neon icon indicating copy to clipboard operation
neon copied to clipboard

Neon: Serverless Postgres. We separated storage and compute to offer autoscaling, code-like database branching, and scale to zero.

Results 520 neon issues
Sort by recently updated
recently updated
newest added

## Problem Now we don't run Pylint as a part of a workflow. However, it can be useful in automatic error detection and style enforcement. Unfortunately, we cannot turn on...

Within the LayerMap, we store `Arc`. In all places, we iterate the existing layers by calling: https://github.com/neondatabase/neon/blob/27fe7f8963e5227d24cdd56aab419fa973dba369/pageserver/src/tenant/layer_map/historic_layer_coverage.rs#L501-L510 This results in incrementing the reference counters needlessly, because we rarely hold on...

t/bug
c/storage/pageserver
a/tech_debt

The basic node removal API in https://github.com/neondatabase/neon/pull/8226 has some caveats: ```[tasklist] ### Tasks - [ ] When we reschedule shards away from the failing node, we do it naively wrt...

t/feature
c/storage/controller

The PostgreSQL frontend-backend protocol supports protocol negotiation, so that if the client supports a higher version of the protocol than the server, they can agree to use the lowest common...

t/feature
c/cloud/proxy

The s3 client setup code in scrubber https://github.com/neondatabase/neon/blob/3a2f10712ad557c978f966579c9bfa89ad6f4bae/s3_scrubber/src/lib.rs#L279-L326 looks quite copy-pasted from `remote_storage` https://github.com/neondatabase/neon/blob/c18d3340b5e3c978a81c3db8b6f1e83cd9087e8a/libs/remote_storage/src/s3_bucket.rs#L84-L138 The differences appear to be: - scrubber explicitly handles `AWS_ENDPOINT_URL`, pageserver doesn't - best practice would...

c/storage/pageserver
a/tech_debt
c/storage/scrubber

Expose endpoint lag to console users

Set up a Grafana dashboard that show RO endpoints which are: * lagging behind the master more than X bytes / Y minutes * endpoints that encountered 'query canceled due...

c/compute

Postgres has a mechanism to write all currently-held AccessExclusiveLocks to the WAL, before each running-xacts record. A hot standby replica uses that to acquire the equivalent AccessExclusiveLocks in the standby...

t/bug
c/compute

The error means that manager exited earlier than `ResidenceGuard` and it's not unexpected with current deletion implementation. This commit changes log level to reduse noise.