Results 1 to 9 of 9
  1. #1

    Monitoring HTTP Support Providers

    My servers occasionally stop responding now and again. The fear I have is when sleeping or travelling there will be a huge time difference between downtime and recovery.

    I'm wondering if anyone knows or recommends somebody who can actively monitor http (could be alertra) and then within X minutes react.

    Reaction could be logging into Choopa or Pacificrack to submit a support ticket telling them to reboot the server.

    I know providers have monitoring, but I had 1 bad experience in my past when the server was down for 2 hours and when I finally realised the engineer said he was busy in another part of the DC.

    Better to be safe then sorry.

    Last edited by Gigaron; 06-12-2009 at 11:39 PM.

  2. #2
    If the engineer was "busy in another part of the DC", I doubt having a monitoring / management faculty look over your server help. The best monitoring in the world may be unable to get a datacenter that is always "busy" to respond. I personally would recommend looking for a new provider.

    I'd also recommend Plantium Server Management.
    From their website:
    24x7x365 (5 minute interval) Server Monitoring with Reboot
    If your server is down, we will follow your instructions to reboot it by either contacting data center or accessing reboot port.
    Our response time on reboots is Guaranteed within 15 minutes!
    We've used them before, and they know what they are doing. They offer full management with monitoring for 30$ or monitoring only for 10$.

    Personally, I enjoy the monitoring that Liquidweb provides, as if your server becomes unresponsive, we're notified within one minute of it happening, and they are already on their way to walk right out to the data center and reboot it. Of course this would require moving to a different data center, but if that is not a choice, PSM is your best bet

  3. #3

  4. #4
    Thank you. Good help. I'll contact and checkout this cron job.

  5. #5
    Why don't order management form your current web hosting company while you are on vocation?

  6. #6
    Join Date
    Dec 2007
    Indiana, USA
    All of the solutions mentioned are bandaids.

    Don't put a bandaid over the problem but fix the root problem. Why is your server/apache after time giving up or failing? Are you running out of ram? What is causing it to fail, fix the issue not the symptoms.

    A stable server should be able to run without shutting down for very extended periods of time barring reboots for kernel upgrades and other such security maintenance windows.
    Michael Denney - MDDHosting LLC
    New shared plans for 2016! Check them out!
    Highly Available Shared, Premium, Reseller, and VPS

  7. #7
    Join Date
    Mar 2003
    Moved > Managed Hosting and Services.

  8. #8
    Join Date
    May 2007
    I agree with Michael - you should be careful with bandaid fixes that ignore the underlying problem. In particular, a cron job that reboots the server when things get overloaded is pretty dangerous, as it could result in a loop of reboot-thrash-reboot-thrash-reboot that makes actually logging in to diagnose the problem very difficult. Also, forcing a reboot like that can result in database corruption, depending on how it is done.

    It sounds like a management service, either through your hosting provider or a third party might be a big help to you.

  9. #9
    Perfect advice from Panopta & MikeDVB - find the problem and fix it. Does't sound wise to reboot the box when your http daemon is not responding. What does the logs say? What does the system performance utilities (sar, iostat,vmstat...) say? The box should stay up and be accessible all the time (well, as much as possible), otherwise you start losing customers...

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts