The following image shows Java Message Service (JMS) queue messages and client failed requests during a failure of all the web servers on one site. Time to live (TTL) equals 120 seconds, and downtime is approximately three minutes.

This image includes two time-series graphs that correlate Oracle HTTP Server (OHS) request failures with Java Message Service (JMS) queue growth.

Both graphs use the same x-axis, Time, and their trends align to show that the failure surge coincides with rising JMS queue counts during the same period.

The first graph is titled “Failed Requests / sec.” The y-axis label is “Trans Failed/sec.” The x-axis label is “Time.” The failure rate increases during the outage and then clears.

The second graph is titled “FOD JMS MessagesCurrentCount per server's queue.” The y-axis shows message counts from 0K to 6K. The x-axis label is “Time.” Multiple per‑server series rise during the outage and continue afterward.

The outage starts when the call-out states, "When the OHS servers in region 1 stop (the outage starts; all the requests sent to region 1 fail)." Failed requests per second increases during this period, and message counts increase.

A call out states, "When setting DynamicServerList to ON in the region 2 OHS servers (hence, the region 2 OHS servers send the requests to the WLS servers in the other region too)."

The outage ends when the call-out states, "When all the requests are redirected to region 2 LBR (the outage ends)."