ELB

Monitoring¶

ActiveConnectionCount
HealthyHostCount, UnHealthyHostCount, HostCount
BackendConnectionErrors
- The count of the number of connections that were NOT successfully established between the Load Balancer and the registered instance
- Average stat is most useful
- Will report errors for all AZ's
- Could indicate an issue with the web server
HTTPCode_Backend_2XX,3XX,4XX,5XX
- Generated by registered instances
HTTPCode_ELB_4XX,5XX
- Generated by Load Balancer
ActiveFlowCount (Network Load Balancer)
Metrics for performance
- Latency
  - Measures the time elapsed (in seconds) after the request leaves the Load Balancer until the response is received
  - Average stat is most useful
  - Will report latency for all AZs
  - Page Load Time
- RequestCount
  - Number of requests completed/connections made during specified interval (1 or 5 mins)
- SurgeQueueLenght (Classic only)
  - A count of the total number of requests that are pending submissions to a registered instance
  - Max queue size is 1024
  - Additional requests will be rejected
  - Scaling up instances to ensure that never increases beyond the maximum queue capacity
- SpilloverCount (Classic only)
  - Number of requests rejected because the surge queue is full
  - Direct consequence of a surge queue length increase
  - Should monitor SpilloverCount and SurgeQueueLenght
- High numbers in these metrics can indicate a performance issue, need to scale infrastructure, etc