rust/neon
1
0
Fork 0
mirror of https://github.com/neondatabase/neon.git synced 2026-01-05 20:42:54 +00:00
Code Issues Packages Projects Releases Wiki Activity

Rename contrib/zenith to contrib/neon. Rename custom GUCs:

Browse Source 
- zenith.page_server_connstring -> neon.pageserver_connstring
- zenith.zenith_tenant -> neon.tenantid
- zenith.zenith_timeline -> neon.timelineid
- zenith.max_cluster_size -> neon.max_cluster_size
This commit is contained in:
Anastasia Lubennikova
2022-04-18 19:43:57 +03:00
committed by Anastasia Lubennikova
parent e3b320daab
commit 3accde613d
29 changed files with 72 additions and 72 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
docs/rfcs/cluster-size-limits.md
View File

@@ -22,8 +22,8 @@ so we don't want to give users access to the functionality that we don't think i
* pageserver - calculate the size consumed by a timeline and add it to the feedback message.
* safekeeper - pass feedback message from pageserver to compute.
* compute - receive feedback message, enforce size limit based on GUC `zenith.max_cluster_size`.
* console - set and update `zenith.max_cluster_size` setting
* compute - receive feedback message, enforce size limit based on GUC `neon.max_cluster_size`.
* console - set and update `neon.max_cluster_size` setting
## Proposed implementation
@@ -49,7 +49,7 @@ This message is received by the safekeeper and propagated to compute node as a p
Finally, when compute node receives the `current_timeline_size` from safekeeper (or from pageserver directly), it updates the global variable.
And then every zenith_extend() operation checks if limit is reached `(current_timeline_size > zenith.max_cluster_size)` and throws `ERRCODE_DISK_FULL` error if so.
And then every zenith_extend() operation checks if limit is reached `(current_timeline_size > neon.max_cluster_size)` and throws `ERRCODE_DISK_FULL` error if so.
(see Postgres error codes [https://www.postgresql.org/docs/devel/errcodes-appendix.html](https://www.postgresql.org/docs/devel/errcodes-appendix.html))
TODO:
@@ -75,5 +75,5 @@ We should warn users if the limit is soon to be reached.
### **Security implications**
We treat compute as an untrusted component. That's why we try to isolate it with secure container runtime or a VM.
Malicious users may change the `zenith.max_cluster_size`, so we need an extra size limit check.
Malicious users may change the `neon.max_cluster_size`, so we need an extra size limit check.
To cover this case, we also monitor the compute node size in the console.

6
docs/sourcetree.md
View File

@@ -42,13 +42,13 @@ Integration tests, written in Python using the `pytest` framework.
`/vendor/postgres`:
PostgreSQL source tree, with the modifications needed for Zenith.
PostgreSQL source tree, with the modifications needed for Neon.
`/vendor/postgres/contrib/zenith`:
`/vendor/postgres/contrib/neon`:
PostgreSQL extension that implements storage manager API and network communications with remote page server.
`/vendor/postgres/contrib/zenith_test_utils`:
`/vendor/postgres/contrib/neon_test_utils`:
PostgreSQL extension that contains functions needed for testing and debugging.