Health check not working for our application

[detushar]

We have to install NGINX 1.16 recently with this module using the below command. In PRD, if any backend down. Nginx upstream still sending traffic to that dead instance.

patch -p1 <../nginx_upstream_check_module-master/check_1.14.0+.patch

Am i doing anything wrong here? What does this patch mean?

[solio]

Do you enable bellow directive in 'upstream'： check interval=3000 rise=2 fall=5 timeout=1000 type=http; check_http_send "HEAD / HTTP/1.0\r\n\r\n"; check_http_expect_alive http_2xx http_3xx;

[mojodevops]

We have to install NGINX 1.16 recently with this module using the below command. In PRD, if any backend down. Nginx upstream still sending traffic to that dead instance.

patch -p1 <../nginx_upstream_check_module-master/check_1.14.0+.patch

Am i doing anything wrong here? What does this patch mean?

See https://github.com/yaoweibin/nginx_upstream_check_module/issues/217, thinking that your nginx is 1.16, maybe you have used a wrong version of patch file?

[DonatasFe]

I have similar issue with random two

    random two;
    keepalive 300;
    server 192.168.0.1:80  max_fails=3 fail_timeout=15s weight=3;
    server 192.168.0.2:80  max_fails=3 fail_timeout=15s weight=3;
    server 192.168.0.3:80  max_fails=3 fail_timeout=15s weight=3;
    check_http_send "GET /available HTTP/1.0\r\nHost: test_internalhealthcheck\r\n\r\n";
    check_http_expect_alive http_2xx http_3xx;

if i set one upstream to respond with 503, nginx still sends requests to it. with least_conn it works better, still some of request are being send to the upstream which responds with http_5xx

Build info:

nginx version: nginx/1.16.1 built by gcc 4.8.5 20150623 (Red Hat 4.8.5-28) (GCC) built with OpenSSL 1.0.2k-fips 26 Jan 2017 patch -p1 < ../nginx_upstream_check_module/check_1.16.1+.patch