market. We have been able to replace homemade scripts, which took a few hours to create, by very much simpler workflows provided natively by Nexus, which are working in a few minutes, or tens of minutes. It has saved us about 40 percent, in terms of time. But more than the duration, it's helpful that we don't have to maintain or make scripts. That's the most important thing.

It has improved developer productivity and accuracy. That happened a long time ago because we have been using it for so long, but as soon as we deployed the Nexus solution in the company, people didn't need to locally build a lot of stuff. So we were much more easily able to work together, to collaborate, and consume other teams' products. That was a long time ago, but it was definitely an important step.

The REST API is the most useful for us because it allows us to drive it remotely and, ideally, to automate it.

We have worked a lot on the configuration of its capabilities. This is something very new in Nexus and not fully supported. But that's one of the aspects we are the most interested in.

And we like the ability to analyze the libraries. There are a lot of filters to output the available libraries for our development people and our continuous integration.

The solution integrates well with our existing DevOps tools. It's mainly a Maven plugin, and the REST API provides the compliance where we have everything in a giant tool.

We do not use it for more because it is still too immature, not quite "finished." It is missing important features for making it a daily tool. It's not complete, from my point of view. All the Lifecycle tools are not yet finished and usable in production. We are using it in production, but it's not fulfilling all our needs. It's not yet finalized.

It's the right kind of tool and going in the right direction, but it really needs to be more code-driven and oriented to be scaled at the developer level. Nowadays, developers need to be autonomous, so we need to be able to supply them tooling and then everything should be orchestrated around the principle of GitHubs. That is really important in IQ because developers will want to make changes and they need things very quickly. Everything should be driven by that but that is not yet the case. The features are interesting but the way those features are configured and tuned is not quite there yet.

There is room for improvement in the way it is managed, having code-driven configuration, and automation. It needs not to be an old-style tool. Today, a computer tool must be usable in many ways: as a client for developers, as a webpage and reporting tool for managers, and as an automated blocker for continuous integration. It must have a REST API and it must have many features that make it usable in many ways. Currently, that's not the case. 

One thing I can say that is very positive is that it is much better than the other tools. But regarding usage, it's not perfect. It's missing everything around the tuning and usage.

We have been using Nexus Repository, version 2, for about 10 years or so. We switched to the Nexus 3 a little bit more than one year ago, along with Nexus Lifecycle.

Nexus 3 is not yet stable enough. IQ is perfectly stable. We have not had any stability issues with it.

It is currently not scalable. While we haven't encountered a scalability issue regarding Nexus IQ directly, but for maintenance and configuring there is a scalability issue because developers need to make modifications and reports. And those modifications must follow our workflow model, things like a code review and evaluation by a manager. Currently, this is not possible. They cannot make a request for changes in the software. There is no solution to contribute changes and that is a scalability issue. That is with respect to Nexus Lifecycle.

With Nexus Repository, we had a lot of scalability issues with version 2. With the new version 3, we tried to set up a certain type of architecture but it is not available. So scalability is an issue regarding the load, not the amount of data. We have been using Nexus software for 10 years now with very big storage and that is not an issue. But when the number of users increases, that's an issue. We are an open-source company, so we have many consumers of our artifacts, and that means there can be a heavy load on the projects.

Twelve years ago we tried other solutions, like Artifactory. But we quickly moved to Nexus. We may change the solution in the next month or year. It's a possibility. It depends on the pricing and whether the solution provides HA.

The main purpose of using the IQ solution was to have an efficient solution to spot and block security risks. We tested and compared a lot of solutions and found that IQ was the best and the most evolved. But a lot of it is not completed, it's still a prototype, from my point of view. That means it cannot yet be used exactly the way it is marketed. There are some features that are missing. But compared to other products on the market, it appeared to be the most accurate one.

My team was responsible for setting up the whole system. It was complex in the end because the way we wanted to set it up was not yet defined. It was not designed to be used the way we want to use it.

For IQ, the deployment only took a few days, but it's not usable yet. We set it but we are not really using it as we wanted to. For Nexus Repository, it took many months and many issues are not resolved yet. The fact is that IQ is not very useful without the repository.

Internally, we had a deployment plan, but a lot of those requirements were not met by Sonatype and we have had to work with support to make it work. But there are still remaining issues.

Sonatype assisted us with the initial setup and their help was average. It was pretty good most of the time, but sometimes it was not because we were outside of their defined environment. I feel that they are a little bit behind in this field, that they are missing modern requirements. So their support was pretty good, but not enough.

We evaluated Artifactory by JFrog. It also seemed very good, very similar. The other solutions we tried were not as good. Nexus and Artifactory were the finalists.

At the time, the UI was a difference between them, but that is not true anymore. The two are very similar now. The integration with development tools was very important; the ability to implement GitHubs, and so on. The fact of being open-source is very important to us; being able to contribute to and look at the code for reliability and quality.

My advice would be to look at your needs and the features the solution provides. In the last version they released, we were a little bit disappointed by the difference between the marketing and the reality. The product was not yet finished compared to how it was described. Aside from this bad aspect, it's mainly about good practices and looking at common, standard practices. Start with the basics and common stuff and try to evolve it eventually and change some things. Don't try doing something too much complex at the beginning.

The tool's default policies and the policy engine seem pretty good. We have been using it for so long that we are using our own policies. Regarding Lifecycle, we have not gone far enough with it to have a real opinion on it, although it looks consistent.

In terms of its integrations into developer tooling like IDEs, Git repos, and automated pull requests, we have not used it enough. It is a little bit too soon for us. It's a goal, but we want it to be done in a transparent way through the automation. It's not used yet because all of the developers are free to choose the tools they use, the IDEs, and only a few people are looking at this kind of stuff.

Internally, we have about 50 developers for Nexus Repository. We have five to 10 specialists for the Nexus IQ. We don't know many customers there are for Nexus Repository in our public sphere.

It is used across the whole company. It's a central tool and most people in the company cannot work without it. Everybody uses it either directly or indirectly or by proxy. Some people use it without knowing it, but all the technical people in the company are using it, so around 150 depend on it.

For deployment and maintenance of the solution there have been three people on my team doing this, but that was not the only responsibility of the team and they were not enough. We spent a lot of time setting up automation, including maintenance and deployment, and that made it scalable for three people in maintenance. We had to work on this. It was not provided natively, but now three people are enough.

I would rate IQ at six out of ten. That's very good compared to other products on the market.