You absolutely need to know about the servers you have under management in your estate. You need to know about the databases. Uptime, performance, behaviors, errors, corruption and a much longer list of items that will frankly fill this blog post, are very important to you.
I’ve been doing work in and around data and databases for over 30 years now (OMG, I’m so old). I feel like I know how monitoring works, what’s important, what’s not, what’s vital and what’s trivial. However, I know that you have different needs, different requirements, different problems. Those problems need different solutions.
Tell Us The Problems You’re Solving, and How
Or not.
I’m assuming that you’re doing the work you need to do to solve these issues, but maybe you just don’t know how to solve the problem. Maybe the problem isn’t even defined well enough, but you’re feeling the pain.
Redgate Software wants to know what you’re doing to monitor your systems. We have a short survey up that will only take a couple of minutes to fill out. We’re putting together a report on the “State of SQL Server Monitoring” based on the information you give us. We’re going to share that information with all of you that take part.
You’ll be able to see where your problems are unique and where they’re common. You can help to influence development of our products to help you solve your problems better, easier, or faster.
Please, if you have a couple of minutes, follow this link now and fill out the survey.
Also, please share this survey with your peers. Everyone has a unique perspective on what’s working and what’s not.
Thanks for your time and attention.
As I started reading the post, I was hoping you might be asking for folks to share unique monitoring challenges here, rather than offering a survey (though I hope that’s helpful for you and others).
If you’re indeed open to hearing one here (perhaps readers can chime in), I’m looking for an on-premise solution for tracking historically the top 10 processes running on a Windows machine (top by cpu, memory, disk, and/or network), tracked in something like 5-second intervals, and kept for at least a few hours if not days or longer (whether in its own DB or one we would point it to).
The goal is to be able to look at a point in recent time, when there were other indications of problems on a machine, to be able to see if some perhaps-unexpected process was the culprit for high resource use. (I see this often in my troubleshooting work, so it’s not just a theoretical/hypothetical interest.)
And I know already that a) there are cloud solutions that offer this (like New Relic Infrastructure), but again I seek an on-prem one. And I know that b) some DB monitoring tools do capture *total* resource use on the box at such intervals, but again I want the info tracked *per process*, for the top 10 running.
Finally, before anyone proposes c) “you can do that with perfmon”, no, you can’t do what I am asking. With that (and many tools that try to mimic it), you have to *identify the processes to watch*.
Again my point here is to help identify *perhaps-unexpected* processes that one may not see or think to have been watching. And I don’t just mean viruses or the like. It could be virus scanners, security scanners, network sync processes, and so much more.
I’ve looked around for such a Windows-based solution for a long time, to no avail (which really surprises me). If someone may point me to something I’ve just missed, I’d love to hear of it, and perhaps others reading this would appreciate it also.
Yeah, please. Survey sure, but if you’ve got some interesting stuff going on, we want to know it.