What is our primary use case?
We use it for network monitoring. We don't do a lot of systems monitoring with it. We predominantly use it for core switches, external network adapters, and other similar things but not for the endpoints or server hardware.
We are currently not using any of the automation functions.
How has it helped my organization?
It definitely assists with visibility into our remote and distributed networks. We don't necessarily use the Windows side of things, but we do have a couple of systems that are monitored. It is almost like a ping test or just a sanity check. From that perspective, it definitely helps. This visibility is 100% critical.
It helps keep device inventories up-to-date. It has saved us time because we have pretty much everything at a glance. It does allow us to prioritize what needs to be replaced or anything that would be necessitated by the device inventory, such as software updates or vulnerability patches. It definitely helps, especially with the end-of-life hardware. We're able to determine that and apply a device lifecycle to it.
In terms of helping our teams focus on high-value tasks and delegating low-level tasks to junior staff, in our network team, we do everything. There are all types of tasks that would be normally assigned to juniors. It definitely provides a lot more visibility and helps in delegating specific things. For example, when an interface is flapping or a port is shut down, it is a lot easier to delegate such a task. We're an MSP staff that doesn't necessarily deal with high-end network equipment. Turning a port back on is something most of us can do as long as we can log into a command prompt. Even the server admins can do some network tasks if need be. Within that, it does allow us to prioritize and state, "Okay, a senior network admin can figure out why this entire site is down," versus, "We need to update a switch."
Our mean time to resolution has reduced due to the alerting system.
What is most valuable?
We are seeing that the monitoring is very accurate. We are seeing that in terms of problems and solutions, there is a lot of functionality to it, such as APIs. So, you can dig down. You can dig deep into it. It does provide very useful metrics, and it has improved a large portion of our day-to-day operations.
The network monitoring portion of it has pretty sane defaults, and it is fairly good as a product. It is probably one of the better ones that I've used.
What needs improvement?
The visibility on the site itself is a bit of a problem. We do have the alerts panel, but there is no central monitoring. When we had requested how we would do this to place it up in the NOC and how we would view it and everything else, their answer was to use a third-party tool, such as Power BI. That was the response that we got. A front-end component to show the actual NOC operations at a glance is not present. That would be a major con in my opinion, especially for what we do as a data center.
The actual adding of networks, systems, and everything like that is fairly easy, but the problem that I have is getting the metrics out. Specifically, if you go to the Auvik webpage, sign in, and go through the alerts and everything else, they don't offer any plasma display with a red light, green light, or stop light indicating this device is in an error state, down, etc. To get around this, we have to use their API. I had to code an entire interface to work around that lack of information.
One of my major concerns or my major problems is the API hasn't always been super reliable. Sometimes things get broken. Sometimes it is down for a little bit. It doesn't seem to have the same reliability as their primary service, the actual web page itself. The API reliability is problematic when you apply a user account. I have a super admin account, and I have an API user that is a super admin as well. I create a new site as a super admin, and you'd expect everything to fall through, where the top level is the super admin and the subsites don't have access. We have network admins that create sites and DCOM sites and everything else all the time. When that happens, it breaks the alerts API and gives a 403, forbidden error, and that's across everything. If it can't access the top-level tenant, it just breaks the site. There are ways of counteracting that, and we're aware of the pitfalls there.
We have had the API function in erratic ways where we do filtration based on various criteria, for example, if a ticket has been dismissed, if it is in maintenance, or if it is critical. We have filters for all the metrics. Sometimes, we had a couple of tickets where it doesn't acknowledge those filtrations or the filters, which causes a little bit of a problem, and we have to do a little bit of a sanity check within our code itself. It almost seems a little bit like they do focus on the front end and making it visible, but it seems like the API is almost a second-class citizen.
For how long have I used the solution?
We've been using Auvik for about two years.
What do I think about the stability of the solution?
Because the API is what we use frequently, we've had various issues. It could use some work, but in the front-end portion of it, where I'm assuming most of the customers would be looking, we haven't had any downtime that hasn't been pre-planned and reported to us in advance.
What do I think about the scalability of the solution?
Its scalability seems to be fairly good. We're not having any real problems. With the way we're doing things now, it seems to be fairly scalable, and I don't think we'll have any problems.
In terms of our environment, we have operations predominantly in New York. Specifically, there are a few in Manhattan. We have a few in Queens and Suffolk County. We do have one location that is in Singapore, which is one of the smaller operations that we have, but it is predominantly located in the New York, Long Island region.
How are customer service and support?
They were very friendly. They were very good. Generally, if there was a problem, I was able to talk to an engineer on their side relatively quickly, which was a good thing. I was able to very easily prove the point that I had with the calls and everything else, and it worked flawlessly. After I was able to show them the output and everything else, they were able to resolve the problem. I believe they were able to resolve it after six hours or eight hours of having the call with them. That was a pretty good response time in my opinion.
I would definitely rate them a 9 out of 10. Getting a 10 is almost unheard of. All things considered, support is one of the better parts of Auvik in my opinion.
How would you rate customer service and support?
Which solution did I use previously and why did I switch?
We have been using OpsRamp. We got Auvik because one of our larger customers used OpsRamp and then changed to Auvik. In our case, my boss said, "Well, why don't we use it too? They seem to be really enthused by it." However, that customer doesn't use it with the same use case. They monitored a lesser number of sites and locations. Their use case is slightly different and their monitoring is not the same, so it works for them, but it doesn't necessarily have the same impact on us.
In terms of the consolidation of tools, we are still using multiple software types. Auvik is what we use exclusively to monitor network areas. We are currently using OpsRamp with which we are monitoring network hardware as well. We also used Kaseya, which was the worst software for monitoring anything. That was the reason why we immediately binned it as fast as we could, but we also have a couple of other different software. We are using an application manager. We do have Zabbix, and we monitor some things through that as well. That's mostly the ESXi and servers that are on-prem. We are a data center, but we also provide managed services as well. We have a lot of different systems within a lot of different operating systems and environments. Some are PCI. Some are non-PCI. So, we do use other software, and Auvik fulfills some of the same monitoring purposes but for different clients or different hardware.
In terms of time-saving by switching to Auvik, the OpsRamp software has some faults and after the actual interface that I wrote was deployed and started to be used by our NOC, there have been time savings. However, getting to that point took a little bit more frustration in setting up compared to some of the other products that we've used.
How was the initial setup?
We spent substantially less time with Auvik than with our previous solution. The initial setup was relatively straightforward, but my experience level is closer to DevOps than a traditional Systems Administrator. Between my own level of experience and my network team, it was fairly easy to get it deployed.
We were able to deploy it, but then we found that for our monitoring needs, it was a little bit lackluster. I had to code the webpage.
In terms of maintenance, with regard to the API and the coding work, maintenance is required, but it is infrequent.
What about the implementation team?
Which other solutions did I evaluate?
There were quite a few options that we looked at. It has been a while, but there was a large selection of software that we've tried, both on-prem and cloud-based. We did monitor or look at NinjaRMM and ScriptLogic. There was Nagios for the on-prem and Applications Manager from ManageEngine. We checked out Enable as well.
When comparing Auvik's cloud-based solution versus the on-prem network monitoring solutions, they serve different use cases, but the cloud-based Auvik has its advantages due to the fact that we don't have to have firewall ports opened. We can very easily monitor various devices and various client sites without having to be concerned about any leakages because we have the accumulator of the agent gateway and whatever the terminology that they use. It definitely has its pros and cons in the sense of firewall access, deployment speed, and monitoring aspects. We can apply a template across all different types of devices, and the scanning works perfectly in that sense.
What other advice do I have?
My advice would be dependent on how many sites you are monitoring and what you are intending on monitoring. For network equipment, Auvik is very good. For hardware and software, such as Linux, Windows, ESXi, and other similar things, it is very poor in those regards. That would be the major thing. If you are intending on having one tool to rule them all, I would probably steer you toward that limitation because it is quite limited in the endpoint monitoring and server monitoring, but it very well exceeds in network monitoring.
In terms of providing a single integrated platform, the API access to it is good. It does provide that, but the actual OS and software side of things that are not network devices is a little bit lacking.
Overall, I would rate Auvik a 7 out of 10.
Disclosure: PeerSpot contacted the reviewer to collect the review and to validate authenticity. The reviewer was referred by the vendor, but the review is not subject to editing or approval by the vendor.