The Application Monitor is a GREAT tool...but there are some things that we (and our Boss) wish we could do with it.
1. I wish that it were a little bit easier to add applications to multiple servers.
For example, YES you can add an applications to all servers with the 'Add to All Servers' command, but most of the time, I only want to add a certain application to a group of servers...with the 'Add to All Servers' this is a pain because I would have to go through and uncheck the 240 or so servers that I don't need to monitor...and I have to do this for every application. So it would also be helpful in the new 'Add application to Selected Servers' command that there would be a "Add another application" button, and then a "Done" button to close the window as apposed to just "OK" and "Cancel" that do one Application at a time. Or if there is an SQL script that can do this easily, that would be great too!
2. As mentionend in another post, the ability to pull App Monitor data into the Report Writer. Yes it can be pulled through SQL, then the data converted, but only if you know the tables well, (and SQL well enough)...even then, it is still not a Report that can be viewed/emailed etc.
3. Graphs/Better Views- I think there should be a view that would allow a comparitive graph with user definable variables such as date/time range, which apps to show, etc. I know all of the data is there, and I would like to view it similar to how response time/packet loss graphs are seen. I created the following chart just by dumping cooked chart data into excel:
4. Accuracy of polling data etc- It seems that no matter what I do, I cannot get the total CPU load or Memory load to match what the apps are saying. I would say that I also see a disconnect in polling times etc. For example, if I receive an alert that a CPU is at 100%, when I look at the node details, none of the apps show a large amount of CPU usage. This has not happened often, but my boss brought it to my attention when he did that exact thing.
Here is a better explanation of what I mean when I'm talking about missing data- In the screenshot below, the CPU LOAD values are gone, yet MEMORY USED values are populated. I don't think I have ever seen Memory values dissappear, but I could be wrong. This seems to happen fairly frequently...
They do show up...I am thinking it may have to do with the fact that I have the polling rate set for 1 minute for all servers. I had done this to try to keep the application data in sync with data that came from cpu/memory monitoring- to try to figure out what was causing CPU spikes, etc.
I'm wondering if I should spread the polling out further??
SolarWinds solutions are rooted in our deep connection to our user base in the THWACK® online community. More than 150,000 members are here to solve problems, share technology and best practices, and directly contribute to our product development process.