To be fair at my current place we have a Prometheus instance for our Network team with a scrape interval of 5 minutes; and we have just adapted 2 things:
1. Loopback delta is set to 15 minute 2. We use irate, not rate. However this Prometheus instance is isolated and is only doing this. ----- Original Message ----- From: Stuart Clark <[email protected]> To: Arun Raj <[email protected]> Cc: Prometheus Users <[email protected]> Sent: Sun, 16 Feb 2020 16:13:06 +0100 (CET) Subject: Re: [prometheus-users] Prometheus exporter collection behaviour On 2020-02-16 15:06, Arun Raj wrote: > hi Clark, > > Our applications are telco based applications. So far, they have > been using push model for every 5 minutes. The every 5 minute is a > telecom standard guideline. We thought of consume those files every 5 > minutes from our custom exporter. We can't miss the second, Is there > any way in prometheus to accomplish this ? > The maximum scrape rate for Prometheus is about 2 minutes due to staleness. Can you switch to say scraping every minute? > Regards, > Arun Raj. R > > On Sunday, February 16, 2020 at 11:53:54 PM UTC+9, Stuart Clark wrote: > >> Both are expected. >> >> The scrape interval is an average, so gaps between scrapes might be >> a bit longer or shorter each time. Different scrapes with the same >> scrape interval will happen at different times to prevent everything >> happening at once, overwhelming the server and network. >> >> Ideally most metrics should be counters rather than gauges which are >> a lot more resistant to slight differences in timings between >> scrapes and across jobs & targets. >> >> On 16 February 2020 14:42:32 GMT, Arun Raj <[email protected]> >> wrote: >> >>> hi All, >>> >>> We are running prometheus server in our testing environment >>> which scraps 3 different exporters.( node exporter, kafka exporter >>> and custom exporter). >>> >>> 1. All the three exporters are pulling its metrics in different >>> timing even though the scrap interval are same. Is it expected >>> behaviour ? >>> 2. Our application's very time sensitive, we would like to >>> scrap for every 60 seconds. But we could see the collection time >>> difference are like 61 seconds or 62 seconds at sometimes. Is >>> there any way to make it consistent ? Or Is it expected behaviour >>> ? >>> >>> Regards, >>> Arun Raj. R >> >> -- >> 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/db38d816-c77d-46ff-be54-0cde43e27ac1%40googlegroups.com > [1]. > > > Links: > ------ > [1] > https://groups.google.com/d/msgid/prometheus-users/db38d816-c77d-46ff-be54-0cde43e27ac1%40googlegroups.com?utm_medium=email&utm_source=footer -- Stuart Clark -- 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/be85e01278a5f6a8ec15861bf0d833d3%40Jahingo.com. -- 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/22084893.7026.1581866423708.JavaMail.zimbra%40inuits.eu.
