Technical Review: Deep Citrix Monitoring and Troubleshooting with Goliath

Recently I published a blog post titled: Citrix is Slow, Or is it Really? In that article I briefly mentioned using a product such as Goliath Performance Monitor. If you have XenApp or XenDesktop 7.x in your environment, you may rely on Citrix Director for monitoring and troubleshooting. If so, I’m sorry for you. It has been my experience that Director can miss quite a lot. At least in version 7.9 and earlier, which is what I have used the most. If I could put a value on the time I have spent over the last 6 months alone troubleshooting slowness issues which Director couldn’t help with, it would likely pay for a good monitoring product such as this two times over!

Over the last few months, I have been trying out Goliath Performance Monitor in my Citrix environment. The product covers a pretty wide spectrum of things. I likely won’t even scratch the surface on what it can do BUT I will share some of my experience and what I like about the product.

Before I started using the product for myself, something that stood out to me was how they presented their product on webinars. I have been doing a lot of research into many different “monitoring” products over the last year. Goliath provided a long list of specific customer use cases, some of which were very relevant to my current workplace.

In the above screenshot, you can see an example of a breakdown of session launch performance. Harkening back to the Citrix is slow paradox!

If you rely on Director or some of the other products out there you may get a display of the performance of different Citrix processes but very little around specifics of things like GPO or what is happening in that interactive session (after the session has been brokered).

You can see in the previous screenshot that Goliath provides a pretty comprehensive and detailed breakdown of the entire login process. You can get very granular and pinpoint a specific policy which may be bogging down your user’s login. It takes a lot of the time and effort out of troubleshooting.

Goliath provides a plethora of monitoring rules out of the box, which you can customize as you see fit. You can also create your own additional rules. These rules can help highlight obvious things like a server is down, an application has crashed or a server has become overloaded but you can also get setup rules based on specific events too. Other products also give event level logging but not every product does. This is a very important feature for us Citrix admins!

I had a particular use case for this myself. I found a bug in XenApp 7.6 CU1 LTSR which was causing hosts to constantly unregister and re-register. Director was oblivious that this was going on. It couldn’t discover the un-registration event and report it before the server had already re-registered. It made troubleshooting a nightmare. The issue was made all the worse by the simple fact Citrix does not automatically put a server in this state into maintenance mode so we had quite a high call volume from user’s complaining getting launch errors.

With Goliath, I can trigger off the event and automatically put the server in maintenance mode and then take it back out of maintenance mode if it has re-registered after 3 minutes. This could save a lot of heartache whilst waiting for the next CU1 or private hotfix!

One more additional point. The product does also display both event and system logs. It also keeps an archive of these logs, which is pretty useful in a PVS environment as you can go back weeks ago and find logs despite server reboots. It’s a lot more graceful than event log redirection.

The new topology view is awesome! I really like how it has been broken down into three distinct layers. Above you can see a topology view which shows my entire Citrix infrastructure showing at a glance the health of things like my VMware or XenServer backend, Delivery Controllers, License Server, Database etc.

The next layer shows my Delivery Groups and machine catalog, as well as info related to user logon performance and number of concurrent users per delivery group, as well as the breakdown of Receiver versions. Of course, you can view a lot of this info in Studio and Director but I think it’s pretty sweet to have it included in this layered topology approach.

Finally, a machine level view which shows the health at a machine level displaying data around CPU, Memory, IOPS, Network Latency and more.

Before using this tool, the setup of Director I was using also wasn’t reliable on alerting for things like high memory and CPU because it was pointing to a vCenter environment that contained more than just the Citrix environment. This provides only the alerts and topology that I care about. This is also dynamic. You don’t configure the topology yourself. The agent determines the type of server and plots it accordingly. If you add a new StoreFront server or DDC, it will show up instantly in this topology.

You can, of course pull general performance data across your servers, desktops and sessions. I could have lead with this but I figure all products do this much. While it’s obviously essential, I was more interested in what makes the product stand out.

You can also drill in on a per server basis or a particular session and get performance info. This was also an interesting point with Goliath. When I asked questions whilst researching, I asked about a particular problem I encountered and asked them how I could detect that using their product. They showed me in less than 2 minutes. Whilst with other products, navigating the data for the problem was pretty complicated.

You can also view real-time metrics for a given machine.

This can obviously be useful for real-time troubleshooting.

My use was focused primarily around XenDesktop and XenApp but I also used the product for some physical servers, workstations, my Hyper-V environment and some of my network storage. It’s very easy to use the guided wizard and auto-detection abilities of the product to get out of the box dashboards and features related to your key focus areas.

Aside from the great features, Goliath also have a keen focus on Healthcare and provide some integration with many EMR products. I found the data extremely value and the presentation gave me what was pertinent to what I wanted. It wasn’t a full overall view of everything in my environment but it was what I wanted to see, which a plus in itself is. Sometimes when you get too much data, it’s actually a detriment as it makes getting to the data that’s relevant to your issue much more cumbersome. I liked this because I had exactly what I needed in very few clicks.

If you are interested in reading more about general IT problem solving, check out the article HERE.

About author

Rorymon

Microsoft MVP in App-V. Citrix CTA. VMware vExpert. Unidesk Certified Engineer

Citrix App Layering

Over the years, I have posted quite a lot of content about Unidesk, which you can read HERE. While I did try out Unidesk with ...