What is our primary use case?
Our primary use case is to monitor the live game observability. When our game is live, we can see how the servers are performing and how players are connecting around the world. Generally, we just monitor everything to make sure the stability is running smoothly.
The secondary use of it is when we have things we want to dig deeper into, then we have a backup of the logging system. We can jump from a point in the metrics of the observability to a point in logging in which it happened, then observe what the server saw at that time in order to diagnose and fix whatever the issue might be.
How has it helped my organization?
Lockdown and COVID-19 made it impossible to go to work anymore. Having a central place where everybody could have a login, then we could share, take snapshots of graphs, or send people live data. Then, if anybody wants to know, they can just go and have a look for themselves. They don't have to ask us, "What is the state of the services like?" or, "What does the player population look like?" We don't have to answer those questions. People can just go and find the answers by looking at it themselves. Also, the customization means we can very quickly build graphs to analyze data or check something out. If somebody else wants something else that isn't there, then it is a very quick turnaround to add it very quickly.
Logz.io is useful for helping us to correlate between our log, metrics, and tracing data. They are integrated altogether, so we have our observability metrics dashboard with a link in there. You click it, then it goes to the logs. It is a holistic approach to having everything together.
It has definitely improved our troubleshooting: To have visibility on what is going on so we can identify things as early as possible. The alerting system is super powerful and hooked up to all our systems so we get alerts. Whenever there is an inkling of a problem, we know about it. That is just incredibly useful rather than having to wait for someone to tell us, which is theoretically what the old world used to be like: Someone would have to notice. Now, we can build alerts in it and the solution will tell us.
For anyone who uses it, Logz.io has increased their productivity. There is no restrictive policy on who can go and find out some basic information or look at graphs. Our design team, customer service team, engineering team, and executive team can all look at it from different levels. They can just go to the dashboard or graph where they want to see the information, without having to ask anybody. This adds a layer of transparency and demystifies a lot of stuff going on, which helps productivity for the sheer fact that anyone can see it. There is a lot of data in Logz.io to be consumed, but it is very clearly laid out and not confusing.
What is most valuable?
The most valuable feature is the tight integration they have between services that they provide. In our use case, it is great to see all the observability metrics, but also be able to go from that high-level view of important stats that we have deliberately recorded down into the low level of any particular time or server to see exactly what was happening at that moment, when we saw whatever we saw in the metrics.
They have the logging and metrics all in one platform. They are all tightly tied together, so we can look at the metrics and logs at the same time. They have features in there that tie the integration together. This makes them one product as opposed to three. This is super important to us. We were originally looking for something that did just the observability stuff, but then they also had something that did the logging, and we were like, "Well, this solves two problems for us at the same time." They have other features too, but they weren't within our requirements at the time. It was more valuable than having to find two products that did the same thing. It has such tight integration, support, training. etc. It has just made it super easy for us to use because we only have to write one integration.
We use Logz.io's Log Patterns and Drop Filters all the time. The Drop Filters are amazing and hugely beneficial. The Log Patterns are just so useful. We don't have to do any work to set them up. We can just start the main log, then I just say, "Could you parse this into this particular format?" They do. Then, I instantly have all this data coming through, formatted in a way that I need it, which is really useful. We didn't have to think about that upfront. We just look at it afterwards, and say, "Okay, this is really useful."
What needs improvement?
One of those areas would be allowing it to have public dashboards where you don't need a login. The reason is if we were in a studio right now, we would have a big screen TV with the dashboards on it. That would mean someone would have to log into it. Right now, it is not a problem. However, in the future, this will be. We will want these dashboards on big TVs so we can just visually see the state of the world. You have to log in right now to see anything. Whereas, it would be nice if we could have a link to a dashboard that we could give to anybody without having to log in. That way, we can put it into a kiosk. So, we could put it into a kiosk mode, then just have graphs on a big screen all the time without having to log in. That is a feature that would be great to have and a feature that I have asked for a couple of times.
I would like granularity on alerting so we can get tentative alerts and major alerts, then break it down between the two, but they do have a lot of good alerting stuff in there already.
Buyer's Guide
Logz.io
July 2022
Learn what your peers think about Logz.io. Get advice and tips from experienced pros sharing their opinions. Updated: July 2022.
621,548 professionals have used our research since 2012.
For how long have I used the solution?
I have been using it for about a year.
What do I think about the stability of the solution?
It has been super stable. There were some stability issues in Q1 this year. Other than that, it has been super stable. They were really aware of them and let us know when there are stability issues. There is a whole status monitoring page where we get alerts from if they are stability issues.
They are aware that we find it important that it has stability and uptime. I don't think we have ever had a cause where we thought, "This is actually bad."
It was just unfortunate that it did have a stability issue, rather than catastrophic. We put massive amounts of data through to them last yeah, and it was very tolerant and stable. When there was minor stability issues they were very attentive and their support team was fully available. Stability-wise, we have been very happy with it. It is certainly well within any SLAs.
What do I think about the scalability of the solution?
We scale up on the amount of data that we send them. If we want to, we could also just scale it back. This is good because we don't feel like we are strangled by the contract, costs, etc. We feel pretty free to use it as we need to.
How are customer service and support?
The technical support is the best. They have an app just embedded in their site, where you can just click on it, type in a request, and then they respond really quickly. We also have access to them via email, and they will respond super quickly. If we feel like it is super urgent, and we really want that red button to press, then they respond very quickly.
Which solution did I use previously and why did I switch?
I haven't got anything to compare this solution to right now, because any solution I have ever used has always been a home-brewed one. The value this solution has is measurable in that it is a fundamental thing we use now, and I look at it daily to see what is going on. I can't give you a monetary value to it, but it's definitely something that I wouldn't want to be without.
Previously, I never used a service like this. It normally was in-house. However, we didn't want to make another home-grown solution. We were really looking for something that could solve our problem, and we just happened upon Logz.io. This solution was actually recommended to me by someone I know at another company, and it was a good recommendation.
All home-grown solution used to take time to set up. We needed people with knowledge in that area of expertise, i.e., ELK stack. Generally, we needed a full-time person to integrate it, set it up, manage and monitor it. Even if that work tapers off a bit, there was always a maintenance cost to it. It is useful to be able to offload that into a service that has these features plus additional integration features that we don't necessarily want to write ourselves.
What was our ROI?
It has removed a lot of stress. Anybody can go and look for stuff. We don't feel that we have a black box so much as we have complete control over what we see and do. We can just put anything we want in, being able to see more (or less) information if we wanted. I can't imagine how I ever worked without it before. I never thought it would be that useful, but it certainly is. I wouldn't want to start another game project/product without it. It is crazy useful.
Logz.io has massively reduced our troubleshooting time. Before we had this in play, for some of the testing phases that we went through, we used to manually collect all the logs from the servers then run Python scripts over them in order to produce the data that we wanted. Now, we don't do that at all. That is potentially four hours a week per team until the end of a playtest that is just completely absorbed. So, half a person's day went away just by having this solution in place.
It is massively useful and great for testing. We can just go, find logs, and attach them easily. It has a very quick lookup. Whereas, before we would have to go, dig around, and find the server that the logs were connected to, then go to the server, download the log, and attach it. Now, we can just go straight to this solution, type in the log ID and server ID, and obtain the information that we want.
What's my experience with pricing, setup cost, and licensing?
You pay for what you need, and that is a good model. They are also quite happy to talk to you about your uses and your use case. They will even go as far as suggesting things that you don't need to do in order to save you money. At one point, I was quite surprised at how cheap it could be if we wanted it to be or how much they would help us manage our costs.
This was probably a good sales approach, but it is definitely true that they will look at stuff, and say, "We think you could save money by not necessarily doing this," and, "You don't need to do this," and, "The Drop Filters are the same. You will save money by doing this." Then, they put that feature in there to help you save money. This is a really good way of doing it because it also means that we can run as many services as we want without having to worry about scaling up.
Which other solutions did I evaluate?
It is more than comparable to other solutions. There are a couple other solutions that I have looked at and reviewed during our due diligence, and price was secondary to features in a lot of ways.
What other advice do I have?
Logz.io is great. For any other project we create it will be one of the staple products where I will say, "We should just start with this upfront, as opposed to trying to find another solution later or reassessing it, because there is no reason not to start with the product right now." Then, we will find its niche for whatever it is we are making in the future. I think this solution is super useful. It is a common go-to for a lot of the team who use it.
We use the solution’s AI-powered Insights feature, but not as heavily as maybe we should. It is definitely interesting to see what is going on and what the AI-powered Insights are telling us is going on. That can be really useful. We don't necessarily look at AI-powered Insights for troubleshooting and root cause. We are looking at it more to tell us something it thinks is going on. I will look at Insights maybe once a week, and say, "Oh, I get it. This is interesting. This is a new thing that seems to be happening. Let's have a look, see if we can dig down, and determine why it is coming up with that insight." If we see a lot of problems, then we go and look at Insights first. Normally, we are too busy trying to diagnose problems to think about Insights. While it is a good feature, we probably don't lean on it enough.
I would rate this solution as a solid nine 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.