Hello,
Let's say I am bumping all my prometheus instances to at least 2 replicas to achieve HA. I add Thanos, I have a global view, metrics are deduplicated, all is fine. One day, I have a huge spike because of high cardinality metrics, and this will kill my prometheus instance. But wouldn't this kill all of my replicas as I am scraping the very same targets with each of those which is the best practice, right? So I guess, all of the replicas will be killed by oomkill, and not event Thanos can really help me in this case. Or am I missing something? Thanks in advance! Another quick question: let's say I am using Istio. Would it make sense to add Thanos to an istio-injected namespace with envoy-proxies or the overhead/complexity is not necessary and there's no real advantage to do so? -- 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/1aa4463e-51e9-497e-935f-4e7c2ede11e7n%40googlegroups.com.
