← Back to context Comment by otabdeveloper4 10 hours ago No thanks. In 2026 I want HA and replication out of the box without the insanity. 6 comments otabdeveloper4 Reply eikenberry 9 hours ago Came to say the same thing. Personally I'd only touch Postgres in a couple cases.1. Downtime doesn't matter. 2. Paying someone else (eg. AWS) to manage redundancy and fail-over.It just feels crazy to me that Postgres still doesn't have a native HA story since I last battled with this well over a decade ago. nrvn 9 hours ago Exactly my thoughts immediately after reading the word “just”. Also, PITR. groundzeros2015 8 hours ago You exceeded the step of maxing out the best server you can buy? throwaway7783 8 hours ago HA is not about exceeding the limits of a server. Its about still serving traffic when that best server I bought goes offline (or has failed memory chip, or a disk or... ). groundzeros2015 8 hours ago Replication? 1 reply →
eikenberry 9 hours ago Came to say the same thing. Personally I'd only touch Postgres in a couple cases.1. Downtime doesn't matter. 2. Paying someone else (eg. AWS) to manage redundancy and fail-over.It just feels crazy to me that Postgres still doesn't have a native HA story since I last battled with this well over a decade ago.
groundzeros2015 8 hours ago You exceeded the step of maxing out the best server you can buy? throwaway7783 8 hours ago HA is not about exceeding the limits of a server. Its about still serving traffic when that best server I bought goes offline (or has failed memory chip, or a disk or... ). groundzeros2015 8 hours ago Replication? 1 reply →
throwaway7783 8 hours ago HA is not about exceeding the limits of a server. Its about still serving traffic when that best server I bought goes offline (or has failed memory chip, or a disk or... ). groundzeros2015 8 hours ago Replication? 1 reply →
Came to say the same thing. Personally I'd only touch Postgres in a couple cases.
1. Downtime doesn't matter. 2. Paying someone else (eg. AWS) to manage redundancy and fail-over.
It just feels crazy to me that Postgres still doesn't have a native HA story since I last battled with this well over a decade ago.
Exactly my thoughts immediately after reading the word “just”. Also, PITR.
You exceeded the step of maxing out the best server you can buy?
HA is not about exceeding the limits of a server. Its about still serving traffic when that best server I bought goes offline (or has failed memory chip, or a disk or... ).
Replication?
1 reply →