nginxinc/nginx-amplify-agent

Missing metrics when using Syslog for collection

AlexanderMatveev opened this issue · 2 comments

Switched to using syslog instead of log files.

These metrics no longer delivered to Amplify:

  • http.request.time
  • nginx.upstream.response.time

log_format is same as when I was using log files:

log_format  main_ext  '$remote_addr - $remote_user [$time_local] "$request" '
                      '$status $body_bytes_sent "$http_referer" '
                      '"$http_user_agent" "$http_x_forwarded_for" '
                      '"$host" sn="$server_name" '
                      'rt=$request_time '
                      'ua="$upstream_addr" us="$upstream_status" '
                      'ut="$upstream_response_time" ul="$upstream_response_length" '
                      'cs=$upstream_cache_status' ;

# this works fine:
#access_log /var/log/nginx/access.log main_ext;

# this variant hides two metrics I mentioned (all other metrics are fine):
access_log syslog:server=127.0.0.1:12000,tag=amplify,severity=info main_ext;

Screenshot_2021-01-31 NGINX Monitoring Made Easy(5)
Screenshot_2021-01-31 NGINX Monitoring Made Easy(3)
Screenshot_2021-01-31 NGINX Monitoring Made Easy(2)

Looks like some other metrics no longer collecting or showing incorrect zeros.

Screenshot_2021-01-31 NGINX Monitoring Made Easy(4)
Screenshot_2021-01-31 NGINX Monitoring Made Easy(1)
Screenshot_2021-01-31 NGINX Monitoring Made Easy

were you able to fix this? Im having the same problem