Enterprise Monitoring

bledd · 27 Jul 2019 at 16:15

Jumping on the back of this thread, any suggestions for vcenter / esxi monitoring?

We have What's Up Gold. Not looked into how much that can do for us yet.

KubyNeeTees · 28 Jul 2019 at 17:08

bledd said:
Jumping on the back of this thread, any suggestions for vcenter / esxi monitoring?

We have What's Up Gold. Not looked into how much that can do for us yet.

Zabbix is pretty darn good. ZenOss was also decent last time I tried it. All depends if you want an out of the box solution or willing to work on something a bit more open source / custom made (like a Grafana, Telegraph, etc) setup.

It irked me massively when they went from ESX to ESXi and dropped full snmp agent support. Literally spent 6 months trawling through new providers and OID's trying to get metrics out :mad:

#Chri5# · 29 Jul 2019 at 10:42

bledd said:
Jumping on the back of this thread, any suggestions for vcenter / esxi monitoring?

We have What's Up Gold. Not looked into how much that can do for us yet.

SolarWinds have a specific module for virtualisation management - VMAN;

https://www.solarwinds.com/virtualization-manager

Spammeh · 29 Jul 2019 at 23:25

Had a demo of Dynatrace at the end of last week, and I have to say, rather impressive - especially if you're very developer / code orientated. Sadly for us, its lack of support for physical infrastructure (routers, switches, etc) is somewhat of a deal breaker for us, so we're back to the drawing board.

Have demos of AppD (which I suspect is similar) and ExtraHop early next week.

#Chri5# · 30 Jul 2019 at 11:30

Spammeh said:
Had a demo of Dynatrace at the end of last week, and I have to say, rather impressive - especially if you're very developer / code orientated. Sadly for us, its lack of support for physical infrastructure (routers, switches, etc) is somewhat of a deal breaker for us, so we're back to the drawing board.

Have demos of AppD (which I suspect is similar) and ExtraHop early next week.

What are you after? IIRC (as we don't sell that much AppD), it doesn't look to target SNMP (etc) for infrastructure monitoring. I've seen some community extensions but not tried them myself.

I'm a big fan of ExtraHop but again, that's possibly not the right solution for you given it's a passive wire-data solution. It can offer up a bucket load of network related metrics eg RTT, TCP zero windows, re-transmits etc etc but it's blind as to the health of a switch or if you want to see something like % interface utilisation.

Spammeh · 31 Jul 2019 at 22:47

#Chri5# said:
What are you after? IIRC (as we don't sell that much AppD), it doesn't look to target SNMP (etc) for infrastructure monitoring. I've seen some community extensions but not tried them myself.

I'm a big fan of ExtraHop but again, that's possibly not the right solution for you given it's a passive wire-data solution. It can offer up a bucket load of network related metrics eg RTT, TCP zero windows, re-transmits etc etc but it's blind as to the health of a switch or if you want to see something like % interface utilisation.

We're after a bit of a mix of infrastructure monitoring (health of switches, routers, hosts, etc) mixed with the user experience, service driven approach. Dynatrace was superb at the application level, but lacked any infrastructure monitoring. By their own admission - they are purely focused on public facing web monitoring, whereas I need this for our public facing sites, and devops teams, but also to monitor our services for our internal business (230+ sites, 3500+ users).

I think App-D is gonna fall short in the infra monitoring, extrahop I suspect will lack in the granular host level monitoring - so I'm struggling at the moment

A[L]C · 1 Aug 2019 at 13:44

Spammeh, sounds very similar to our requirements. Drop me a trust and I'll send you a copy of our completed vendor questionnaire

bledd · 27 Sep 2019 at 13:34

Installed Observium this week, very impressed with it (free version)

https://www.observium.org/

Moving to the paid version today.

Found a small bug with the dashboards, emailed the author and he offered the fix via email and said it's already deployed in the paid version.

Super easy to install on Ubuntu / Debian

https://docs.observium.org/install_debian/

After installation, adding each device, we had graphs for devices within around 10 minutes.

Already handy for identifying a CCTV camera that had wildly wrong settings which meant it was using far more bandwidth than we'd want it to. We have the graphs displaying on Raspberry Pi in the office on a couple of screens.

Nate--IRL-- · 27 Sep 2019 at 14:03

Although I'm a big PRTG fan - it let me down when it came to monitoring our SANs - I've used Stor2RRD https://www.stor2rrd.com/ for this instead and have to say for free software it is very very good. The list of Compatible Enterprise SANs is impressive. Like the post above, I discovered a bug monitoring my particular SAN, and had a hotfix in 3 days. Can definitely recommend.

Nate

droyden · 2 Oct 2019 at 20:39

I have deployed lpar2rrd (by the same people as stor2rrd), as above very good. Forum support (currently unpaid) has been more responsive than many big USA listed companies who are certainly not unpaid!

Would recommend deploying the OS agent even if you are using vmware as it does offer additional information.

DRZ · 24 Oct 2019 at 22:09

Depending on what you want/need, at a smaller scale LibreNMS (forked from Observium) is absolutely excellent for what it is and if you know what you're doing it scales quite well too. For true global enterprise scale stuff you need a mix of tools. We use a lot of the big names mentioned in this thread - Splunk, Dynatrace etc etc plus some home-grown tools and heavily customised COTS products.