For how long are your containers running? If your containers are
short-lived (like a minute or less), Prometheus will not be able to
monitor them reliably. In that case, you might want to look into the
Pushgateway
https://github.com/prometheus/pushgateway . Your jobs
needed to inquire the average memory usage from the container and push
them to the Pushgateway before they exit.
If your containers are longer running (multiple minutes or longer), a
carefully crafted `average_over_time` expression with a long range
selector should do the trick.
--
Björn Rabenstein, Engineer
http://soundcloud.com/brabenstein
SoundCloud Ltd. | Rheinsberger Str. 76/77, 10115 Berlin, Germany
Managing Director: Alexander Ljung | Incorporated in England & Wales
with Company No. 6343600 | Local Branch Office | AG Charlottenburg |
HRB 110657B