I'm starting a discussion to express grievances about the Sonar Discovery tool for the entire Thwack community. It appears that larger environments have multiple issues concerning Sonar.
It's obvious that Solarwinds expects us to deal with a sub-par tool that is so critical to discovering nodes for Management and Monitoring.
Full disclosure, I am not a Solarwinds Admin, I am a Network Engineer. My job description does not include learning the in's and out's of Solarwinds.
IT shops for larger environments do not always know what is occurring at the farthest reaches of their networks, especially when local IT is thrown into the mix.
I didn't choose Solarwinds, it already existed at my shop. Local IT staff add devices all the time that will never be discovered due to Sonar limitations, but guess whose getting chewed out when those devices go down and nobody at Central was notified. Upper management only cares that the issue was not reported and not resolved in a timely manner, they do not care for excuses about software the company has already paid for.
I hope that people express their grievances with Sonar and that Solarwinds/Thwack pays attention.
- The arbitrary 10 hour max time limit imposed on the discovery process. The seed router function exists for a reason. I should have every expectation that adding a seed router to the discovery process and selecting all found routes will ensure that Sonar will discover all nodes until every subnet selected has been searched.
- The discovered node purge when the above max time has been reached. If you can't/won't fix the arbitrary timeout, then at least design Sonar to not purge the discovered node database. I'm not sure Thwack/Solarwinds knows how frustrating this is when doing a full network scan.
- Better custom dates for auto-discovery. The current custom date selection is atrocious. How about a start date and time, and the ability to select run every 1 day, 2 day, 1 week, 2 week, 1 month, 2 month, etc... I suggest Solarwinds look at Bluecat IPAM for inspiration. The current run every x-hours makes no sense.
- Discovery is slow. Latency for me averages less than 30ms between the Poller and any point on the network, but the discovery process takes to long. A side-by-side SNMP sweep comparison between Lansweeper and Sonar on the same exact subnet, on the same exact ESX host shows Lansweeper mopping the floor with Sonar. I can't believe that Solarwinds would allow Sonar to be that resource intensive.
I appreciate flamers to not post. Post your "solutions" on other threads. This is for legitimate grievances that have gone unanswered for far to long.
To Thwack moderators, this discussion is not for asinine workarounds, and/or broken promises for future functionality. Don't ask me specifics, you already know the limitations of your software. Take it all in, and get the designers to fix it!!! Please don't take this personal, it's just pure frustration from the software, and interactions between between users and moderators viewed on the Thwack forums.