Home > Knowledge Base
> Failover Service
Failover Service
by ZoneEdit staff, edited by Mohan Kapoor, Supervisor, Himalayan Websites
To make our failover monitoring service work you need to have at least two web servers running with the same content, e.g. you use a site-mirror. We monitor these servers every few minutes with our failover service. Typically a web page is checked for a predefined text string in the first 500 bytes of the results. If this text is not found the failover service will consider it as a failure.
One of the following codes is returned after each check by the failover service:
- O - All sites are OK
- E - Some sites are on error, but some are OK
- F - All sites have failed, failure site has been activated
- X - All sites have failed, no failure site is available
- Y - All sites have failed, no failure site is configured
- U, W - Benchmark site failure, failover temporary disabled
- K - Timeout while waiting for connection
When the system detects that one of the servers is having an error, and the others aren't, it pulls the IP number out of the DNS list. If none of the IP's are responding and there is a predefined failure IP or web page, then the system points to that page. If the system can't get to all of the IP's, then it assumes that the fault is at it's own connection - and takes no action.
The system will automatically notify the staff at helloyou.ws when a site goes on error, and when the site is restored. If the system is unsure, because neither site is responding or because of network errors, then it always restores the primary site.
This effectively and safely keeps your site online - even if one of your web servers is down.
The average failure detection time is 5 minutes. This time varies depending on the speed of your site and the nature of the failure. Recovery times are faster, averaging 3 minutes.
The failover service is included in our Information Logistics, Managed Mail, Managed Processing and Managed Web Space solutions. See Domain Routing for prices.