jbiggley ✭✭✭✭✭

Back in the BLL days (Before adatole​ Left), we were running our Orion DB on a shared storage frame that was configured RAID 5 but "emulated" RAID10 with some pre-caching. We routinely crushed that shared storage environment. In the early months of ALL (After Leon Left), we migrated to an All-SSD array. Much better. Just…

What sort of impact is that going to have on the database? I know that NTA was recently peeled off to its own DB to help ease the pressure on the core DB. Just wondering how this will scale especially for large enterprise customers.

Budget 20% of the purchase price for maintenance renewals annually. That is pretty standard.

We run a single instance with nearly 98K elements, 25K application monitors across 17 additional pollers. I have been resisting the urge to split this into 2 distinct environments since I started 2 years ago. Growth and retention at scale, as cobrien​ said, are instrumental for SolarWinds to make the leap from "Yeah, we…

Sweetness! Can't wait to get all of the geeks in PEI together for some Solarwinds goodness! Best Christmas present EVER.

1. Is there an all of the above option? 2. Dedicated DBA team 3. DPA and SAM though we don't have the SQL AppInsight enabled yet, 4. See answer to question 2. (That's my cheater way of saying no idea!)

Honestly -- can we please make this happen so that zackm​ can stop making some many stinkin' alerts in our environment

Of course I'd be amazingly happy if we could see IBM XiV support too (hint, hint!)

Love the look of the new Orion interface and can't wait for NPM 12 to hit the streets. As one of those participants in the UX design sessions for this re-work, I echo the comments thus far -- great job meech​, katieb​ and the rest of the UX team. familyofcrowes​ nailed it with his comment on the importance of the UX. I…

Ready Player One got a call out in a recent Wired article on VR. Must add it to my books to read list.

I've taking university classes through Coursera when I find that I want to learn something that really stretches my brain. I'm currently taking a course from Johns Hopkins called The Data Scientist's Toolbox. It explores data and the how and why of analysis. Expanding horizons and indulging your passions will keep you…

'I wonder how long we can bump this'-bump

Sometimes my urge to be cheeky overrides my desire to win Thwack points. Today's question was quite tempting.

I see the role of Head Geek as one of building trust. The Head Geek is the spokesperson for the entire Solarwinds brand but, in addition to the brand, the company, the people, the technology and, I dare say, contributory to the trust in Enterprise Systems Monitoring in general. Being a Head Geek means being honest (and not…

What if I already wear a cape? <insert awkward silence>

+1 for silverbacksays​. My vision of the future for NetPath is a map view of all of the endpoints and their status checks. If one of those points happens to be yellow/red then you could drill into it and investigate which service was generating the path degradation. I think that would be especially valuable for outside-in…

Check Cisco's documentation for CDR configuration and file storage.

7 time zones -- and here I am whining about 1 while visiting Ohio this week What a trooper, indeed!

This is a killer explanation of DPI. Thanks darragh.delaney for taking the time to lay all of that out. If I opened the DPI rabbit hole I might never come back out! It's like the first time I saw Netflow data --- oooooo sparklies!

Holy smokes it wasn't just a bad dream, Leon Adato really did leave me after 2 weeks of working together to be a Head Geek at Solarwinds. Our loss, Solarwinds gain! It couldn't have happened to a nice guy. I look forward to that same quirky humour and generally geekery that have marked the past two weeks to come through in…

A comment and a question --- First, the introduction of the AppStack view (which I think is powered by ground unicorns and other magical creations, BTW!) is forcing us to re-evaluate how we do application monitoring in our environment. It isn't just about 'tell me what you want to monitor in an application when it fails'…

I have to admit that I'm struggling to recognize the use-case for an always-on DPI system that *wasn't* an appliance. Don't get me wrong, I'm definitely a Solarwinds fanboy but I also know when some things are better left to dedicated hardware. In my previous life I had a client that need to monitor 4 distinct networks…

Yeah, we're in the same boat. 20 CPUs, SSD RAID10 and RAID 1 arrays in our DB server for data and logs respectively. 768GB RAM. What we found, at least in the DB schema with NPM 11.5, is that some queries related to UnDPs are very inefficient as per DPA. At least when we get to the size of our single instance environment.…

arodrigo.oliveira@gmail.com This is a known "issue" with NPM, but it is based on some decent logic. If the SNMP service restarts then it was likely due to a server reboot, but I agree that it does cause some problems. I would build a custom poller (universal device poller) for the hrSystemUptime OID (OID repository -…

Count limits the number of processes and timeout just keeps things from getting out of hand. I would add them and ensure that the timeout value is less than the delayed time for the action that needs to send the process details. For example, if you have a 60 second delay then make the alert action 90 seconds (or something…

Just an hour or 2 per week. Setting it up takes a little bit of work but the ongoing care and feeding is pretty lightweight.

Eww, you said SCOM We actually leverage SCOM to help with collecting some very specific data that needs to send changing keys. I think that is the lesson every monitoring engineer learns -- there is no single tool for every situation. Sure, you can pick a platform to do most of the things you want to do but you'll almost…

We just did a major DB upgrade this weekend and didn't see any chance. Scratch the DB as the cause. I ended up adding a quick and dirty batch file to the trigger actions to measure the time it takes to process the results for the process query. @ECHO OFF echo %date% %time% >> WinCPUCritResults.txt…

The issue with actually with 11.5.X. I could not get it to work after we upgraded from 11.0.x to 11.5.x. Upgrading to 12.x is the only option at this point. Or use reporting outside of Orion to get and expose that data. Your problem with "It works in SSMS but not alerts/reports/SWQL" was the same problem I had.

Hey rsprim​, F5 pool monitoring was natively supported in NPM 12.0.1. Have you looked at the available monitoring instead of this custom SQL query? This was really a stop gap put in place my adatole​. I also know it broke after the 11.5 upgrade so you may want to install the latest/greatest version of NPM and then look at…