Hi,
I’m using a container with the official Fastly Prometheus Exporter image to retrieve metrics every minutes from rt.fastly.com.
It has been working flawlessly for weeks now until last tuesday. Now every hour I have an interval of 5-10 minutes where no metrics is retrieved at all.
On Grafana it shows like this:
First it was twice an hour, and it lasted until 13:00 UTC, before disappearing.
It reappeared the next day around 12:00 and now is extremely stable: there’s a window of mising data each hour between hh:00 and hh:17.
Of course it’s not a bandwidth or target server issue. The traffic continues as usual but the RT API “hides” it.
Any recent change that could explain this issue? Any help?


