Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!


Uptimerobot Alerts - how long does it usually take?
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.

Uptimerobot Alerts - how long does it usually take?

blackblack Member
edited July 2014 in Help

So I'm writing a script that gets alerts from uptimerobot. If it's a website that has gone down, it'll change the A record of that domain to point to a backup IP address via cloudflare API.

Currently, I'm testing notifications from uptimerobot. It's been 20 minutes and I haven't gotten a notification yet.... how long is it supposed to take?

Comments

  • NeoonNeoon Community Contributor, Veteran

    takes sometimes up to hours.

  • Give us a shot. Some say our alerts are the fastest anywhere.

    Thanked by 1lukesUbuntu
  • blackblack Member

    @Infinity580 said:
    takes sometimes up to hours.

    Ugh. Spent the past 3 hours coding everything for cloudflare and uptimerobot, only to find out it the notification can take hours :(

  • DylanDylan Member
    edited July 2014

    Uptime Robot and StatusCake have both been inconsistent for me. If I were paying for them I'd be pretty unhappy. For free they're okay.

    I've never seen either of them take hours, though sometimes one or the other does miss a short(er) downtime entirely.

  • blackblack Member

    @NodePing said:
    Give us a shot. Some say our alerts are the fastest anywhere.

    Thanks for the offer. I actually have my own set of monitor nodes, but uptimerobot is something a lot of people use (and I use it as a backup notifier). I thought if I used uptimerobot and if I open source my HA script, it'd target a broader audience. I'll contact uptimerobot and see if they can make it any shorter.

  • BellaBella Member

    I've always gotten them within 5 minutes of being down, as it checks every 5 minutes.

  • blackblack Member

    @Bella said:
    I've always gotten them within 5 minutes of being down, as it checks every 5 minutes.

    I turned the webserver off for 35 minutes and turned it back on because I didn't want to wait any longer. 40 minutes in downtime, I got an UP notification (saying my website has been down for 35 minutes), which isn't even the notification I was looking for. I contacted their support, going to try to work with them and see what's going on.

  • NekkiNekki Veteran

    I've found the alerts to be appalling slow at times. I was demoing it to my boss once as an alternative to pricey monitors, took 30 minutes to spot multiple sites were down.

  • Wow, most of the alerts I've gotten from them have been right after the box had gone down. Wonder if I should be using a separate monitoring solution...

  • ATHKATHK Member

    @Bella said:
    I've always gotten them within 5 minutes of being down, as it checks every 5 minutes.

    This.

  • geekalotgeekalot Member
    edited July 2014

    @black, @Charles I've usually gotten notifications immediately from UptimeRobot, it is in fact what I am currently using to notify Rage4 for failover purposes on a bunch of nodes.

    But, as I mentioned in a previous thread on this topic (I think it was the StatusCake thread), there are occasional hiccups: the last one was when the UptimeRobot admin panel said "maintenance mode" for a couple of hours and then when it came back up, the monitors were "frozen" (not reporting uptime changes) for about 20 minutes.

    I still use it, but I have been searching & researching like crazy to build my own solution -- I have at least 8 vps' that would make perfect distributed monitors (separate from the webservers) if I can just find the right framework. Hate to say it but I have had the most reliability with custom Bash + Cron (in fact, that is what I am using as the local "agent" for UptimeRobot to monitor on the webservers).

    No offense to all the various uptime projects posted here and elsewhere ... I just need more control, lightweight, distributed, secure, AND to monitor as many hosts & processes as I need, without ridiculous costs.

  • Spotty for me as well. Never seems to matter where the server is located either. I now use three different ones including the robot.

  • i have a php wrapper i wrote changes the A record gets called off uptimerobots web hook and it works great notify always happens within the 5mins

  • blackblack Member

    @lukesUbuntu said:
    i have a php wrapper i wrote changes the A record gets called off uptimerobots web hook and it works great notify always happens within the 5mins

    I'm using webhooks as well, but the notification comes in late. Looks like we have fairly similar projects going on :)

    Thanked by 1lukesUbuntu
  • lukesUbuntulukesUbuntu Member
    edited July 2014

    black said: I'm using webhooks as well, but the notification comes in late. Looks like we have fairly similar projects going on :)

    20+mins don't sound right, i did have to re-enable the notification on each node alert when they added there last feature somehow the setting got lost on each node alert...

    edit : just tested webhook notification worked fine

  • dedicadosdedicados Member
    edited July 2014

    it depends of wich email you use, i have gmail and other, in gmail, it arrives at much 5 min later, because Uptimerobot check is each 5 min.

    Never arrives longer than 5.

    also check with them, because if appears offline, then check is done from other place to confirm is down. (failover)

  • blackblack Member
    edited July 2014

    When uptimerobot detects the website is down image

    When I received web hook notification, this is from my nginx logs image It's actually an UP notification too, not a down one.

    Thanked by 1lukesUbuntu
  • lukesUbuntulukesUbuntu Member
    edited July 2014

    black said: When I received web hook notification, this is from my nginx logs It's actually an UP notification too, not a down one

    err thats weird, have you tested the response on the uptimerobot api for results to see if they match, and do you use custom request's with your webhook or just default?

  • blackblack Member
    edited July 2014

    I use a custom web hook that contains an extra get variable called "key." I'm receiving the correct parameters, it's just that I'm not getting them in a reasonable time frame. I also disabled cloudflare so it runs as a DNS only. In this scenario, it works fine. I'm working with uptimerobot's support staff to see what's going on.

    Thanked by 1lukesUbuntu
  • wojonswojons Member

    @black

    things like this are not very fun to code in mass and it really requires queuing alerts and checking at the same time. You first thing that is get a list of what to check then to run the checks. then to send messages for who is down check who was down and now backup and then to send the message its self. and considering ther are 5 steps in that idea if you have 100k for your product it can be a lot to do in a 1-5 minutes when ur not getting paid.

  • blackblack Member

    @wojons said:
    black

    things like this are not very fun to code in mass and it really requires queuing alerts and checking at the same time. You first thing that is get a list of what to check then to run the checks. then to send messages for who is down check who was down and now backup and then to send the message its self. and considering ther are 5 steps in that idea if you have 100k for your product it can be a lot to do in a 1-5 minutes when ur not getting paid.

    I agree it's not easy. They have some pretty nifty features as well, which is why instead of giving up or switching to another uptime monitor, I'm speaking to their support team and helping them debug the issue.

  • @Bella said:
    I've always gotten them within 5 minutes of being down, as it checks every 5 minutes.

    Mine have also always been within 5 minutes. It could be the difference of using HTTP vs Ping monitor type. I use Ping.

  • BellaBella Member

    @phantoms said:

    I use ping as well.

  • In average, 2 to 5 minutes, in my experience (several years).

  • wojonswojons Member

    @black said:

    that is good to here. i would love your feedback when i add uptime monitoring to my project.

  • praveenpraveen Member
    edited July 2014

    Uptimerobt is giving me very reliable alerts and also quick usually I get within 2 to 3 minutes. Really great product for free!

  • blackblack Member

    Bella said: I use ping as well.

    phantoms said: Mine have also always been within 5 minutes. It could be the difference of using HTTP vs Ping monitor type. I use Ping.

    I'm using http monitor as the code I'm working on targets websites.

Sign In or Register to comment.