jbiggley ✭✭✭✭✭

We just did a very similar upgrade -- everything exactly the same except for NCM. Here are a couple of things I learned: 1) Upgrade NPM first. It is the glue that holds everything else together. That means NPM primary server, additional pollers, additional web servers, etc. the move on to the next module. 2) As of Sept…

I think the very best alerts proactive. deverts nailed it with that design but I think that design of alerts that are proactive takes a collaborative approach from the monitoring team. You have to know the business and all of the moving bits in the business so that you can understand how to best serve the needs of…

Oh ya! Geeky goodness, here I come!

We played with the entire Elastic.io stack prior to make the investment in Splunk. (There, I said it!) We also used MQ (not Rabbit though) to handle the flow of data in the Elastic.io stack. There are lots of reasons that Splunk won out. Cost was not one of them

I'd really love to see every action logged to a file. Not just the ones AuditingActionTypes table. One of my big pet peeves is that adding/removing/changing/assigning/unassigning alert actions is not audited (as of NPM 12.1). That hurts when we are trying to trace who modified a specific shared alert action.

This is a great idea. So often we forget to 'watch the watcher' -- in this case, SNMP or WMI. If they aren't responding in a timely manner than all of the dependent monitoring and alerting won't be relevant.

I 100% agree with clubjuggle. I think that I've said it these forums before, but I am huge advocate for the monitoring team being a cost-savings centre, not just a cost centre. When infrastructure, application and other business teams need help wrangling a problem they should always be able to come to the monitoring team…

Love this line mikegrocket‌. Folks will probably get tired of me repeating myself, but I will hire for curiosity and drive to root out and fix problems any day of the week. Mindful curiosity ought to be the superpower of every enterprise monitoring engineer mikegrocket wrote: Adopting this attitude certainly wouldn't hurt…

Bonus points if it is Solarwinds-branded wrapping paper!

You are correct. Our standard is 7 minutes As adatole​ is know to say, salt to taste. Thanks for the cool query too!

After reading this thread I might need to surrender my MVP status. You all ROCK! Holy cow! The level of awesomeness here is amazing. Humbled to be included among so many deserving MVPs. I seriously need to spend more time on the care and feeding of my environment and less time on the "let's make new and shiny"

You incorrectly assume that this customer was able to articulate business services or that they were doing application monitoring. When you are ripping out CA Nimsoft and displacing WhatsUp Gold going from go-kart to Cadillac requires some work that isn't exactly covered in the SolarWinds 101 guide. But I 100% agree. PS: I…

Amen. With nearly 13,000 nodes under management in our environment we need enterprise-scale functionality. Admittedly, I've just learned how to do things with the interface available but I would love to see some better tools and functionality for large-scale enterprises. I bet meech​ would be very interested in that…

If you think racing go-karts is dangerous, wait until we go paintballing

IMO, the very best compliment one could every receive is repeat business. As a consultant in a previous life, the fact that previous customers still contact me asking if I am available to do freelance work is worth its weight in gold. (No, seriously -- you should see my freelancing bill rates ) Of course a cold Dr. Pepper…

I really like the new layout of the questions -- kudos to the design team. Makes my eyeballs happy

I'm interested to see how startups influence the idea of Software Defined Data Centers. As the definition of a data center begins to change, so will the storage technologies that support it. Swinging the ship of a massive supply chain makes the pivot harder when storage (and other tech!) suddenly shifts.

Strange times we live in! Thanks for the quick response maria.bungau

Let's just go with "it didn't happen". Tragically *this* did happen. Maybe I should have worn a cape? Computer Problem Solving Team circa 1993

I reference you back to aLTeReGo​ and his comment about using the SQL monitor. You could do something as simple as SELECT count(nodesdata.nodeid) FROM NodesData and execute that application monitor every X minutes. (Unless you are adding nodes like crazy, I would say once per day would be fine.) You can also find the same…

Thanks for the clarification. I'm definitely geeked out by the prospect of DPI but I'm still not convinced that I would leave it running 24x7. That said, if I had the computing resources and manpower resources to respond to a given trigger then I would definitely add it to my arsenal of data. He who has the most data wins.…

Here is a nice little primer on zombie VMs (aka orphaned VMs) that should help admins that don't have a lot of virtualization experience. http://www.virtualizationsoftware.com/classification-detection-zombie-virtual-machines-tools-use/

After the last two days of 'in your opinion' questions that is a touchy subject

Since I want to be able to backfill dev with some tweaks we made to our prod view -- I want this now. Is that the kind of feedback you wanted?

I didn't even think about this when answering the question but I second Jay Garcia's comment. With 16+ pollers and multiple products, some of them virtual appliances, keeping things up to date can be a little hairy. Doing it in the right order is even more mind-bending.

Is there a crown and a scepter? Thanks for the congrats!

I am paranoid as the next person but I take the tact that any time I sign up for rewards card (can you say Thwack! points?!) or any other type of system that provides a reward of information that I am selling my privacy. I guess, in the end, I am OK with it. An active invasion of one's privacy (telemarketers) is much…

What you're looking for is this line FROM [dbo].[InterfaceErrors] ie ie is just an alias for [dbo].[interfaceerrors]. Instead of typing SELECT [InterfaceErrors].[InterfaceID] You would just type SELECT ie.[InterfaceID] Shortcuts are great for writing queries, horrendous when you are trying to reverse engineer queries,…

The one thing that confuses me (and probably a few others) when working across different monitoring platforms is the granularity of the data. If it is polled data, how often is it polled? Every 5 minutes? 1 minute? Streamed!?! What does that data look like if it isn't 'streamed' data? Is it min/max/avg for the interval or…

It's a great justification for our obscene bill rates