Hate to be that guy… I’m still rocking Nagios…
2 Likes
PHP Server Monitor + LibreNMS (incl. Nagios plugin) + Smokeping + uptime robot.
2 Likes
Also using Freenom for the .cf? Hehe 
1 Like
You bet’cha
1 Like
But rsyslog can do that tho…
Yeah, one line in rsyslog and all system logs fly away to Graylog <3
I was just being a dick, but I plan on looking further into greylog. Thanks for the tip.
1 Like
Hm. Technically speaking, wouldn’t HetrixTools be classified as a “self hosted” tool for you?
5 Likes
It has to be. I was teasing someone else about “blaming their tools”, and @Andrei jumped into the comments about it being HIS tool, rather than the person commenting.
E: 100% across the board this week. Then, ns1 networking bounced a few times for no reason several times afterwards. I’d still like a way to centralize testing on HT before it sends me “END OF THE WORLD” notices, but I’m quite happy with it.
2 Likes
Technically true indeed, it’d fall under both the ‘HetrixTools’ and ‘Self-hosted’ categories, since it’s both
Guilty as charged. 
Any more info on how you’d see such a feature function? Sort of an escalation feature? Or…?
2 Likes
Well, not explicitly, but having your services able to pool together and say ‘well, this isn’t working in Atlantis, but it works fine in Constantinople, so it might just be local’ would be awesome. I literally got paged 5 times about ns1 going down within an hour because a local monitor was having shitfits to that location.
ns1:~$ uptime
07:48:39 up 274 days, 18:04, load average: 0.00, 0.00, 0.00
Its’ just fine.
All locations data is being pooled before any notifications are sent out. Just one location facing issues cannot trigger an alert under any circumstances.
I’d suggest you look through your Location Fail Log because I’m seeing a lot of locations timing out towards your NS1 monitor, all trying to connect multiple times within the same minute. Different providers, different geo-locations, trying multiple times, virtually impossible to be a false positive:
And your Network Diagnostics (which are collected after your monitor is declared as being down) also show network issues from multiple locations.
The ‘uptime’ command does not take into account network connectivity issues to/from your server.
1 Like
I really appreciate the work you did, but it was more of the frequency of it showing, and not across all monitors that I was requesting possible work on.
Shit, you’re my only SaaS other than MailCheap. That should mean something about your product.
3 Likes
Self-hosted SaaS
2 Likes
Jumping on the bandwagon of features suggestions I’d really like to be able to setup recurring maintenance slots. I ran my Nextcloud backups every day at same time, so it would be good to be able to set these as maintenance slots and not be alerted about the outage.
A mix of HetrixTools + UptimeRobot, plus recently put up an Observium instance to play around with that a bit.
1 Like
Smoke signals and a monkey that flings poo when he sees downtime. Great setup!
2 Likes
Hold on sport- I haven’t signed the contract yet.
2 Likes
I use UptimeRobot for most things (status page at https://uptime.d.sb/), but I’ve been meaning to move some of the monitors to HetrixTools. I also use Prometheus + blackbox_exporter + Grafana to monitor ping times for a few servers.
1 Like
Hey, if this is in regards to HetrixTools, you can schedule your maintenances with a simple cronjob to wget or curl our API’s maintenance endpoint:
4 Likes