[prometheus-users] Re: not able to see metrics from query browser even though end point is up and showing metrics through curl
Then you're probably not scraping it, or the custom exporter is returning invalid data. Check the value of the "up" metric in query browser for the given job and instance, i.e. up{job="foo",instance="bar"} "up" missing - scraping not configured. "up" has value 0 - unable to communicate or data is bad. "up" has value 1 - maybe partial scrape Check logs at prometheus and exporter; check traffic between prometheus and exporter with tcpdump. Also show example of the curl output. -- 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 prometheus-users+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/prometheus-users/ab0b5481-a040-45db-b2cd-e9ed70c37f72o%40googlegroups.com.
Re: [prometheus-users] Re: not able to see metrics from query browser even though end point is up and showing metrics through curl
Hi Brian Candler wrote: "up" missing - scraping not configured. "up" has value 0 - unable to communicate or data is bad. "up" has value 1 - maybe partial scrape Check logs at prometheus and exporter; check traffic between prometheus and exporter with tcpdump. Can you show a sample tcpdump syntax for this purpose? Thanks. -- 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 prometheus-users+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/prometheus-users/130d9158-d6d7-f49f-1884-b2dd1f66e314%40pobox.com.
Re: [prometheus-users] Re: not able to see metrics from query browser even though end point is up and showing metrics through curl
tcpdump -i eth0 -nn -s0 -A host x.x.x.x and tcp port y where x.x.x.x is the IP address of the remote host, and y is the port you're running the exporter on. Replace "eth0" with your network interface name. If prometheus and the exporter are on the same machine then use "lo" (loopback) as the interface name. -- 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 prometheus-users+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/prometheus-users/a3819232-4410-46fb-b78a-b3cf72d01cebo%40googlegroups.com.
Re: [prometheus-users] Re: not able to see metrics from query browser even though end point is up and showing metrics through curl
up metric showing value 1 and alos targets it is showing up as well. is there nay time mismatch between prometheus server and exporter? if yes,, how could we disable it. On Sunday, June 28, 2020 at 6:44:57 AM UTC-5 b.ca...@pobox.com wrote: > tcpdump -i eth0 -nn -s0 -A host x.x.x.x and tcp port y > > where x.x.x.x is the IP address of the remote host, and y is the port > you're running the exporter on. Replace "eth0" with your network interface > name. If prometheus and the exporter are on the same machine then use "lo" > (loopback) as the interface name. > -- 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 prometheus-users+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/prometheus-users/be6a5035-d446-42ca-a44a-2e42a15dba5cn%40googlegroups.com.
Re: [prometheus-users] Re: not able to see metrics from query browser even though end point is up and showing metrics through curl
I don't really understand what you're trying to say. Does your exporter include timestamps for each metric? These are optional. It's almost always best not to include them unless you are *really* sure you know what you're doing. Notice that none of the standard exporters, like node_exporter, include timestamps. In that case, clock sync between prometheus server and exporter is irrelevant. If you do include timestamps, and the values are too far in the past, or out of sequence, then prometheus will discard the metrics and log an error. You might need to run prometheus with --log.level=debug to see them. If you're trying to export the time when some event happened - e.g. the time when a job was last run - then *don't* use the prometheus metric timestamp for this. Instead, expose a separate metric which is the timestamp as a Unix time. For an example of this in action, see how the node_exporter textfile collector returns the timestamp of the textfile it returned. -- 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 prometheus-users+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/prometheus-users/1f5f6727-5dd5-4e41-bdf7-3074216af84co%40googlegroups.com.