1 reply to this topic

[Not a Cloud]

I upgraded to Cloud service from Shared, thinking that maybe it would solve downtime woes.  I also thought that as I launch a few new sites this year I would want the scalability, responsiveness, and most importantly, autofailover.

Cloud is apparently not actually a cloud.

• Day one, there were 20-30 minutes of downtime.
  
• Day two, there has been 1 hour of downtime and we're still going.

Downtime does exist.  I understand this much.  A few minutes here and there isn't a big deal, especially not on a shared server.

However, this is supposedly a cloud service with autofailover.  If the site goes down, we're supposed to go right back up.  Right back up isn't 1 hour and counting.  

There's my testimonial.  As for a question, why is it that uptime is poor on a cloud service?  Secondly, if there is no autofailover, what's the benefit to the cloud service?  What does cloud mean to ASO?

[-ASO- Dan]

Greetings,

I apologize for the inconvenience experienced.  You are correct in which the system is designed to have an automatic failover in the case that a server goes offline.  Recently I was put in charge of managing the technical infrastructure here at ASO and am still feeling my way around things.  Once issues were seen yesterday with the cloud service I immediately noticed there was a previous configuration error with the setup which was preventing the automatic failover process to work correctly.  This failure affected approximately one third of the virtual machines on the system.

Since this has occurred I have corrected the configuration of the system, added additional hardware to better handle the load, and spread virtual machines throughout the hardware more evenly.  Through testing that was performed a few hours ago, the failover process is now confirmed to be functional again and I do not expect any further issues from here on out.  

If you have any questions or concerns feel free to let us know.