The scrape interval of 10m is too long. 2m is the maximum recommended; this is because Prometheus considers timeseries which have not been updated for 5m to be stale. However, that would make the timeseries vanish, not report a status of 0.
I suggest you take Prometheus out of the loop entirely, and just test blackbox_exporter by itself: curl -g 'http://127.0.0.1:9115/probe?module=icmp_ipv4&target=10.23.45.223&debug=true' Does it return probe_success 0 or 1? If 0, then look at tcpdump in another window: tcpdump -i <devname> -nn -s0 -X host 10.23.45.223 This is a long shot, but are you running blackbox_exporter or the target host in a KVM virtual machine by any chance? There are some cases where the virtual NIC doesn't do UDP or ICMP checksums properly, I think when connected to a virtual bridge. Also, check these issues: https://github.com/prometheus/blackbox_exporter/issues/360 https://github.com/prometheus/blackbox_exporter/issues/411 https://github.com/prometheus/blackbox_exporter/issues/705 BTW, it's perfectly OK to replace the "instance" label in cases like this, e.g. relabel_configs: - source_labels: [__address__] target_label: __param_target * - source_labels: [__address__] target_label: instance* - target_label: __address__ replacement: localhost:9115 If you don't set the instance label then it gets set to __address__, which is always localhost:9115 (hence not very interesting) On Wednesday, 9 March 2022 at 08:37:42 UTC [email protected] wrote: > Hi, > > Currently, I need to monitor a hundred hosts with the ICMP module using > blackbox_exporter with file_sd_configs for every target host. > > But the result I've got is almost the host result the '0' / 'down' value > whereas the host is up and running and can be ping to that. Is there any > misconfiguration on my own? > > Here is my config example. > > > *prometheus config prometheus.yml:* > scrape_configs: > - job_name: "rtu_status" > scrape_interval: 10m > metrics_path: /probe > params: > module: [icmp_ipv4] > file_sd_configs: > - files: > - 'ping/*.json' > relabel_configs: > - source_labels: [__address__] > target_label: __param_target > - source_labels: [__param_target] > target_label: ip_rtu > - target_label: __address__ > replacement: localhost:9115 > > > > *target config hosts.json:*[ > {"targets":["10.23.45.223"]," > labels": > {"rtu_sname":"RTU00-D1-ARK"}} > ] > > > *blackbox config blackbox.yml*modules: > icmp_ipv4: > prober: icmp > # timeout: 1s > icmp: > preferred_ip_protocol: ip4 > > > *blackbox daemon config:*[Unit] > Description=blackbox_exporter > Wants=network-online.target > After=network-online.target > > [Service] > User=blackbox > AmbientCapabilities=CAP_NET_RAW > ExecStart=/opt/blackbox-exporter/blackbox_exporter > --config.file=/opt/blackbox-exporter/blackbox.yml > StandardOutput=syslog > StandardError=syslog > LimitNOFILE=8192 > -- You received this message because you are subscribed to the Google Groups "Prometheus Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/prometheus-users/80c7910a-df0e-4fa6-8fa3-fa863cd4a1a0n%40googlegroups.com.
