New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Best commercial monitoring system
AnthonySmith
Member, Patron Provider
in General
Hi Folks,
Just gathering opinions and ideas.
What is the best commercial/enterprise grade monitoring system?
In terms of monitoring I really mean uptime/alerting, I have tried statuscake, pingdom and uptimerobot but they are all pretty flaky and inaccurate or lacking information, pingdom being the better one out of the 3 by a million miles.
Money being no object what would you suggest (excluding build your own) ?
I am looking for something that has multiple levels of validation, not just a few pings were dropped, has better options in terms of monitoring specific services and actually gives some log info, and most importantly can be relied on to accurate 999 times out of 1000.
I am sure there must be something out there?
Comments
Virtualizor lounched a monitoring website take a look.
http://pinguzo.com/ < This?
Yeah
NodePing has been great for us and matches up with our in-house monitoring solutions nearly 100%. The ping tests are not as accurate as the port tests, but our data centers claim this is due to "ICMP de-prioritization". The number of false alerts for ping tests are much fewer than other services (they are rare, but do happen) but more importantly they actually alert us for actual downtime unlike every other paid service I tried before them.
Been using Hyperspin for my servers, and it's been very accurate, but I haven't tried much else...
I have seen virtualizors logging first hand, no chance I would trust them to provide anything reliable or useful for monitoring, appreciate the suggestion though.
@KuJoe Sorry I forgot to add nodeping to the list, I have tested that as well, to many false positives when I used them.
nixstats ? srvtrack ?
had a look at hyperspin just now thanks, looks interesting, cant see any mention of analysis logs or anything similar, do they provide the information they used to determine your XYZ was down?
Also $840p/month for 100 monitors is a bit high ... i know I said if money was no object, but at that price I would build my own in SL DC's
Even when you don't use the ping tests? That's pretty strange. Maybe @NodePing can help you out?
I have used StatusCake for a public monitoring page, I've never had any false positives with them. Only downside is that it might be more expensive than needed for some people if they require 1 minute intervals instead of the 5 minute interval that's the default on free accounts. I use a few self hosted scripts for my backend monitoring and alerts though.
Edit: Link to my public page is in my signature if you want to see what it looks like.
both sites only have a login page I cant seem to get any info/pricing.
Well to be fair it was a while ago, what is the reporting like in the event they do report something down, mtr reports, raw outputs from the checks with source locations?
Actually they seem to have seriously upped their game since I last used them, like the 30 second monitor and some of the features, might give the business plan a whirl!
No MTR reports but it tells you which locations were used to check before marking it as down along with the response (different depending on the test) and how long the check ran.
You can see an example of their public status pages here: http://status.securedragon.net and clicking on the name on the left will show you the public reporting for that specific "check".
Monitive.com is quite good IMO.
+1 for @NodePing. Shawn created an incredible product and very fairly priced. 10/10 would recommend.
I use Monitis at work. I tested a lot of allocations and that was the best of the bunch (for my use case, obviously). With a properly configured monitor I have (to date) never had a false positive. Not sure about pricing, but uptime monitors are reasonably cheap from memory, it's the others that cost the big bucks.
https://www.datadoghq.com/ have a trial. I tested lightly so cannot tell but worth trying.
Solarwinds N-able
Uptimerobot if you do not need all the nodeping features. Reliability is pretty much the same on the pro plan.
Monitive looks to be the winner, thanks!
Well just throwing ideas here, we use zabbix on our environment with different proxies that check our servers for uptime, packet loss and other server components such as raid issues, etc. Pretty good and it is free
Probably because both are in beta.
Hyperspin or HetrixByte are solid choices for Commercial Monitoring, although the paid UptimeRobot is also solid.
I have good experience with Nodeping - it has been solid so far.
+1 For Datadoghq.com
I use it for 150 servers.
Not too interested in something I cant get any info on or that is unproven, thanks though, ones to keep an eye o for the future.
Looks interesting, I will contact them for Enterprise pricing as I need 18 months worth of data retention minimum.
I guess they have got better then, I don't consider it Enterprise though as they generally use LES hosts for nodes iirc and don't provide any logging data per event worth having.
It's quite nice, you can lay out your whole infrastructure in it; automated tasks and etc. It's very commercialized for a-lot of enterprise companies that use them.
I paid them 2,000 annually a few months ago for 200 hosts. So far they're working great for us.
Here's a screenshot of a "metric" you can customize to add to your dashboard
Really simple to add your own network endpoint checks
@AnthonySmith
Not to mention, their API is really easy. I whipped up something to monitor solusvm nodes just from the solusvm master in like a hour to monitor everything(network,i/o,high cpu, etc) in 10secs intervals
We use Uptime Robot at our bankrupt company since it was the only one to fit the budget.