Page MenuHomePhabricator
Create Task
Maniphest T260053

Reduce Prometheus retention/coverage once Thanos has collected enough data
Closed, ResolvedPublic
Actions

Description

On Sept 21st 2020 there will be full overlap between Thanos data and Prometheus data. After that we should progressively reduce Prometheus' retention and increase Thanos coverage (i.e. the time span of queries that are served by Thanos)

Also note that ATM Prometheus compactions are disabled to be able to upload data to Thanos, meaning that queries to Prometheus that contain both a lot of metrics and long time ranges will be slower than usual.

Related Objects

Event Timeline

fgiunchedi created this task.Aug 10 2020, 1:18 PM
Restricted Application added a subscriber: Aklapper. · View Herald TranscriptAug 10 2020, 1:18 PM
lmata moved this task from Inbox to Backlog on the observability board.Aug 10 2020, 3:35 PM
fgiunchedi mentioned this in T260241: Grafana/Thanos serves 503s for long-time-window requests.Aug 12 2020, 2:22 PM
fgiunchedi closed this task as Resolved.Aug 26 2020, 8:52 AM
fgiunchedi claimed this task.

This has happened (earlier than expected) as part of T260241, we're now serving 0-15 days from Prometheus and the rest from Thanos.

Content licensed under Creative Commons Attribution-ShareAlike (CC BY-SA) 4.0 unless otherwise noted; code licensed under GNU General Public License (GPL) 2.0 or later and other open source licenses. By using this site, you agree to the Terms of Use, Privacy Policy, and Code of Conduct. · Wikimedia Foundation · Privacy Policy · Code of Conduct · Terms of Use · Disclaimer · CC-BY-SA · GPL