From: Christopher Faulet <cfaulet@haproxy.com>
Date: Thu, 27 Feb 2020 15:12:07 +0000 (+0100)
Subject: MINOR: contrib/prometheus-exporter: Add the last heathcheck duration metric
X-Git-Tag: v2.2-dev4~108
X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=2711e51016cfc408deb03e66f46bea0fa5bf954c;p=thirdparty%2Fhaproxy.git

MINOR: contrib/prometheus-exporter: Add the last heathcheck duration metric

ST_F_CHECK_DURATION is now part of exported server metrics, named
haproxy_server_check_duration_seconds and expressed in seconds. For a given
server, this value is exported only if the healthcheck is finished (the status
is greater or equal to HCHK_STATUS_CHECKED).

This patch fixes the issue #519. It may be backported as fat as 2.0.
---

diff --git a/contrib/prometheus-exporter/README b/contrib/prometheus-exporter/README
index a0df4b21de..a63102028a 100644
--- a/contrib/prometheus-exporter/README
+++ b/contrib/prometheus-exporter/README
@@ -272,6 +272,7 @@ Exported metrics
 | haproxy_server_weight                              | Service weight.                                                           |
 | haproxy_server_check_status                        | Status of last health check, if enabled. (see below for the mapping)      |
 | haproxy_server_check_code                          | layer5-7 code, if available of the last health check.                     |
+| haproxy_server_check_duration_seconds              | Total duration of the latest server health check, in seconds.             |
 | haproxy_server_check_failures_total                | Total number of failed check (Only when the server is up).                |
 | haproxy_server_check_up_down_total                 | Total number of UP->DOWN transitions.                                     |
 | haproxy_server_downtime_seconds_total              | Total downtime (in seconds) for the service.                              |
diff --git a/contrib/prometheus-exporter/service-prometheus.c b/contrib/prometheus-exporter/service-prometheus.c
index fa80147865..6e7eca04bd 100644
--- a/contrib/prometheus-exporter/service-prometheus.c
+++ b/contrib/prometheus-exporter/service-prometheus.c
@@ -388,8 +388,8 @@ const int promex_srv_metrics[ST_F_TOTAL_FIELDS] = {
 	[ST_F_RATE_LIM]       = 0,
 	[ST_F_RATE_MAX]       = ST_F_LASTSESS,
 	[ST_F_CHECK_STATUS]   = ST_F_CHECK_CODE,
-	[ST_F_CHECK_CODE]     = ST_F_CHKFAIL,
-	[ST_F_CHECK_DURATION] = 0,
+	[ST_F_CHECK_CODE]     = ST_F_CHECK_DURATION,
+	[ST_F_CHECK_DURATION] = ST_F_CHKFAIL,
 	[ST_F_HRSP_1XX]       = ST_F_HRSP_2XX,
 	[ST_F_HRSP_2XX]       = ST_F_HRSP_3XX,
 	[ST_F_HRSP_3XX]       = ST_F_HRSP_4XX,
@@ -552,7 +552,7 @@ const struct ist promex_st_metric_names[ST_F_TOTAL_FIELDS] = {
 	[ST_F_RATE_MAX]       = IST("max_session_rate"),
 	[ST_F_CHECK_STATUS]   = IST("check_status"),
 	[ST_F_CHECK_CODE]     = IST("check_code"),
-	[ST_F_CHECK_DURATION] = IST("check_duration_milliseconds"),
+	[ST_F_CHECK_DURATION] = IST("check_duration_seconds"),
 	[ST_F_HRSP_1XX]       = IST("http_responses_total"),
 	[ST_F_HRSP_2XX]       = IST("http_responses_total"),
 	[ST_F_HRSP_3XX]       = IST("http_responses_total"),
@@ -715,7 +715,7 @@ const struct ist promex_st_metric_desc[ST_F_TOTAL_FIELDS] = {
 	[ST_F_RATE_MAX]       = IST("Maximum observed number of sessions per second."),
 	[ST_F_CHECK_STATUS]   = IST("Status of last health check (HCHK_STATUS_* values)."),
 	[ST_F_CHECK_CODE]     = IST("layer5-7 code, if available of the last health check."),
-	[ST_F_CHECK_DURATION] = IST("Time in ms took to finish last health check."),
+	[ST_F_CHECK_DURATION] = IST("Total duration of the latest server health check, in seconds."),
 	[ST_F_HRSP_1XX]       = IST("Total number of HTTP responses."),
 	[ST_F_HRSP_2XX]       = IST("Total number of HTTP responses."),
 	[ST_F_HRSP_3XX]       = IST("Total number of HTTP responses."),
@@ -2037,6 +2037,12 @@ static int promex_dump_srv_metrics(struct appctx *appctx, struct htx *htx)
 							goto next_sv;
 						metric = mkf_u32(FN_OUTPUT, (sv->check.status < HCHK_STATUS_L57DATA) ? 0 : sv->check.code);
 						break;
+					case ST_F_CHECK_DURATION:
+						if (sv->check.status < HCHK_STATUS_CHECKED)
+						    goto next_sv;
+						secs = (double)sv->check.duration / 1000.0;
+						metric = mkf_flt(FN_DURATION, secs);
+						break;
 					case ST_F_CHKFAIL:
 						metric = mkf_u64(FN_COUNTER, sv->counters.failed_checks);
 						break;