Hey there, since this discussion seems rather persistent and there still seems to be a need for expiration of jobs in the pushgateway, I decided to also leave a note here to share our current solution to the problem which might be of use to others:
https://github.com/jorinvo/prometheus-pushgateway-cleaner This solution is only one possibility and it might be similar to other custom solutions, but it is open source. The README tries to explain the reasoning behind the project and tries to direct people to go with the option that is right for them. Feel free to leave feedback in Github. I hope this is helpful and contributes to slowing down these arguments going on for years now :) Regards, Jorin On Tuesday, November 12, 2019 at 2:10:02 PM UTC+1, Björn Rabenstein wrote: > > On 11.11.19 23:07, 'Chris Scott-Thomas' via Prometheus Developers wrote: > > When you have no option but to push metrics, than scrape, what is the > > better solution? > > It really depends on details. But if you have a whole complex system > that doesn't really fit the Prometheus approach, and you cannot really > change it to fit better, a possible answer is to not use Prometheus at > all. > > If a statsd approach works well for you, you might consider using > statsd instrumentation and then use the statsd_exporter to lift things > over into the Prometheus world. > > But really, no silver-bullet recommendation is possible here. > > > Perhaps this is just being missed? Pushgateway does kind of > > allude to this function in its name. > > Yes, in hindsight the naming was a mistake. > > -- > Björn Rabenstein > [PGP-ID] 0x851C3DA17D748D03 > [email] [email protected] <javascript:> > -- You received this message because you are subscribed to the Google Groups "Prometheus Developers" 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-developers/816838e7-7a08-42ea-b609-a56e48d0440b%40googlegroups.com.
