Hi! I'd say the experimental flag can be removed now. I already bumped the version of it to 1.0 on master. I don't know anyone who has it in heavy use though, but there was some feedback either on this list, or in the issue tracker. The short term plans for a 1.0 release are to change the postgresql requirement to 9.6 as minimum and use the "UPSERT" functionality instead of the current stored procedure. That should improve the reliability a bit further, but with the usage we have it shouldn't be a problem. It will mostly just make the code cleaner. Long term plans are an actual relational data model, but the storage backend in devpi-server would have to change for that. A general issue even with sqlite/file storage we recently identified are duplicated files if you push releases to other indexes. This is most likely easier to solve in the postgresql backend than in the sqlite/file storage backend. Regards, Florian Schulze On 19 Mar 2019, at 16:29, Kevin Patterson wrote:
I was curious if anyone was using the devpi-postgresql package in a live/heavy use environment. It still carries the experimental warning, so figured I'd ask here before diving too deep :-)
We were working with an internal cloud deployment suite and it seemed that the sqlite with smb mounts was not very reliable....so we were considering a switch to this postgres module if possible Or any other ideas for cloud deployments where the database needs to be remote from the app?
Thanks! Kevin _______________________________________________ devpi-dev mailing list -- devpi-dev@python.org To unsubscribe send an email to devpi-dev-leave@python.org https://mail.python.org/mailman3/lists/devpi-dev.python.org/