Nginx入口超时/连接断开

时间:2018-08-24 16:57:17

标签: nginx kubernetes-ingress nginx-ingress

考虑此NodePort服务:

k describe service myservice
...
Type:                     NodePort
IP:                       10.15.248.153
Port:                     myservice  8080/TCP
TargetPort:               8080/TCP
NodePort:                 myservice  30593/TCP
Endpoints:                10.12.223.56:8080

考虑一个耗时120秒的请求:

# time curl -vkH 'http://myservce:8080/test?timeout=120'
*   Trying 10.15.248.153...
* TCP_NODELAY set
* Connected myservice (10.15.248.153) port 8080 (#0)
...
< HTTP/1.1 200

real    2m0.023s
user    0m0.009s
sys 0m0.009s

这很好。因此,我配置了nginx-inress超时:

nginx.ingress.kubernetes.io/proxy-send-timeout: "900"
nginx.ingress.kubernetes.io/proxy-read-timeout: "900"
nginx.ingress.kubernetes.io/proxy-connect-timeout: "60"

我可以确认它在nginx.conf中:

proxy_connect_timeout                   60s;
proxy_send_timeout                      900s;
proxy_read_timeout                      900s;

所以现在,我尝试输入nginx入口pod并尝试通过nginx-ingress访问myservice:

time curl -vkH 'http://127.0.0.1/myservice/test?timeout=120'

但是这次我得到的答复是空的-有时在35s之后,有时在90s之后,但是它将到达那里:

* Curl_http_done: called premature == 0
* Empty reply from server
* Connection #0 to host 127.0.0.1 left intact
curl: (52) Empty reply from server

我不知道可能会发生什么。好像nginx随机重启了,我的连接掉了。

1 个答案:

答案 0 :(得分:3)

由于某种原因,nginx进程会每30秒不断重新加载一次。每次重新加载时,所有连接均被删除。

解决方案是设置:

worker-shutdown-timeout: "900s"

在nginx元素config-map中。