Downly Wiki

At present, our DNS Failover and Monitoring system supports the following monitoring methods:

ICMP Ping: The system sends 12 ICMP ping checks to the specified IP address, with 4 packets of 64 bytes, 4 packets of 512 bytes, and 4 packets of 1024 bytes.

15% Threshold Ping: The check is marked as "DOWN" if 15% or more of the packets fail.

25% Threshold Ping: The check is marked as "DOWN" if 25% or more of the packets fail.

50% Threshold Ping: The check is marked as "DOWN" if 50% or more of the packets fail.

Web Checks: Requests are sent to the specified IP using either HTTP or HTTPS, with options for customizing hostname, port, and path. A check is marked as down if the request times out, the response code is not 200, or the response text does not match a specified custom string.

HTTP: A request is sent over HTTP, and marked as down if the request times out or the response code is not 200.

HTTPS: A request is sent over HTTPS, and marked as down if the request times out or the response code is not 200.

HTTP with custom string in response: A request is sent over HTTP, and marked as down if the request times out, the response code is not 200, or the response text does not match a specified custom string.

HTTPS with custom string in response: A request is sent over HTTPS, and marked as down if the request times out, the response code is not 200, or the response text does not match a specified custom string.

The system supports the following monitoring types for TCP, UDP, and DNS:

TCP - The system establishes a TCP connection to the designated IP address on the specified port. If the connection fails, the check is marked as DOWN.

UDP - The system tests the designated UDP port on the given IP address. If the connection fails, the check is marked as DOWN.

DNS - The system performs a DNS query for the selected hostname and query type on the specified IP address. If the response does not match the expected result, the check is marked as DOWN.

To configure monitoring regions, you have the following options:

Global: The default option that monitors your failover records from all nodes globally.

Europe: Monitors your failover records only from nodes located in Europe.

North America: Monitors your failover records only from nodes located in North America (USA and Canada).

These monitoring regions are useful when running GeoDNS and you want to monitor your nodes from the closest locations.

To be declared as "DOWN," the DNS record must be identified as "DOWN" by two separate checks from two different locations (check nodes). If only one node reports it as "DOWN" while others report it as "UP," the DNS record will not be deemed "DOWN."

The "DOWN" event handlers determine what will happen to the monitored DNS record when it is detected as "DOWN." The available options are:

Monitoring only: No changes will be made to the DNS record. You will receive an email notification if the DNS record is detected as "DOWN."

Deactivate the DNS record: If the monitored DNS record is detected as "DOWN," it will be deactivated. This option is useful for Round-Robin and GeoDNS configurations.

Replace with working IP: This option allows you to define up to five backup IPs. If the main DNS record IP is detected as "DOWN," the system will check the backup IPs in order, starting from the first and proceeding to the last. The first IP that is detected as "UP" and working will be set as the DNS record.

The DNS record is considered UP if it's detected as such from two separate checks in two different locations. If only one check reports UP but others show DOWN, the DNS record won't be marked as UP.

The UP event handlers determine what happens to the monitored DNS record when its main IP address is detected as UP after being marked as DOWN. Options are:

Monitoring only: The DNS record will continue to be monitored but no changes will be made when it's detected as UP again. Monitoring continues even after the UP detection and notifications will be sent if the check goes UP and DOWN multiple times.

Do not monitor if UP: Once the DNS record is marked as DOWN and the down event handler is processed, monitoring stops. It must be manually reactivated for checks to continue.

Activate main IP if UP: The main IP address will be monitored even if it's marked as DOWN. If it's detected as UP again, it will be reactivated for the DNS record. This option works with both "Deactivate the DNS record" and "Replace with working IP" down events.

To activate the DNS Failover and Monitoring check for all supported DNS records, simply click the icon:  

Once activated, the first monitoring check will be carried out within a minute, and the results will be recorded.

If you need to deactivate the DNS Failover and Monitoring check, use the "Deactivate this DNS Failover and Monitoring check" button at any time.