Improve metrics logging for network io and nginx workers (#16)

Author: dhurley

src/core/metrics/sources/net_io.go

@@ -87,7 +87,7 @@ func (nio *NetIO) Collect(ctx context.Context, wg *sync.WaitGroup, m chan<- *pro
 	// collect net overflow. This is not easily obtained by gopsutil, so we exec netstat to get these values
 	overflows, err := nio.env.GetNetOverflow()
 	if err != nil {
-		log.Warnf("Error occurred getting network overflow metrics, %v", err)
+		log.Debugf("Error occurred getting network overflow metrics, %v", err)
 	}
 
 	if nio.netOverflows < 0 {

src/core/metrics/sources/nginx_worker.go

@@ -183,40 +183,40 @@ func (client *NginxWorkerClient) GetWorkerStats(childProcs []*proto.NginxDetails
 
 		pidAsInt, err := strconv.Atoi(nginxDetails.ProcessId)
 		if err != nil {
-			log.Warnf("failed to convert %s to int: %v", nginxDetails.ProcessId, err)
+			log.Debugf("failed to convert %s to int: %v", nginxDetails.ProcessId, err)
 			continue
 		}
 
 		proc, err := ps.NewProcess(int32(pidAsInt))
 		if err != nil {
-			log.Warnf("failed to retrieve process from pid %d: %v", pidAsInt, err)
+			log.Debugf("failed to retrieve process from pid %d: %v", pidAsInt, err)
 			continue
 		}
 
 		if times, err := proc.Times(); err == nil {
 			usr = usr + times.User
 			sys = sys + times.System
 		} else {
-			log.Warn("unable to get CPU times metrics")
+			log.Debug("unable to get CPU times metrics")
 		}
 
 		if memstat, err := proc.MemoryInfo(); err == nil {
 			memRss += float64(memstat.RSS)
 			memVms += float64(memstat.VMS)
 		} else {
-			log.Warn("unable to get memory info metrics")
+			log.Debug("unable to get memory info metrics")
 		}
 
 		if mempct, err := proc.MemoryPercent(); err == nil {
 			memPct += float64(mempct)
 		} else {
-			log.Warn("unable to get memory percentage metrics")
+			log.Debug("unable to get memory percentage metrics")
 		}
 
 		if fd, err := proc.NumFDs(); err == nil {
 			fdSum = fdSum + float64(fd)
 		} else {
-			log.Warn("unable to get number of file descriptors used metrics")
+			log.Debug("unable to get number of file descriptors used metrics")
 		}
 
 		if rlimit, err := proc.Rlimit(); err == nil {
@@ -228,7 +228,7 @@ func (client *NginxWorkerClient) GetWorkerStats(childProcs []*proto.NginxDetails
 			}
 			stats.Workers.RlimitNofile = float64(rlimitMax)
 		} else {
-			log.Warn("unable to get resource limit metrics")
+			log.Debug("unable to get resource limit metrics")
 		}
 
 		if ioc, err := proc.IOCounters(); err == nil {