The easiest thing to do is change your scrape interval to 1m. This would eliminate the problem. Don't worry, this is a perfectly normal interval for Prometheus.
On Wed, Jul 24, 2024 at 4:29 AM Tareerat Pansuntia < [email protected]> wrote: > And I would like to know if Prometheus can store metrics for only five > minutes or up to five minutes before they disappear. Why does the graph > show gaps at certain intervals?" > Thanks in advance, > > ในวันที่ พ. 24 ก.ค. 2024 เวลา 09:20 Tareerat Pansuntia < > [email protected]> เขียนว่า: > >> Thank you for help me >> but >> >> I am still a beginner in this field. Could you please help me understand >> how to use the avg_over_time(metric[$__rate_interval]) function? I will >> look for more information on my own afterward. For the metric in question, >> if I use probe_success, should I write it as >> avg_over_time(probe_duration_seconds{instance="xx", >> job="blackbox_exporter"}[30m])? >> Cloud you recommend me about function prometheus ? >> >> Thanks in advance, >> >> ในวันที่ พฤ. 18 ก.ค. 2024 เวลา 15:10 Ben Kochie <[email protected]> >> เขียนว่า: >> >>> This is Prometheus staleness handling. In order to allow queries to >>> function, Prometheus graph query evaluation (range query) is actually a >>> series of query evaluations over time. Each one is independent of the next. >>> >>> In order to support millisecond accurate timestamps, Prometheus will >>> look back up to 5 minutes for samples from each evaluation timestamp. >>> >>> There are a couple ways around this. >>> * Scrape faster. It's perfectly common and normal to scrape every >>> 15s-60s in Prometheus. It basically doesn't take any more storage space due >>> to the way Prometheus compresses samples. In fact, scraping slower wastes >>> data by creating incomplete blocks. >>> * Use a function like `avg_over_time(metric[$__rate_interval])` to >>> effectively increase the lookback. This needs to be combined with setting >>> the "min step" in the Grafana query options to 30min to match your scrape >>> interval. >>> >>> On Thu, Jul 18, 2024 at 8:48 AM Tareerat Pansuntia < >>> [email protected]> wrote: >>> >>>> hi, everyone >>>> >>>> I have configured Prometheus to scrape data every 30 minutes. This >>>> means Prometheus will scrape data every thirty minutes and then display it >>>> in Grafana. However, the graph in Grafana shows the data for only a brief >>>> moment and then disappears. After thirty minutes, the graph reappears >>>> again. What could be causing this issue, and how can I fix it? >>>> >>>> [image: Screenshot 2024-07-18 112523.png] >>>> >>>> In the graph , the green line appears every thirty minutes and then >>>> disappears until the next thirty-minute interval. I have set the >>>> scrape_interval for that website to every 30 minutes. What could be causing >>>> the data to disappear in between, and how can I fix this issue? >>>> >>>> Thanks in advance, >>>> >>>> -- >>>> 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/13d491cd-f8c0-42a3-bb3e-9f5a0fb94454n%40googlegroups.com >>>> <https://groups.google.com/d/msgid/prometheus-users/13d491cd-f8c0-42a3-bb3e-9f5a0fb94454n%40googlegroups.com?utm_medium=email&utm_source=footer> >>>> . >>>> >>> -- 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/CABbyFmp9t7E-aXXcoHwvt%2BhqizX1jYmHfTciU%3DtRswY2k1b9CQ%40mail.gmail.com.
