Is that ID something to do with the event or related to something else? Label should be very carefully used to ensure cardinality is controlled. Things like user IDs, IP addresses or other unique references should not be used.
If that is a unique ID that label should be removed. If you are needing information at the per event level you would want an events sys3(such as Elasticsearch or Loki) rather than Prometheus (which is for metrics). On 18 October 2020 16:27:56 BST, "[email protected]" <[email protected]> wrote: >Hi, >We have a program which exports metric whenever an event happens in the > >system, so we are generating a metric like this: >custom_metric{id="39a"} 1 >and once another event happens in the system, a new metric generates >like >this: >custom_metric{id="15c"} 1 >The problem is, since this metric is generated once in a while, the >metrics >get stalled after 5 minutes and e.g. there will be a gap between id=39a >and >id=15c when I want to see the history of "custom_metric", like the >picture: > >Is it possible to tell prometheus to keep the last metric until new one > >scraped by prometheus so the graph would be connected > >-- >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/fe4d0c65-7368-4162-a929-0acb7a44304dn%40googlegroups.com. -- Sent from my Android device with K-9 Mail. Please excuse my brevity. -- 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/58F3D9D4-D990-4DC9-A944-228546994A4B%40Jahingo.com.
