StatsMan Chart Drilldowns

  • Article

As you probably already know, the root StatsMan page shows a great overview of your Skype for Business server environment usage and performance. Each chart on that page displays an aggregation of all the servers in your environment in real-time. So what do we do when we notice a spike or drop in those lines that is unexpected? How do we figure out which server is actually causing that anomaly?  That's where drill-downs come in.

Counter hierarchy

It's important to realize that there is an implied hierarchy of server populations in StatsMan, which is what allows much of the StatsMan interface to function as it does.  The hierarchy is laid out like this:

  1. * (Root population, all servers)
  2. Site
  3. Pool
  4. Server

If you're viewing a chart where all data is coming from the same server population, you'll be able to drill down to see all of the server populations in the next level below.  For example, if all series are from "*", you can drill down to all the sites, then drill down from a site to see all of the pools in that site, and then again to see the servers in one of the pools.

Drill baby drill

To the right side of any chart in StatsMan, there are a number of controls.  If all series in the chart are derived from the same server population, you'll see a drill-down option in that list of controls. The default view in StatsMan is a great place to start for this, since all of the charts will be using the "*" population.  Pick a chart that is interesting to you, and click the site drill down button.

You'll now be shown a page with the parent population at the top, and a chart for all of the sites below the parent. The scale of the child charts will sync together as they load to aid comparing their data.  You can also use the compact view button in the side bar to see more charts on the screen at the same time.

You'll notice that in the role view, the drill down button will change to pool.  You can now drill from any site to see all the pools within that site, and then again from pool to servers.  Note when you get to the server level, the population aggregation will be dropped since you're now looking at raw counter data from the server directly.

No limits

A great feature of StatsMan is that the drill down functionality is dynamic.  This means that you can drill down from any custom chart you create, as long as every series on that chart is from the same population.

 

As always, if you have any questions about this or any StatsMan functionality, drop us a line in the comments or on Twitter @SkypeStatsMan.

Comments

  • Anonymous
    May 13, 2016
    Glad you folks have a blog...love Stats Man! It's been a great help for monitoring the environment during the day, especially with uOne thing I had a question on...we updated to v. 1.1 a couple weeks back, and it looks like the aggregate SIP messages/sec value for all servers is off. With the previous version, we were seeing 1000-1300 messages/sec during the day for all servers. Since the upgrade, that number has dropped to 600-800 messages/sec for all servers...but, if you change the filter to each of the front end pools, you see 1000-1200 messages/sec.Is this being calculated differently now?
    • Anonymous
      May 13, 2016
      First - thanks for the feedback, we're happy to know it's working well for your needs!Regarding the values changing, note this is an average value on the default view. If you have some servers reporting zero traffic you will see this number drop when looking at the environment as a whole. Is it possible you have some newly reporting agents (because of the upgrade) that may be dragging the average lower? Try using the site/pool drill downs on that chart to see if there are some servers reporting low numbers. Alternatively, you could use the counter details view (Click the Counter menu just below the logo - choose Counter Details View, then choose one of the counters from the chart) and look for servers reporting very low numbers to narrow down the search.If you can't find a way to make the numbers make sense, let us know and we can look into it further. Thanks!
      • Anonymous
        May 17, 2016
        Thanks...I think you were on to something. If I recall, I don't think we previously had the agent on our secondary edge pool. We only have federation setup now, so the three servers in that secondary edge pool are at 0 msg/sec. That must be what's bringing down the numbers.I didn't realize that counter also encompassed the edge servers.