Ask YC: What tools do you use for notifications of server/site uptime? | Hacker News Reader

Ask YC: What tools do you use for notifications of server/site uptime?

16 pointsshafqat22 comments

Wondering what the best tools for notifications are. Specifically, something to monitor the server and the site to make sure its up. What do people use? Is it best to use a built-in notification tool on the server for the site uptime monitoring?

View on HN

We use a monitoring service called Wormly... http://www.wormly.com/

It's run by a friend of ours who used to work at Sitepoint and was frustrated by the limited options for good, affordable, distributed monitoring. He's been running it for over 2 years and is growing fast.

It's a great service and recommended as a YC-style startup worth supporting.

We use Pingdom (http://pingdom.com)

Simple but it works.

I use two different scripts:

1. dwatch http://nubyonrails.com/articles/2006/03/29/surviving-rails-1...

2. a PHP script that accesses a URL on the other server; if the server does not respond as expected, it sends me a SMS and emails the server admin

To monitor more than uptime and service availability, the high end of this is automated performance measurement analysis by Keynote Systems (keynote.com). They have a lower end product called http://redalert.com/ (formerly its own service) which is comparable to pingdom.

Cacti and Pingdom (www.pingdom.com) for external monitoring.

I use Monit on my server to make sure my services are running. It also restarts them if they fail.

I use a free uptime service to make sure that my servers are reachable: http://uptime.openacs.org/uptime/

Python script + e-mail to your cell phone number.

I've used nagios and before that, ganglia for some monitoring. I've been looking into cacti lately for monitoring qmail message status and an asterisk installation for the number of active channels.

Many of these systems use rrdtool and even though I've used that tool directly before, with Cacti at least its a bit tricky to get familiar with the way the gui and the different templates are interconnected. The look and feel sure does seem polished though.

Wow, thanks for all the great feedback. We're going to try Pingdom first. I also liked nagios for its ability to restart services automatically - very cool.

If anyone wants an invite to NewsCred, shoot me a mail: shafqat at newscred.com. Its not much, but thought I should say thanks!

Monit! Highly recommended. You can easily set it up to monitor local daemons, watch disk space, memory, external systems and more. Monit can also restart failing daemons.

http://www.tildeslash.com/monit/

GroundWork (http://www.groundworkopensource.com/). It uses Nagios under the hood, but I wouldn't want to set up Nagios by hand (without the GroundWork config tools) at a largish site like ours.

Pingdom and Alertra. Alertra has had a perfect record for us for three years, meaning no false reports over that time (as far as we can tell). A bit more expensive, we use a single Alertra alert and multiple Pingdom alerts.

I've been using hosttracker.com for years because they have a 2-site free service.

nagios

Is there a way to configure Nagios to send MSM text alerts when your site is down? I haven't fully dove into the documentation yet.

u mean SMS text (not sure what's MSM as you put it). If SMS, yup, we're doing it with Nagios.

rackspace provides support for verifying our sites are operating properly.

in addition we use monit and then finally we use wormly as a catchall

Nagios and Pingdom

Monit.

Hyperspin

munin and nagios

pingdom

Reply on news.ycombinator.com