问题原文
原帖链接 · user307927 · CC BY-SA 3.0
I'm building a prometheus stack that uses the remote_read/remote_write feature for long term storage, and I have some questions about if/how prometheus' tsdb needs to be backed up, given that it is using long term storage.
1) Do I need to back up the prometheus TSDB if data is being pushed into remote storage?
2) If a failure occurs and the prometheus instance needs to be rebuilt, would prometheus' tsdb need to be restored? Or will it attempt to query the long term storage for instance if it does not have metrics in the local tsdb for its retention period?
已采纳解答
原帖链接 · brian-brazil · CC BY-SA 3.0
1) Do I need to back up the prometheus TSDB if data is being pushed into remote storage?
You don't have to.
Or will it attempt to query the long term storage for instance if it does not have metrics in the local tsdb for its retention period?
This is how it works.
摘自公开归档,保留原文语言,版式经整理,外部媒体与邮箱地址已移除。
