Compare commits

..

315 Commits

Author SHA1 Message Date
Folke Behrens
8a6ee79f6f Merge pull request #9921 from neondatabase/rc/release-proxy/2024-11-28
Proxy release 2024-11-28
2024-11-28 11:09:06 +01:00
github-actions[bot]
9052c32b46 Proxy release 2024-11-28 2024-11-28 06:02:15 +00:00
Ivan Efremov
995e729ebe Merge pull request #9832 from neondatabase/rc/release-proxy/2024-11-21
Proxy release 2024-11-21
2024-11-21 09:41:31 +02:00
github-actions[bot]
76077e1ddf Proxy release 2024-11-21 2024-11-21 06:02:11 +00:00
Ivan Efremov
0467d88f06 Merge pull request #9756 from neondatabase/rc/proxy/2024-11-14
Proxy release 2024-11-14
2024-11-14 09:46:52 +02:00
Ivan Efremov
f5eec194e7 Merge pull request #9674 from neondatabase/rc/proxy/2024-11-07
Proxy release 2024-11-07
2024-11-07 12:07:12 +02:00
Conrad Ludgate
7e00be391d Merge pull request #9558 from neondatabase/rc/proxy/2024-10-29
Auth broker release 2024-10-29
2024-10-29 12:10:50 +00:00
Folke Behrens
d56599df2a Merge pull request #9499 from neondatabase/rc/proxy/2024-10-24
Proxy release 2024-10-24
2024-10-24 10:34:56 +02:00
Folke Behrens
9d9aab3680 Merge pull request #9426 from neondatabase/rc/proxy/2024-10-17
Proxy release 2024-10-17
2024-10-17 12:18:51 +02:00
Folke Behrens
a202b1b5cc Merge pull request #9341 from neondatabase/rc/proxy/2024-10-10
Proxy release 2024-10-10
2024-10-10 09:17:11 +02:00
Folke Behrens
90f731f3b1 Merge pull request #9256 from neondatabase/rc/proxy/2024-10-03
Proxy release 2024-10-03
2024-10-03 11:01:41 +02:00
Conrad Ludgate
7736b748d3 Merge pull request #9159 from neondatabase/rc/proxy/2024-09-26
Proxy release 2024-09-26
2024-09-26 09:22:33 +01:00
Conrad Ludgate
9c23333cb3 Merge pull request #9056 from neondatabase/rc/proxy/2024-09-19
Proxy release 2024-09-19
2024-09-19 10:41:17 +01:00
Conrad Ludgate
66a99009ba Merge pull request #8799 from neondatabase/rc/proxy/2024-08-22
Proxy release 2024-08-22
2024-08-22 10:04:56 +01:00
Conrad Ludgate
5d4c57491f Merge pull request #8723 from neondatabase/rc/proxy/2024-08-14
Proxy release 2024-08-14
2024-08-14 13:05:51 +01:00
Conrad Ludgate
73935ea3a2 Merge pull request #8647 from neondatabase/rc/proxy/2024-08-08
Proxy release 2024-08-08
2024-08-08 15:37:09 +01:00
Conrad Ludgate
32e595d4dd Merge branch 'release-proxy' into rc/proxy/2024-08-08 2024-08-08 13:53:33 +01:00
Conrad Ludgate
b0d69acb07 Merge pull request #8505 from neondatabase/rc/proxy/2024-07-25
Proxy release 2024-07-25
2024-07-25 11:07:19 +01:00
Conrad Ludgate
98355a419a Merge pull request #8351 from neondatabase/rc/proxy/2024-07-11
Proxy release 2024-07-11
2024-07-11 10:40:17 +01:00
Conrad Ludgate
cfb03d6cf0 Merge pull request #8178 from neondatabase/rc/proxy/2024-06-27
Proxy release 2024-06-27
2024-06-27 11:35:30 +01:00
Conrad Ludgate
d81ef3f962 Revert "proxy: update tokio-postgres to allow arbitrary config params (#8076)"
This reverts commit 78d9059fc7.
2024-06-27 09:46:58 +01:00
Conrad Ludgate
5d62c67e75 Merge pull request #8117 from neondatabase/rc/proxy/2024-06-20
Proxy release 2024-06-20
2024-06-20 11:42:35 +01:00
Anna Khanova
53d53d5b1e Merge pull request #7980 from neondatabase/rc/proxy/2024-06-06
Proxy release 2024-06-06
2024-06-06 13:14:40 +02:00
Anna Khanova
29fe6ea47a Merge pull request #7909 from neondatabase/rc/proxy/2024-05-30
Proxy release 2024-05-30
2024-05-30 14:59:41 +02:00
Alexander Bayandin
640327ccb3 Merge pull request #7880 from neondatabase/rc/proxy/2024-05-24
Proxy release 2024-05-24
2024-05-24 18:00:18 +01:00
Anna Khanova
7cf0f6b37e Merge pull request #7853 from neondatabase/rc/proxy/2024-05-23
Proxy release 2024-05-23
2024-05-23 12:09:13 +02:00
Anna Khanova
03c2c569be [proxy] Do not fail after parquet upload error (#7858)
## Problem

If the parquet upload was unsuccessful, it will panic.

## Summary of changes

Write error in logs instead.
2024-05-23 11:44:47 +02:00
Conrad Ludgate
eff6d4538a Merge pull request #7654 from neondatabase/rc/proxy/2024-05-08
Proxy release 2024-05-08
2024-05-08 11:56:20 +01:00
Conrad Ludgate
5ef7782e9c Merge pull request #7649 from neondatabase/rc/proxy/2024-05-08
Proxy release 2024-05-08
2024-05-08 06:54:03 +01:00
Conrad Ludgate
73101db8c4 Merge branch 'release-proxy' into rc/proxy/2024-05-08 2024-05-08 06:43:57 +01:00
Anna Khanova
bccdfc6d39 Merge pull request #7580 from neondatabase/rc/proxy/2024-05-02
Proxy release 2024-05-02
2024-05-02 12:00:01 +02:00
Anna Khanova
99595813bb proxy: keep track on the number of events from redis by type. (#7582)
## Problem

It's unclear what is the distribution of messages, proxy is consuming
from redis.

## Summary of changes

Add counter.
2024-05-02 11:56:19 +02:00
Anna Khanova
fe07b54758 Merge pull request #7507 from neondatabase/rc/proxy/2024-04-25
Proxy release 2024-04-25
2024-04-25 13:50:05 +02:00
Anna Khanova
a42d173e7b proxy: Fix cancellations (#7510)
## Problem

Cancellations were published to the channel, that was never read.

## Summary of changes

Fallback to global redis publishing.
2024-04-25 13:42:25 +02:00
Anna Khanova
e07f689238 Update connect to compute and wake compute retry configs (#7509)
## Problem

## Summary of changes

Decrease waiting time
2024-04-25 13:20:21 +02:00
Conrad Ludgate
7831eddc88 Merge pull request #7417 from neondatabase/rc/proxy/2024-04-18
Proxy release 2024-04-18
2024-04-18 12:03:07 +01:00
Conrad Ludgate
943b1bc80c Merge pull request #7366 from neondatabase/proxy-hotfix
Release proxy (2024-04-11 hotfix)
2024-04-12 10:15:14 +01:00
Conrad Ludgate
95a184e9b7 proxy: fix overloaded db connection closure (#7364)
## Problem

possible for the database connections to not close in time.

## Summary of changes

force the closing of connections if the client has hung up
2024-04-11 23:38:47 +01:00
Conrad Ludgate
3fa17e9d17 Merge pull request #7357 from neondatabase/rc/proxy/2024-04-11
Proxy release 2024-04-11
2024-04-11 11:49:45 +01:00
Anna Khanova
55e0fd9789 Merge pull request #7304 from neondatabase/rc/proxy/2024-04-04
Proxy release 2024-04-04
2024-04-04 12:40:11 +02:00
Anna Khanova
2a88889f44 Merge pull request #7254 from neondatabase/rc/proxy/2024-03-27
Proxy release 2024-03-27
2024-03-27 11:44:09 +01:00
Conrad Ludgate
5bad8126dc Merge pull request #7173 from neondatabase/rc/proxy/2024-03-19
Proxy release 2024-03-19
2024-03-19 12:11:42 +00:00
Anna Khanova
27bc242085 Merge pull request #7119 from neondatabase/rc/proxy/2024-03-14
Proxy release 2024-03-14
2024-03-14 14:57:05 +05:00
Anna Khanova
192b49cc6d Merge branch 'release-proxy' into rc/proxy/2024-03-14 2024-03-14 14:16:36 +05:00
Conrad Ludgate
e1b60f3693 Merge pull request #7041 from neondatabase/rc/proxy/2024-03-07
Proxy release 2024-03-07
2024-03-08 08:19:16 +00:00
Anna Khanova
2804f5323b Merge pull request #6997 from neondatabase/rc/proxy/2024-03-04
Proxy release 2024-03-04
2024-03-04 17:36:11 +04:00
Anna Khanova
676adc6b32 Merge branch 'release-proxy' into rc/proxy/2024-03-04 2024-03-04 16:41:46 +04:00
Nikita Kalyanov
96a4e8de66 Add /terminate API (#6745) (#6853)
this is to speed up suspends, see
https://github.com/neondatabase/cloud/issues/10284


Cherry-pick to release branch to build new compute images
2024-02-22 11:51:19 +02:00
Arseny Sher
01180666b0 Merge pull request #6803 from neondatabase/releases/2024-02-19
Release 2024-02-19
2024-02-19 16:38:35 +04:00
Conrad Ludgate
6c94269c32 Merge pull request #6758 from neondatabase/release-proxy-2024-02-14
2024-02-14 Proxy Release
2024-02-15 09:45:08 +00:00
Anna Khanova
edc691647d Proxy: remove fail fast logic to connect to compute (#6759)
## Problem

Flaky tests

## Summary of changes

Remove failfast logic
2024-02-15 07:42:12 +00:00
Conrad Ludgate
855d7b4781 hold cancel session (#6750)
## Problem

In a recent refactor, we accidentally dropped the cancel session early

## Summary of changes

Hold the cancel session during proxy passthrough
2024-02-14 14:57:22 +00:00
Anna Khanova
c49c9707ce Proxy: send cancel notifications to all instances (#6719)
## Problem

If cancel request ends up on the wrong proxy instance, it doesn't take
an effect.

## Summary of changes

Send redis notifications to all proxy pods about the cancel request.

Related issue: https://github.com/neondatabase/neon/issues/5839,
https://github.com/neondatabase/cloud/issues/10262
2024-02-14 14:57:22 +00:00
Anna Khanova
2227540a0d Proxy refactor auth+connect (#6708)
## Problem

Not really a problem, just refactoring.

## Summary of changes

Separate authenticate from wake compute.

Do not call wake compute second time if we managed to connect to
postgres or if we got it not from cache.
2024-02-14 14:57:22 +00:00
Conrad Ludgate
f1347f2417 proxy: add more http logging (#6726)
## Problem

hard to see where time is taken during HTTP flow.

## Summary of changes

add a lot more for query state. add a conn_id field to the sql-over-http
span
2024-02-14 14:57:22 +00:00
Conrad Ludgate
30b295b017 proxy: some more parquet data (#6711)
## Summary of changes

add auth_method and database to the parquet logs
2024-02-14 14:57:22 +00:00
Anna Khanova
1cef395266 Proxy: copy bidirectional fork (#6720)
## Problem

`tokio::io::copy_bidirectional` doesn't close the connection once one of
the sides closes it. It's not really suitable for the postgres protocol.

## Summary of changes

Fork `copy_bidirectional` and initiate a shutdown for both connections.

---------

Co-authored-by: Conrad Ludgate <conradludgate@gmail.com>
2024-02-14 14:57:22 +00:00
John Spray
78d160f76d Merge pull request #6721 from neondatabase/releases/2024-02-12
Release 2024-02-12
2024-02-12 09:35:30 +00:00
Vlad Lazar
b9238059d6 Merge pull request #6617 from neondatabase/releases/2024-02-05
Release 2024-02-05
2024-02-05 12:50:38 +00:00
Arpad Müller
d0cb4b88c8 Don't preserve temp files on creation errors of delta layers (#6612)
There is currently no cleanup done after a delta layer creation error,
so delta layers can accumulate. The problem gets worse as the operation
gets retried and delta layers accumulate on the disk. Therefore, delete
them from disk (if something has been written to disk).
2024-02-05 09:58:18 +00:00
John Spray
1ec3e39d4e Merge pull request #6504 from neondatabase/releases/2024-01-29
Release 2024-01-29
2024-01-29 10:05:01 +00:00
John Spray
a1a74eef2c Merge pull request #6420 from neondatabase/releases/2024-01-22
Release 2024-01-22
2024-01-22 17:24:11 +00:00
John Spray
90e689adda pageserver: mark tenant broken when cancelling attach (#6430)
## Problem

When a tenant is in Attaching state, and waiting for the
`concurrent_tenant_warmup` semaphore, it also listens for the tenant
cancellation token. When that token fires, Tenant::attach drops out.
Meanwhile, Tenant::set_stopping waits forever for the tenant to exit
Attaching state.

Fixes: https://github.com/neondatabase/neon/issues/6423

## Summary of changes

- In the absence of a valid state for the tenant, it is set to Broken in
this path. A more elegant solution will require more refactoring, beyond
this minimal fix.

(cherry picked from commit 93572a3e99)
2024-01-22 16:20:57 +00:00
Christian Schwarz
f0b2d4b053 fixup(#6037): actually fix the issue, #6388 failed to do so (#6429)
Before this patch, the select! still retured immediately if `futs` was
empty. Must have tested a stale build in my manual testing of #6388.

(cherry picked from commit 15c0df4de7)
2024-01-22 15:23:12 +00:00
Anna Khanova
299d9474c9 Proxy: fix gc (#6426)
## Problem

Gc currently doesn't work properly.

## Summary of changes

Change statement on running gc.
2024-01-22 14:39:09 +01:00
Conrad Ludgate
7234208b36 bump shlex (#6421)
## Problem

https://rustsec.org/advisories/RUSTSEC-2024-0006

## Summary of changes

`cargo update -p shlex`

(cherry picked from commit 5559b16953)
2024-01-22 09:49:33 +00:00
Christian Schwarz
93450f11f5 Merge pull request #6354 from neondatabase/releases/2024-01-15
Release 2024-01-15

NB: the previous release PR https://github.com/neondatabase/neon/pull/6286 was accidentally merged by merge-by-squash instead of merge-by-merge-commit.
See https://github.com/neondatabase/neon/pull/6354#issuecomment-1891706321 for more context.
2024-01-15 14:30:25 +01:00
Christian Schwarz
2f0f9edf33 Merge remote-tracking branch 'origin/release' into releases/2024-01-15 2024-01-15 09:36:42 +00:00
Christian Schwarz
d424f2b7c8 empty commit so we can produce a merge commit 2024-01-15 09:36:22 +00:00
Christian Schwarz
21315e80bc Merge branch 'releases/2024-01-08--not-squashed' into releases/2024-01-15 2024-01-15 09:31:07 +00:00
vipvap
483b66d383 Merge branch 'release' into releases/2024-01-08 (not-squashed merge of #6286)
Release PR https://github.com/neondatabase/neon/pull/6286 got
accidentally merged-by-squash intstead of merge-by-merge-commit.

This commit shows how things would look like if 6286 had been
merged-by-squash.

```
git reset --hard 9f1327772
git merge --no-ff 5c0264b591
```

Co-authored-by: Christian Schwarz <christian@neon.tech>
2024-01-15 09:28:08 +00:00
vipvap
aa72a22661 Release 2024-01-08 (#6286)
Release 2024-01-08
2024-01-08 09:26:27 +00:00
Shany Pozin
5c0264b591 Merge branch 'release' into releases/2024-01-08 2024-01-08 09:34:06 +02:00
Arseny Sher
9f13277729 Merge pull request #6242 from neondatabase/releases/2024-01-02
Release 2024-01-02
2024-01-02 12:04:43 +04:00
Arseny Sher
54aa319805 Don't split WAL record across two XLogData's when sending from safekeepers.
As protocol demands. Not following this makes standby complain about corrupted
WAL in various ways.

https://neondb.slack.com/archives/C05L7D1JAUS/p1703774799114719
closes https://github.com/neondatabase/cloud/issues/9057
2024-01-02 10:54:00 +04:00
Arseny Sher
4a227484bf Add large insertion and slow WAL sending to test_hot_standby.
To exercise MAX_SEND_SIZE sending from safekeeper; we've had a bug with WAL
records torn across several XLogData messages. Add failpoint to safekeeper to
slow down sending. Also check for corrupted WAL complains in standby log.

Make the test a bit simpler in passing, e.g. we don't need explicit commits as
autocommit is enabled by default.

https://neondb.slack.com/archives/C05L7D1JAUS/p1703774799114719
https://github.com/neondatabase/cloud/issues/9057
2024-01-02 10:54:00 +04:00
Arseny Sher
2f83f85291 Add failpoint support to safekeeper.
Just a copy paste from pageserver.
2024-01-02 10:54:00 +04:00
Arseny Sher
d6cfcb0d93 Move failpoint support code to utils.
To enable them in safekeeper as well.
2024-01-02 10:54:00 +04:00
Arseny Sher
392843ad2a Fix safekeeper START_REPLICATION (term=n).
It was giving WAL only up to commit_lsn instead of flush_lsn, so recovery of
uncommitted WAL since cdb08f03 hanged. Add test for this.
2024-01-02 10:54:00 +04:00
Arseny Sher
bd4dae8f4a compute_ctl: kill postgres and sync-safekeeprs on exit.
Otherwise they are left orphaned when compute_ctl is terminated with a
signal. It was invisible most of the time because normally neon_local or k8s
kills postgres directly and then compute_ctl finishes gracefully. However, in
some tests compute_ctl gets stuck waiting for sync-safekeepers which
intentionally never ends because safekeepers are offline, and we want to stop
compute_ctl without leaving orphanes behind.

This is a quite rough approach which doesn't wait for children termination. A
better way would be to convert compute_ctl to async which would make waiting
easy.
2024-01-02 10:54:00 +04:00
Shany Pozin
b05fe53cfd Merge pull request #6240 from neondatabase/releases/2024-01-01
Release 2024-01-01
2024-01-01 11:07:30 +02:00
Christian Schwarz
c13a2f0df1 Merge pull request #6192 from neondatabase/releases/2023-12-19
Release 2023-12-19

We need to do a config change that requires restarting the pageservers.
Slip in two metrics-related commits that didn't make this week's regularly release.
2023-12-19 14:52:47 +01:00
Christian Schwarz
39be366fc5 higher resolution histograms for getpage@lsn (#6177)
part of https://github.com/neondatabase/cloud/issues/7811
2023-12-19 13:46:59 +00:00
Christian Schwarz
6eda0a3158 [PRE-MERGE] fix metric pageserver_initial_logical_size_start_calculation
(This is a pre-merge cherry-pick of https://github.com/neondatabase/neon/pull/6191)

It wasn't being incremented.

Fixup of

    commit 1c88824ed0
    Author: Christian Schwarz <christian@neon.tech>
    Date:   Fri Dec 1 12:52:59 2023 +0100

        initial logical size calculation: add a bunch of metrics (#5995)
2023-12-19 13:46:55 +00:00
Shany Pozin
306c7a1813 Merge pull request #6173 from neondatabase/sasha_release_bypassrls_replication
Grant BYPASSRLS and REPLICATION explicitly to neon_superuser roles
2023-12-18 22:16:36 +02:00
Sasha Krassovsky
80be423a58 Grant BYPASSRLS and REPLICATION explicitly to neon_superuser roles 2023-12-18 10:22:36 -08:00
Shany Pozin
5dcfef82f2 Merge pull request #6163 from neondatabase/releases/2023-12-18
Release 2023-12-18-2
2023-12-18 15:34:17 +02:00
Christian Schwarz
e67b8f69c0 [PRE-MERGE] pageserver: Reduce tracing overhead in timeline::get #6115
Pre-merge `git merge --squash` of
https://github.com/neondatabase/neon/pull/6115

Lowering the tracing level in get_value_reconstruct_data and
get_or_maybe_download from info to debug reduces the overhead
of span creation in non-debug environments.
2023-12-18 13:39:48 +01:00
Shany Pozin
e546872ab4 Merge pull request #6158 from neondatabase/releases/2023-12-18
Release 2023-12-18
2023-12-18 14:24:34 +02:00
John Spray
322ea1cf7c pageserver: on-demand activation cleanups (#6157)
## Problem

#6112 added some logs and metrics: clean these up a bit:
- Avoid counting startup completions for tenants launched after startup
- exclude no-op cases from timing histograms 
- remove a rogue log messages
2023-12-18 11:14:19 +00:00
Vadim Kharitonov
3633742de9 Merge pull request #6121 from neondatabase/releases/2023-12-13
Release 2023-12-13
2023-12-13 12:39:43 +01:00
Joonas Koivunen
079d3a37ba Merge remote-tracking branch 'origin/release' into releases/2023-12-13
this handles the hotfix introduced conflict.
2023-12-13 10:07:19 +00:00
Vadim Kharitonov
a46e77b476 Merge pull request #6090 from neondatabase/releases/2023-12-11
Release 2023-12-11
2023-12-12 12:10:35 +01:00
Tristan Partin
a92702b01e Add submodule paths as safe directories as a precaution
The check-codestyle-rust-arm job requires this for some reason, so let's
just add them everywhere we do this workaround.
2023-12-11 22:00:35 +00:00
Tristan Partin
8ff3253f20 Fix git ownership issue in check-codestyle-rust-arm
We have this workaround for other jobs. Looks like this one was
forgotten about.
2023-12-11 22:00:35 +00:00
Joonas Koivunen
04b82c92a7 fix: accidential return Ok (#6106)
Error indicating request cancellation OR timeline shutdown was deemed as
a reason to exit the background worker that calculated synthetic size.
Fix it to only be considered for avoiding logging such of such errors.

This conflicted on tenant_shard_id having already replaced tenant_id on
`main`.
2023-12-11 21:41:36 +00:00
Vadim Kharitonov
e5bf423e68 Merge branch 'release' into releases/2023-12-11 2023-12-11 11:55:48 +01:00
Vadim Kharitonov
60af392e45 Merge pull request #6057 from neondatabase/vk/patch_timescale_for_production
Revert timescaledb for pg14 and pg15 (#6056)
2023-12-06 16:21:16 +01:00
Vadim Kharitonov
661fc41e71 Revert timescaledb for pg14 and pg15 (#6056)
```
could not start the compute node: compute is in state "failed": db error: ERROR: could not access file "$libdir/timescaledb-2.10.1": No such file or directory Caused by: ERROR: could not access file "$libdir/timescaledb-2.10.1": No such file or directory
```
2023-12-06 16:14:07 +01:00
Shany Pozin
702c488f32 Merge pull request #6022 from neondatabase/releases/2023-12-04
Release 2023-12-04
2023-12-05 17:03:28 +02:00
Sasha Krassovsky
45c5122754 Remove trusted from wal2json 2023-12-04 12:36:19 -08:00
Shany Pozin
558394f710 fix merge 2023-12-04 11:41:27 +02:00
Shany Pozin
73b0898608 Merge branch 'release' into releases/2023-12-04 2023-12-04 11:36:26 +02:00
Joonas Koivunen
e65be4c2dc Merge pull request #6013 from neondatabase/releases/2023-12-01-hotfix
fix: use create_new instead of create for mutex file
2023-12-01 15:35:56 +02:00
Joonas Koivunen
40087b8164 fix: use create_new instead of create for mutex file 2023-12-01 12:54:49 +00:00
Shany Pozin
c762b59483 Merge pull request #5986 from neondatabase/Release-11-30-hotfix
Notify safekeeper readiness with systemd.
2023-11-30 10:01:05 +02:00
Arseny Sher
5d71601ca9 Notify safekeeper readiness with systemd.
To avoid downtime during deploy, as in busy regions initial load can currently
take ~30s.
2023-11-30 08:23:31 +03:00
Shany Pozin
a113c3e433 Merge pull request #5945 from neondatabase/release-2023-11-28-hotfix
Release 2023 11 28 hotfix
2023-11-28 08:14:59 +02:00
Anastasia Lubennikova
e81fc598f4 Update neon extension relocatable for existing installations (#5943) 2023-11-28 00:12:39 +00:00
Anastasia Lubennikova
48b845fa76 Make neon extension relocatable to allow SET SCHEMA (#5942) 2023-11-28 00:12:32 +00:00
Shany Pozin
27096858dc Merge pull request #5922 from neondatabase/releases/2023-11-27
Release 2023-11-27
2023-11-27 09:58:51 +02:00
Shany Pozin
4430d0ae7d Merge pull request #5876 from neondatabase/releases/2023-11-17
Release 2023-11-17
2023-11-20 09:11:58 +02:00
Joonas Koivunen
6e183aa0de Merge branch 'main' into releases/2023-11-17 2023-11-19 15:25:47 +00:00
Vadim Kharitonov
fd6d0b7635 Merge branch 'release' into releases/2023-11-17 2023-11-17 10:51:45 +01:00
Vadim Kharitonov
3710c32aae Merge pull request #5778 from neondatabase/releases/2023-11-03
Release 2023-11-03
2023-11-03 16:06:58 +01:00
Vadim Kharitonov
be83bee49d Merge branch 'release' into releases/2023-11-03 2023-11-03 11:18:15 +01:00
Alexander Bayandin
cf28e5922a Merge pull request #5685 from neondatabase/releases/2023-10-26
Release 2023-10-26
2023-10-27 10:42:12 +01:00
Em Sharnoff
7d384d6953 Bump vm-builder v0.18.2 -> v0.18.4 (#5666)
Only applicable change was neondatabase/autoscaling#584, setting
pgbouncer auth_dbname=postgres in order to fix superuser connections
from preventing dropping databases.
2023-10-26 20:15:45 +01:00
Em Sharnoff
4b3b37b912 Bump vm-builder v0.18.1 -> v0.18.2 (#5646)
Only applicable change was neondatabase/autoscaling#571, removing the
postgres_exporter flags `--auto-discover-databases` and
`--exclude-databases=...`
2023-10-26 20:15:29 +01:00
Shany Pozin
1d8d200f4d Merge pull request #5668 from neondatabase/sp/aux_files_cherry_pick
Cherry pick: Ignore missed AUX_FILES_KEY when generating image layer (#5660)
2023-10-26 10:08:16 +03:00
Konstantin Knizhnik
0d80d6ce18 Ignore missed AUX_FILES_KEY when generating image layer (#5660)
## Problem

Logical replication requires new AUX_FILES_KEY which is definitely
absent in existed database.
We do not have function to check if key exists in our KV storage.
So I have to handle the error in `list_aux_files` method.
But this key is also included in key space range and accessed y
`create_image_layer` method.

## Summary of changes

Check if AUX_FILES_KEY  exists before including it in keyspace.

---------

Co-authored-by: Konstantin Knizhnik <knizhnik@neon.tech>
Co-authored-by: Shany Pozin <shany@neon.tech>
Co-authored-by: Arpad Müller <arpad-m@users.noreply.github.com>
2023-10-26 09:30:28 +03:00
Shany Pozin
f653ee039f Merge pull request #5638 from neondatabase/releases/2023-10-24
Release 2023-10-24
2023-10-24 12:10:52 +03:00
Em Sharnoff
e614a95853 Merge pull request #5610 from neondatabase/sharnoff/rc-2023-10-20-vm-monitor-fixes
Release 2023-10-20: vm-monitor memory.high throttling fixes
2023-10-20 00:11:06 -07:00
Em Sharnoff
850db4cc13 vm-monitor: Deny not fail downscale if no memory stats yet (#5606)
Fixes an issue we observed on staging that happens when the
autoscaler-agent attempts to immediately downscale the VM after binding,
which is typical for pooled computes.

The issue was occurring because the autoscaler-agent was requesting
downscaling before the vm-monitor had gathered sufficient cgroup memory
stats to be confident in approving it. When the vm-monitor returned an
internal error instead of denying downscaling, the autoscaler-agent
retried the connection and immediately hit the same issue (in part
because cgroup stats are collected per-connection, rather than
globally).
2023-10-19 21:56:55 -07:00
Em Sharnoff
8a316b1277 vm-monitor: Log full error on message handling failure (#5604)
There's currently an issue with the vm-monitor on staging that's not
really feasible to debug because the current display impl gives no
context to the errors (just says "failed to downscale").

Logging the full error should help.

For communications with the autoscaler-agent, it's ok to only provide
the outermost cause, because we can cross-reference with the VM logs.
At some point in the future, we may want to change that.
2023-10-19 21:56:50 -07:00
Em Sharnoff
4d13bae449 vm-monitor: Switch from memory.high to polling memory.stat (#5524)
tl;dr it's really hard to avoid throttling from memory.high, and it
counts tmpfs & page cache usage, so it's also hard to make sense of.

In the interest of fixing things quickly with something that should be
*good enough*, this PR switches to instead periodically fetch memory
statistics from the cgroup's memory.stat and use that data to determine
if and when we should upscale.

This PR fixes #5444, which has a lot more detail on the difficulties
we've hit with memory.high. This PR also supersedes #5488.
2023-10-19 21:56:36 -07:00
Vadim Kharitonov
49377abd98 Merge pull request #5577 from neondatabase/releases/2023-10-17
Release 2023-10-17
2023-10-17 12:21:20 +02:00
Christian Schwarz
a6b2f4e54e limit imitate accesses concurrency, using same semaphore as compactions (#5578)
Before this PR, when we restarted pageserver, we'd see a rush of
`$number_of_tenants` concurrent eviction tasks starting to do imitate
accesses building up in the period of `[init_order allows activations,
$random_access_delay + EvictionPolicyLayerAccessThreshold::period]`.

We simply cannot handle that degree of concurrent IO.

We already solved the problem for compactions by adding a semaphore.
So, this PR shares that semaphore for use by evictions.

Part of https://github.com/neondatabase/neon/issues/5479

Which is again part of https://github.com/neondatabase/neon/issues/4743

Risks / Changes In System Behavior
==================================

* we don't do evictions as timely as we currently do
* we log a bunch of warnings about eviction taking too long
* imitate accesses and compactions compete for the same concurrency
limit, so, they'll slow each other down through this shares semaphore

Changes
=======

- Move the `CONCURRENT_COMPACTIONS` semaphore into `tasks.rs`
- Rename it to `CONCURRENT_BACKGROUND_TASKS`
- Use it also for the eviction imitate accesses:
    - Imitate acceses are both per-TIMELINE and per-TENANT
    - The per-TENANT is done through coalescing all the per-TIMELINE
      tasks via a tokio mutex `eviction_task_tenant_state`.
    - We acquire the CONCURRENT_BACKGROUND_TASKS permit early, at the
      beginning of the eviction iteration, much before the imitate
      acesses start (and they may not even start at all in the given
      iteration, as they happen only every $threshold).
    - Acquiring early is **sub-optimal** because when the per-timline
      tasks coalesce on the `eviction_task_tenant_state` mutex,
      they are already holding a CONCURRENT_BACKGROUND_TASKS permit.
    - It's also unfair because tenants with many timelines win
      the CONCURRENT_BACKGROUND_TASKS more often.
    - I don't think there's another way though, without refactoring
      more of the imitate accesses logic, e.g, making it all per-tenant.
- Add metrics for queue depth behind the semaphore.
I found these very useful to understand what work is queued in the
system.

    - The metrics are tagged by the new `BackgroundLoopKind`.
    - On a green slate, I would have used `TaskKind`, but we already had
      pre-existing labels whose names didn't map exactly to task kind.
      Also the task kind is kind of a lower-level detail, so, I think
it's fine to have a separate enum to identify background work kinds.

Future Work
===========

I guess I could move the eviction tasks from a ticker to "sleep for
$period".
The benefit would be that the semaphore automatically "smears" the
eviction task scheduling over time, so, we only have the rush on restart
but a smeared-out rush afterward.

The downside is that this perverts the meaning of "$period", as we'd
actually not run the eviction at a fixed period. It also means the the
"took to long" warning & metric becomes meaningless.

Then again, that is already the case for the compaction and gc tasks,
which do sleep for `$period` instead of using a ticker.

(cherry picked from commit 9256788273)
2023-10-17 12:16:26 +02:00
Shany Pozin
face60d50b Merge pull request #5526 from neondatabase/releases/2023-10-11
Release 2023-10-11
2023-10-11 11:16:39 +03:00
Shany Pozin
9768aa27f2 Merge pull request #5516 from neondatabase/releases/2023-10-10
Release 2023-10-10
2023-10-10 14:16:47 +03:00
Shany Pozin
96b2e575e1 Merge pull request #5445 from neondatabase/releases/2023-10-03
Release 2023-10-03
2023-10-04 13:53:37 +03:00
Alexander Bayandin
7222777784 Update checksums for pg_jsonschema & pg_graphql (#5455)
## Problem

Folks have re-taged releases for `pg_jsonschema` and `pg_graphql` (to
increase timeouts on their CI), for us, these are a noop changes, 
but unfortunately, this will cause our builds to fail due to checksums 
mismatch (this might not strike right away because of the build cache).
- 8ba7c7be9d
- aa7509370a

## Summary of changes
- `pg_jsonschema` update checksum
- `pg_graphql` update checksum
2023-10-03 18:44:30 +01:00
Em Sharnoff
5469fdede0 Merge pull request #5422 from neondatabase/sharnoff/rc-2023-09-28-fix-restart-on-postmaster-SIGKILL
Release 2023-09-28: Fix (lack of) restart on neonvm postmaster SIGKILL
2023-09-28 10:48:51 -07:00
MMeent
72aa6b9fdd Fix neon_zeroextend's WAL logging (#5387)
When you log more than a few blocks, you need to reserve the space in
advance. We didn't do that, so we got errors. Now we do that, and
shouldn't get errors.
2023-09-28 09:37:28 -07:00
Em Sharnoff
ae0634b7be Bump vm-builder v0.17.11 -> v0.17.12 (#5407)
Only relevant change is neondatabase/autoscaling#534 - refer there for
more details.
2023-09-28 09:28:04 -07:00
Shany Pozin
70711f32fa Merge pull request #5375 from neondatabase/releases/2023-09-26
Release 2023-09-26
2023-09-26 15:19:45 +03:00
Vadim Kharitonov
52a88af0aa Merge pull request #5336 from neondatabase/releases/2023-09-19
Release 2023-09-19
2023-09-19 11:16:43 +02:00
Alexander Bayandin
b7a43bf817 Merge branch 'release' into releases/2023-09-19 2023-09-19 09:07:20 +01:00
Alexander Bayandin
dce91b33a4 Merge pull request #5318 from neondatabase/releases/2023-09-15-1
Postgres 14/15: Use previous extensions versions
2023-09-15 16:30:44 +01:00
Alexander Bayandin
23ee4f3050 Revert plv8 only 2023-09-15 15:45:23 +01:00
Alexander Bayandin
46857e8282 Postgres 14/15: Use previous extensions versions 2023-09-15 15:27:00 +01:00
Alexander Bayandin
368ab0ce54 Merge pull request #5313 from neondatabase/releases/2023-09-15
Release 2023-09-15
2023-09-15 10:39:56 +01:00
Konstantin Knizhnik
a5987eebfd References to old and new blocks were mixed in xlog_heap_update handler (#5312)
## Problem

See https://neondb.slack.com/archives/C05L7D1JAUS/p1694614585955029

https://www.notion.so/neondatabase/Duplicate-key-issue-651627ce843c45188fbdcb2d30fd2178

## Summary of changes

Swap old/new block references

## Checklist before requesting a review

- [ ] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.

## Checklist before merging

- [ ] Do not forget to reformat commit message to not include the above
checklist

---------

Co-authored-by: Konstantin Knizhnik <knizhnik@neon.tech>
Co-authored-by: Heikki Linnakangas <heikki@neon.tech>
2023-09-15 10:11:41 +01:00
Alexander Bayandin
6686ede30f Update checksum for pg_hint_plan (#5309)
## Problem

The checksum for `pg_hint_plan` doesn't match:
```
sha256sum: WARNING: 1 computed checksum did NOT match
```

Ref
https://github.com/neondatabase/neon/actions/runs/6185715461/job/16793609251?pr=5307

It seems that the release was retagged yesterday:
https://github.com/ossc-db/pg_hint_plan/releases/tag/REL16_1_6_0

I don't see any malicious changes from 15_1.5.1:
https://github.com/ossc-db/pg_hint_plan/compare/REL15_1_5_1...REL16_1_6_0,
so it should be ok to update.

## Summary of changes
- Update checksum for `pg_hint_plan` 16_1.6.0
2023-09-15 09:54:42 +01:00
Em Sharnoff
373c7057cc vm-monitor: Fix cgroup throttling (#5303)
I believe this (not actual IO problems) is the cause of the "disk speed
issue" that we've had for VMs recently. See e.g.:

1. https://neondb.slack.com/archives/C03H1K0PGKH/p1694287808046179?thread_ts=1694271790.580099&cid=C03H1K0PGKH
2. https://neondb.slack.com/archives/C03H1K0PGKH/p1694511932560659

The vm-informant (and now, the vm-monitor, its replacement) is supposed
to gradually increase the `neon-postgres` cgroup's memory.high value,
because otherwise the kernel will throttle all the processes in the
cgroup.

This PR fixes a bug with the vm-monitor's implementation of this
behavior.

---

Other references, for the vm-informant's implementation:

- Original issue: neondatabase/autoscaling#44
- Original PR: neondatabase/autoscaling#223
2023-09-15 09:54:42 +01:00
Shany Pozin
7d6ec16166 Merge pull request #5296 from neondatabase/releases/2023-09-13
Release 2023-09-13
2023-09-13 13:49:14 +03:00
Shany Pozin
0e6fdc8a58 Merge pull request #5283 from neondatabase/releases/2023-09-12
Release 2023-09-12
2023-09-12 14:56:47 +03:00
Christian Schwarz
521438a5c6 fix deadlock around TENANTS (#5285)
The sequence that can lead to a deadlock:

1. DELETE request gets all the way to `tenant.shutdown(progress,
false).await.is_err() ` , while holding TENANTS.read()
2. POST request for tenant creation comes in, calls `tenant_map_insert`,
it does `let mut guard = TENANTS.write().await;`
3. Something that `tenant.shutdown()` needs to wait for needs a
`TENANTS.read().await`.
The only case identified in exhaustive manual scanning of the code base
is this one:
Imitate size access does `get_tenant().await`, which does
`TENANTS.read().await` under the hood.

In the above case (1) waits for (3), (3)'s read-lock request is queued
behind (2)'s write-lock, and (2) waits for (1).
Deadlock.

I made a reproducer/proof-that-above-hypothesis-holds in
https://github.com/neondatabase/neon/pull/5281 , but, it's not ready for
merge yet and we want the fix _now_.

fixes https://github.com/neondatabase/neon/issues/5284
2023-09-12 14:13:13 +03:00
Vadim Kharitonov
07d7874bc8 Merge pull request #5202 from neondatabase/releases/2023-09-05
Release 2023-09-05
2023-09-05 12:16:06 +02:00
Anastasia Lubennikova
1804111a02 Merge pull request #5161 from neondatabase/rc-2023-08-31
Release 2023-08-31
2023-08-31 16:53:17 +03:00
Arthur Petukhovsky
cd0178efed Merge pull request #5150 from neondatabase/release-sk-fix-active-timeline
Release 2023-08-30
2023-08-30 11:43:39 +02:00
Shany Pozin
333574be57 Merge pull request #5133 from neondatabase/releases/2023-08-29
Release 2023-08-29
2023-08-29 14:02:58 +03:00
Alexander Bayandin
79a799a143 Merge branch 'release' into releases/2023-08-29 2023-08-29 11:17:57 +01:00
Conrad Ludgate
9da06af6c9 Merge pull request #5113 from neondatabase/release-http-connection-fix
Release 2023-08-25
2023-08-25 17:21:35 +01:00
Conrad Ludgate
ce1753d036 proxy: dont return connection pending (#5107)
## Problem

We were returning Pending when a connection had a notice/notification
(introduced recently in #5020). When returning pending, the runtime
assumes you will call `cx.waker().wake()` in order to continue
processing.

We weren't doing that, so the connection task would get stuck

## Summary of changes

Don't return pending. Loop instead
2023-08-25 16:42:30 +01:00
Alek Westover
67db8432b4 Fix cargo deny errors (#5068)
## Problem
cargo deny lint broken

Links to the CVEs:

[rustsec.org/advisories/RUSTSEC-2023-0052](https://rustsec.org/advisories/RUSTSEC-2023-0052)

[rustsec.org/advisories/RUSTSEC-2023-0053](https://rustsec.org/advisories/RUSTSEC-2023-0053)
One is fixed, the other one isn't so we allow it (for now), to unbreak
CI. Then later we'll try to get rid of webpki in favour of the rustls
fork.

## Summary of changes
```
+ignore = ["RUSTSEC-2023-0052"]
```
2023-08-25 16:42:30 +01:00
Vadim Kharitonov
4e2e44e524 Enable neon-pool-opt-in (#5062) 2023-08-22 09:06:14 +01:00
Vadim Kharitonov
ed786104f3 Merge pull request #5060 from neondatabase/releases/2023-08-22
Release 2023-08-22
2023-08-22 09:41:02 +02:00
Stas Kelvich
84b74f2bd1 Merge pull request #4997 from neondatabase/sk/proxy-release-23-07-15
Fix lint
2023-08-15 18:54:20 +03:00
Arthur Petukhovsky
fec2ad6283 Fix lint 2023-08-15 18:49:02 +03:00
Stas Kelvich
98eebd4682 Merge pull request #4996 from neondatabase/sk/proxy_release
Disable neon-pool-opt-in
2023-08-15 18:37:50 +03:00
Arthur Petukhovsky
2f74287c9b Disable neon-pool-opt-in 2023-08-15 18:34:17 +03:00
Shany Pozin
aee1bf95e3 Merge pull request #4990 from neondatabase/releases/2023-08-15
Release 2023-08-15
2023-08-15 15:34:38 +03:00
Shany Pozin
b9de9d75ff Merge branch 'release' into releases/2023-08-15 2023-08-15 14:35:00 +03:00
Stas Kelvich
7943b709e6 Merge pull request #4940 from neondatabase/sk/release-23-05-25-proxy-fixup
Release: proxy retry fixup
2023-08-09 13:53:19 +03:00
Conrad Ludgate
d7d066d493 proxy: delay auth on retry (#4929)
## Problem

When an endpoint is shutting down, it can take a few seconds. Currently
when starting a new compute, this causes an "endpoint is in transition"
error. We need to add delays before retrying to ensure that we allow
time for the endpoint to shutdown properly.

## Summary of changes

Adds a delay before retrying in auth. connect_to_compute already has
this delay
2023-08-09 12:54:24 +03:00
Felix Prasanna
e78ac22107 release fix: revert vm builder bump from 0.13.1 -> 0.15.0-alpha1 (#4932)
This reverts commit 682dfb3a31.

hotfix for a CLI arg issue in the monitor
2023-08-08 21:08:46 +03:00
Vadim Kharitonov
76a8f2bb44 Merge pull request #4923 from neondatabase/releases/2023-08-08
Release 2023-08-08
2023-08-08 11:44:38 +02:00
Vadim Kharitonov
8d59a8581f Merge branch 'release' into releases/2023-08-08 2023-08-08 10:54:34 +02:00
Vadim Kharitonov
b1ddd01289 Define NEON_SMGR to make it possible for extensions to use Neon SMG API (#4889)
Co-authored-by: Konstantin Knizhnik <knizhnik@garret.ru>
Co-authored-by: Konstantin Knizhnik <knizhnik@neon.tech>
2023-08-03 16:28:31 +03:00
Alexander Bayandin
6eae4fc9aa Release 2023-08-02: update pg_embedding (#4877)
Cherry-picking ca4d71a954 from `main` into
the `release`

Co-authored-by: Vadim Kharitonov <vadim2404@users.noreply.github.com>
2023-08-03 08:48:09 +02:00
Christian Schwarz
765455bca2 Merge pull request #4861 from neondatabase/releases/2023-08-01--2-fix-pipeline
ci: fix upload-postgres-extensions-to-s3 job
2023-08-01 13:22:07 +02:00
Christian Schwarz
4204960942 ci: fix upload-postgres-extensions-to-s3 job
commit

	commit 5f8fd640bf
	Author: Alek Westover <alek.westover@gmail.com>
	Date:   Wed Jul 26 08:24:03 2023 -0400

	    Upload Test Remote Extensions (#4792)

switched to using the release tag instead of `latest`, but,
the `promote-images` job only uploads `latest` to the prod ECR.

The switch to using release tag was good in principle, but,
reverting that part to make the release pipeine work.

Note that a proper fix should abandon use of `:latest` tag
at all: currently, if a `main` pipeline runs concurrently
with a `release` pipeline, the `release` pipeline may end
up using the `main` pipeline's images.
2023-08-01 12:01:45 +02:00
Christian Schwarz
67345d66ea Merge pull request #4858 from neondatabase/releases/2023-08-01
Release 2023-08-01
2023-08-01 10:44:01 +02:00
Shany Pozin
2266ee5971 Merge pull request #4803 from neondatabase/releases/2023-07-25
Release 2023-07-25
2023-07-25 14:21:07 +03:00
Shany Pozin
b58445d855 Merge pull request #4746 from neondatabase/releases/2023-07-18
Release 2023-07-18
2023-07-18 14:45:39 +03:00
Conrad Ludgate
36050e7f3d Merge branch 'release' into releases/2023-07-18 2023-07-18 12:00:09 +01:00
Alexander Bayandin
33360ed96d Merge pull request #4705 from neondatabase/release-2023-07-12
Release 2023-07-12 (only proxy)
2023-07-12 19:44:36 +01:00
Conrad Ludgate
39a28d1108 proxy wake_compute loop (#4675)
## Problem

If we fail to wake up the compute node, a subsequent connect attempt
will definitely fail. However, kubernetes won't fail the connection
immediately, instead it hangs until we timeout (10s).

## Summary of changes

Refactor the loop to allow fast retries of compute_wake and to skip a
connect attempt.
2023-07-12 18:40:11 +01:00
Conrad Ludgate
efa6aa134f allow repeated IO errors from compute node (#4624)
## Problem

#4598 compute nodes are not accessible some time after wake up due to
kubernetes DNS not being fully propagated.

## Summary of changes

Update connect retry mechanism to support handling IO errors and
sleeping for 100ms

## Checklist before requesting a review

- [x] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
2023-07-12 18:40:06 +01:00
Alexander Bayandin
2c724e56e2 Merge pull request #4646 from neondatabase/releases/2023-07-06-hotfix
Release 2023-07-06 (add pg_embedding extension only)
2023-07-06 12:19:52 +01:00
Alexander Bayandin
feff887c6f Compile pg_embedding extension (#4634)
```
CREATE EXTENSION embedding;
CREATE TABLE t (val real[]);
INSERT INTO t (val) VALUES ('{0,0,0}'), ('{1,2,3}'), ('{1,1,1}'), (NULL);
CREATE INDEX ON t USING hnsw (val) WITH (maxelements = 10, dims=3, m=3);
INSERT INTO t (val) VALUES (array[1,2,4]);

SELECT * FROM t ORDER BY val <-> array[3,3,3];
   val   
---------
 {1,2,3}
 {1,2,4}
 {1,1,1}
 {0,0,0}
 
(5 rows)
```
2023-07-06 09:39:41 +01:00
Vadim Kharitonov
353d915fcf Merge pull request #4633 from neondatabase/releases/2023-07-05
Release 2023-07-05
2023-07-05 15:10:47 +02:00
Vadim Kharitonov
2e38098cbc Merge branch 'release' into releases/2023-07-05 2023-07-05 12:41:48 +02:00
Vadim Kharitonov
a6fe5ea1ac Merge pull request #4571 from neondatabase/releases/2023-06-27
Release 2023-06-27
2023-06-27 12:55:33 +02:00
Vadim Kharitonov
05b0aed0c1 Merge branch 'release' into releases/2023-06-27 2023-06-27 12:22:12 +02:00
Alex Chi Z
cd1705357d Merge pull request #4561 from neondatabase/releases/2023-06-23-hotfix
Release 2023-06-23 (pageserver-only)
2023-06-23 15:38:50 -04:00
Christian Schwarz
6bc7561290 don't use MGMT_REQUEST_RUNTIME for consumption metrics synthetic size worker
The consumption metrics synthetic size worker does logical size calculation.
Logical size calculation currently does synchronous disk IO.
This blocks the MGMT_REQUEST_RUNTIME's executor threads, starving other futures.

While there's work on the way to move the synchronous disk IO into spawn_blocking,
the quickfix here is to use the BACKGROUND_RUNTIME instead of MGMT_REQUEST_RUNTIME.

Actually it's not just a quickfix. We simply shouldn't be blocking MGMT_REQUEST_RUNTIME
executor threads on CPU or sync disk IO.
That work isn't done yet, as many of the mgmt tasks still _do_ disk IO.
But it's not as intensive as the logical size calculations that we're fixing here.

While we're at it, fix disk-usage-based eviction in a similar way.
It wasn't the culprit here, according to prod logs, but it can theoretically be
a little CPU-intensive.

More context, including graphs from Prod:
https://neondb.slack.com/archives/C03F5SM1N02/p1687541681336949

(cherry picked from commit d6e35222ea)
2023-06-23 20:54:07 +02:00
Christian Schwarz
fbd3ac14b5 Merge pull request #4544 from neondatabase/releases/2023-06-21-hotfix
Release 2023-06-21 (fixup for post-merge failed 2023-06-20)
2023-06-21 16:54:34 +03:00
Christian Schwarz
e437787c8f cargo update -p openssl (#4542)
To unblock release
https://github.com/neondatabase/neon/pull/4536#issuecomment-1600678054

Context: https://rustsec.org/advisories/RUSTSEC-2023-0044
2023-06-21 15:52:56 +03:00
Christian Schwarz
3460dbf90b Merge pull request #4536 from neondatabase/releases/2023-06-20
Release 2023-06-20 (actually 2023-06-21)
2023-06-21 14:19:14 +03:00
Vadim Kharitonov
6b89d99677 Merge pull request #4521 from neondatabase/release_2023-06-15
Release 2023 06 15
2023-06-15 17:40:01 +02:00
Vadim Kharitonov
6cc8ea86e4 Merge branch 'main' into release_2023-06-15 2023-06-15 16:50:44 +02:00
Shany Pozin
e62a492d6f Merge pull request #4486 from neondatabase/releases/2023-06-13
Release 2023-06-13
2023-06-13 15:21:35 +03:00
Alexey Kondratov
a475cdf642 [compute_ctl] Fix logging if catalog updates are skipped (#4480)
Otherwise, it wasn't clear from the log when Postgres started up
completely if catalog updates were skipped.

Follow-up for 4936ab6
2023-06-13 13:37:24 +02:00
Stas Kelvich
7002c79a47 Merge pull request #4447 from neondatabase/release_proxy_08-06-2023
Release proxy 08 06 2023
2023-06-08 21:02:54 +03:00
Vadim Kharitonov
ee6cf357b4 Merge pull request #4427 from neondatabase/releases/2023-06-06
Release 2023-06-06
2023-06-06 14:42:21 +02:00
Vadim Kharitonov
e5c2086b5f Merge branch 'release' into releases/2023-06-06 2023-06-06 12:33:56 +02:00
Shany Pozin
5f1208296a Merge pull request #4395 from neondatabase/releases/2023-06-01
Release 2023-06-01
2023-06-01 10:58:00 +03:00
Stas Kelvich
88e8e473cd Merge pull request #4345 from neondatabase/release-23-05-25-proxy
Release 23-05-25, take 3
2023-05-25 19:40:43 +03:00
Stas Kelvich
b0a77844f6 Add SQL-over-HTTP endpoint to Proxy
This commit introduces an SQL-over-HTTP endpoint in the proxy, with a JSON
response structure resembling that of the node-postgres driver. This method,
using HTTP POST, achieves smaller amortized latencies in edge setups due to
fewer round trips and an enhanced open connection reuse by the v8 engine.

This update involves several intricacies:
1. SQL injection protection: We employed the extended query protocol, modifying
   the rust-postgres driver to send queries in one roundtrip using a text
   protocol rather than binary, bypassing potential issues like those identified
   in https://github.com/sfackler/rust-postgres/issues/1030.

2. Postgres type compatibility: As not all postgres types have binary
   representations (e.g., acl's in pg_class), we adjusted rust-postgres to
   respond with text protocol, simplifying serialization and fixing queries with
   text-only types in response.

3. Data type conversion: Considering JSON supports fewer data types than
   Postgres, we perform conversions where possible, passing all other types as
   strings. Key conversions include:
   - postgres int2, int4, float4, float8 -> json number (NaN and Inf remain
     text)
   - postgres bool, null, text -> json bool, null, string
   - postgres array -> json array
   - postgres json and jsonb -> json object

4. Alignment with node-postgres: To facilitate integration with js libraries,
   we've matched the response structure of node-postgres, returning command tags
   and column oids. Command tag capturing was added to the rust-postgres
   functionality as part of this change.
2023-05-25 17:59:17 +03:00
Vadim Kharitonov
1baf464307 Merge pull request #4309 from neondatabase/releases/2023-05-23
Release 2023-05-23
2023-05-24 11:56:54 +02:00
Alexander Bayandin
e9b8e81cea Merge branch 'release' into releases/2023-05-23 2023-05-23 12:54:08 +01:00
Alexander Bayandin
85d6194aa4 Fix regress-tests job for Postgres 15 on release branch (#4254)
## Problem

Compatibility tests don't support Postgres 15 yet, but we're still
trying to upload compatibility snapshot (which we do not collect).

Ref
https://github.com/neondatabase/neon/actions/runs/4991394158/jobs/8940369368#step:4:38129

## Summary of changes

Add `pg_version` parameter to `run-python-test-set` actions and do not
upload compatibility snapshot for Postgres 15
2023-05-16 17:19:12 +01:00
Vadim Kharitonov
333a7a68ef Merge pull request #4245 from neondatabase/releases/2023-05-16
Release 2023-05-16
2023-05-16 13:38:40 +02:00
Vadim Kharitonov
6aa4e41bee Merge branch 'release' into releases/2023-05-16 2023-05-16 12:48:23 +02:00
Joonas Koivunen
840183e51f try: higher page_service timeouts to isolate an issue 2023-05-11 16:24:53 +03:00
Shany Pozin
cbccc94b03 Merge pull request #4184 from neondatabase/releases/2023-05-09
Release 2023-05-09
2023-05-09 15:30:36 +03:00
Stas Kelvich
fce227df22 Merge pull request #4163 from neondatabase/main
Release 23-05-05
2023-05-05 15:56:23 +03:00
Stas Kelvich
bd787e800f Merge pull request #4133 from neondatabase/main
Release 23-04-01
2023-05-01 18:52:46 +03:00
Shany Pozin
4a7704b4a3 Merge pull request #4131 from neondatabase/sp/hotfix_adding_sks_us_west
Hotfix: Adding 4 new pageservers and two sets of safekeepers to us west 2
2023-05-01 15:17:38 +03:00
Shany Pozin
ff1119da66 Add 2 new sets of safekeepers to us-west2 2023-05-01 14:35:31 +03:00
Shany Pozin
4c3ba1627b Add 4 new Pageservers for retool launch 2023-05-01 14:34:38 +03:00
Vadim Kharitonov
1407174fb2 Merge pull request #4110 from neondatabase/vk/release_2023-04-28
Release 2023 04 28
2023-04-28 17:43:16 +02:00
Vadim Kharitonov
ec9dcb1889 Merge branch 'release' into vk/release_2023-04-28 2023-04-28 16:32:26 +02:00
Joonas Koivunen
d11d781afc revert: "Add check for duplicates of generated image layers" (#4104)
This reverts commit 732acc5.

Reverted PR: #3869

As noted in PR #4094, we do in fact try to insert duplicates to the
layer map, if L0->L1 compaction is interrupted. We do not have a proper
fix for that right now, and we are in a hurry to make a release to
production, so revert the changes related to this to the state that we
have in production currently. We know that we have a bug here, but
better to live with the bug that we've had in production for a long
time, than rush a fix to production without testing it in staging first.

Cc: #4094, #4088
2023-04-28 16:31:35 +02:00
Anastasia Lubennikova
4e44565b71 Merge pull request #4000 from neondatabase/releases/2023-04-11
Release 2023-04-11
2023-04-11 17:47:41 +03:00
Stas Kelvich
4ed51ad33b Add more proxy cnames 2023-04-11 15:59:35 +03:00
Arseny Sher
1c1ebe5537 Merge pull request #3946 from neondatabase/releases/2023-04-04
Release 2023-04-04
2023-04-04 14:38:40 +04:00
Christian Schwarz
c19cb7f386 Merge pull request #3935 from neondatabase/releases/2023-04-03
Release 2023-04-03
2023-04-03 16:19:49 +02:00
Vadim Kharitonov
4b97d31b16 Merge pull request #3896 from neondatabase/releases/2023-03-28
Release 2023-03-28
2023-03-28 17:58:06 +04:00
Shany Pozin
923ade3dd7 Merge pull request #3855 from neondatabase/releases/2023-03-21
Release 2023-03-21
2023-03-21 13:12:32 +02:00
Arseny Sher
b04e711975 Merge pull request #3825 from neondatabase/release-2023-03-15
Release 2023.03.15
2023-03-15 15:38:00 +03:00
Arseny Sher
afd0a6b39a Forward framed read buf contents to compute before proxy pass.
Otherwise they get lost. Normally buffer is empty before proxy pass, but this is
not the case with pipeline mode of out npm driver; fixes connection hangup
introduced by b80fe41af3 for it.

fixes https://github.com/neondatabase/neon/issues/3822
2023-03-15 15:36:06 +04:00
Lassi Pölönen
99752286d8 Use RollingUpdate strategy also for legacy proxy (#3814)
## Describe your changes
We have previously changed the neon-proxy to use RollingUpdate. This
should be enabled in legacy proxy too in order to avoid breaking
connections for the clients and allow for example backups to run even
during deployment. (https://github.com/neondatabase/neon/pull/3683)

## Issue ticket number and link
https://github.com/neondatabase/neon/issues/3333
2023-03-15 15:35:51 +04:00
Arseny Sher
15df93363c Merge pull request #3804 from neondatabase/release-2023-03-13
Release 2023.03.13
2023-03-13 20:25:40 +03:00
Vadim Kharitonov
bc0ab741af Merge pull request #3758 from neondatabase/releases/2023-03-07
Release 2023-03-07
2023-03-07 12:38:47 +01:00
Christian Schwarz
51d9dfeaa3 Merge pull request #3743 from neondatabase/releases/2023-03-03
Release 2023-03-03
2023-03-03 19:20:21 +01:00
Shany Pozin
f63cb18155 Merge pull request #3713 from neondatabase/releases/2023-02-28
Release 2023-02-28
2023-02-28 12:52:24 +02:00
Arseny Sher
0de603d88e Merge pull request #3707 from neondatabase/release-2023-02-24
Release 2023-02-24

Hotfix for UNLOGGED tables. Contains #3706
Also contains rebase on 14.7 and 15.2 #3581
2023-02-25 00:32:11 +04:00
Heikki Linnakangas
240913912a Fix UNLOGGED tables.
Instead of trying to create missing files on the way, send init fork contents as
main fork from pageserver during basebackup. Add test for that. Call
put_rel_drop for init forks; previously they weren't removed. Bump
vendor/postgres to revert previous approach on Postgres side.

Co-authored-by: Arseny Sher <sher-ars@yandex.ru>

ref https://github.com/neondatabase/postgres/pull/264
ref https://github.com/neondatabase/postgres/pull/259
ref https://github.com/neondatabase/neon/issues/1222
2023-02-24 23:54:53 +04:00
MMeent
91a4ea0de2 Update vendored PostgreSQL versions to 14.7 and 15.2 (#3581)
## Describe your changes
Rebase vendored PostgreSQL onto 14.7 and 15.2

## Issue ticket number and link

#3579

## Checklist before requesting a review
- [x] I have performed a self-review of my code.
- [x] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [x] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.
    ```
The version of PostgreSQL that we use is updated to 14.7 for PostgreSQL
14 and 15.2 for PostgreSQL 15.
    ```
2023-02-24 23:54:42 +04:00
Arseny Sher
8608704f49 Merge pull request #3691 from neondatabase/release-2023-02-23
Release 2023-02-23

Hotfix for the unlogged tables with indexes issue.

neondatabase/postgres#259
neondatabase/postgres#262
2023-02-23 13:39:33 +04:00
Arseny Sher
efef68ce99 Bump vendor/postgres to include hotfix for unlogged tables with indexes.
https://github.com/neondatabase/postgres/pull/259
https://github.com/neondatabase/postgres/pull/262
2023-02-23 08:49:43 +04:00
Joonas Koivunen
8daefd24da Merge pull request #3679 from neondatabase/releases/2023-02-22
Releases/2023-02-22
2023-02-22 15:56:55 +02:00
Arthur Petukhovsky
46cc8b7982 Remove safekeeper-1.ap-southeast-1.aws.neon.tech (#3671)
We migrated all timelines to
`safekeeper-3.ap-southeast-1.aws.neon.tech`, now old instance can be
removed.
2023-02-22 15:07:57 +02:00
Sergey Melnikov
38cd90dd0c Add -v to ansible invocations (#3670)
To get more debug output on failures
2023-02-22 15:07:57 +02:00
Joonas Koivunen
a51b269f15 fix: hold permit until GetObject eof (#3663)
previously we applied the ratelimiting only up to receiving the headers
from s3, or somewhere near it. the commit adds an adapter which carries
the permit until the AsyncRead has been disposed.

fixes #3662.
2023-02-22 15:07:57 +02:00
Joonas Koivunen
43bf6d0a0f calculate_logical_size: no longer use spawn_blocking (#3664)
Calculation of logical size is now async because of layer downloads, so
we shouldn't use spawn_blocking for it. Use of `spawn_blocking`
exhausted resources which are needed by `tokio::io::copy` when copying
from a stream to a file which lead to deadlock.

Fixes: #3657
2023-02-22 15:07:57 +02:00
Joonas Koivunen
15273a9b66 chore: ignore all compaction inactive tenant errors (#3665)
these are happening in tests because of #3655 but they sure took some
time to appear.

makes the `Compaction failed, retrying in 2s: Cannot run compaction
iteration on inactive tenant` into a globally allowed error, because it
has been seen failing on different test cases.
2023-02-22 15:07:57 +02:00
Joonas Koivunen
78aca668d0 fix: log download failed error (#3661)
Fixes #3659
2023-02-22 15:07:57 +02:00
Vadim Kharitonov
acbf4148ea Merge pull request #3656 from neondatabase/releases/2023-02-21
Release 2023-02-21
2023-02-21 16:03:48 +01:00
Vadim Kharitonov
6508540561 Merge branch 'release' into releases/2023-02-21 2023-02-21 15:31:16 +01:00
Arthur Petukhovsky
a41b5244a8 Add new safekeeper to ap-southeast-1 prod (#3645) (#3646)
To trigger deployment of #3645 to production.
2023-02-20 15:22:49 +00:00
Shany Pozin
2b3189be95 Merge pull request #3600 from neondatabase/releases/2023-02-14
Release 2023-02-14
2023-02-15 13:31:30 +02:00
Vadim Kharitonov
248563c595 Merge pull request #3553 from neondatabase/releases/2023-02-07
Release 2023-02-07
2023-02-07 14:07:44 +01:00
Vadim Kharitonov
14cd6ca933 Merge branch 'release' into releases/2023-02-07 2023-02-07 12:11:56 +01:00
Vadim Kharitonov
eb36403e71 Release 2023 01 31 (#3497)
Co-authored-by: Kirill Bulatov <kirill@neon.tech>
Co-authored-by: Heikki Linnakangas <heikki@neon.tech>
Co-authored-by: Anastasia Lubennikova <anastasia@neon.tech>
Co-authored-by: bojanserafimov <bojan.serafimov7@gmail.com>
Co-authored-by: Christian Schwarz <christian@neon.tech>
Co-authored-by: Alexey Kondratov <kondratov.aleksey@gmail.com>
Co-authored-by: Joonas Koivunen <joonas@neon.tech>
Co-authored-by: Konstantin Knizhnik <knizhnik@garret.ru>
Co-authored-by: Shany Pozin <shany@neon.tech>
Co-authored-by: Sergey Melnikov <sergey@neon.tech>
Co-authored-by: Dmitry Rodionov <dmitry@neon.tech>
Co-authored-by: Rory de Zoete <33318916+zoete@users.noreply.github.com>
Co-authored-by: Rory de Zoete <rdezoete@Rorys-Mac-Studio.fritz.box>
Co-authored-by: Rory de Zoete <rdezoete@RorysMacStudio.fritz.box>
Co-authored-by: Lassi Pölönen <lassi.polonen@iki.fi>
2023-01-31 15:06:35 +02:00
Anastasia Lubennikova
3c6f779698 Merge pull request #3411 from neondatabase/release_2023_01_23
Fix Release 2023 01 23
2023-01-23 20:10:03 +02:00
Joonas Koivunen
f67f0c1c11 More tenant size fixes (#3410)
Small changes, but hopefully this will help with the panic detected in
staging, for which we cannot get the debugging information right now
(end-of-branch before branch-point).
2023-01-23 17:46:13 +02:00
Shany Pozin
edb02d3299 Adding pageserver3 to staging (#3403) 2023-01-23 17:46:13 +02:00
Konstantin Knizhnik
664a69e65b Fix slru_segment_key_range function: segno was assigned to incorrect Key field (#3354) 2023-01-23 17:46:13 +02:00
Anastasia Lubennikova
478322ebf9 Fix tenant size orphans (#3377)
Before only the timelines which have passed the `gc_horizon` were
processed which failed with orphans at the tree_sort phase. Example
input in added `test_branched_empty_timeline_size` test case.

The PR changes iteration to happen through all timelines, and in
addition to that, any learned branch points will be calculated as they
would had been in the original implementation if the ancestor branch had
been over the `gc_horizon`.

This also changes how tenants where all timelines are below `gc_horizon`
are handled. Previously tenant_size 0 was returned, but now they will
have approximately `initdb_lsn` worth of tenant_size.

The PR also adds several new tenant size tests that describe various corner
cases of branching structure and `gc_horizon` setting.
They are currently disabled to not consume time during CI.

Co-authored-by: Joonas Koivunen <joonas@neon.tech>
Co-authored-by: Anastasia Lubennikova <anastasia@neon.tech>
2023-01-23 17:46:13 +02:00
Joonas Koivunen
802f174072 fix: dont stop pageserver if we fail to calculate synthetic size 2023-01-23 17:46:13 +02:00
Alexey Kondratov
47f9890bae [compute_ctl] Make role deletion spec processing idempotent (#3380)
Previously, we were trying to re-assign owned objects of the already
deleted role. This were causing a crash loop in the case when compute
was restarted with a spec that includes delta operation for role
deletion. To avoid such cases, check that role is still present before
calling `reassign_owned_objects`.

Resolves neondatabase/cloud#3553
2023-01-23 17:46:13 +02:00
Christian Schwarz
262265daad Revert "Use actual temporary dir for pageserver unit tests"
This reverts commit 826e89b9ce.

The problem with that commit was that it deletes the TempDir while
there are still EphemeralFile instances open.

At first I thought this could be fixed by simply adding

  Handle::current().block_on(task_mgr::shutdown(None, Some(tenant_id), None))

to TenantHarness::drop, but it turned out to be insufficient.

So, reverting the commit until we find a proper solution.

refs https://github.com/neondatabase/neon/issues/3385
2023-01-23 17:46:13 +02:00
bojanserafimov
300da5b872 Improve layer map docstrings (#3382) 2023-01-23 17:46:13 +02:00
Heikki Linnakangas
7b22b5c433 Switch to 'tracing' for logging, restructure code to make use of spans.
Refactors Compute::prepare_and_run. It's split into subroutines
differently, to make it easier to attach tracing spans to the
different stages. The high-level logic for waiting for Postgres to
exit is moved to the caller.

Replace 'env_logger' with 'tracing', and add `#instrument` directives
to different stages fo the startup process. This is a fairly
mechanical change, except for the changes in 'spec.rs'. 'spec.rs'
contained some complicated formatting, where parts of log messages
were printed directly to stdout with `print`s. That was a bit messed
up because the log normally goes to stderr, but those lines were
printed to stdout. In our docker images, stderr and stdout both go to
the same place so you wouldn't notice, but I don't think it was
intentional.

This changes the log format to the default
'tracing_subscriber::format' format. It's different from the Postgres
log format, however, and because both compute_tools and Postgres print
to the same log, it's now a mix of two different formats.  I'm not
sure how the Grafana log parsing pipeline can handle that. If it's a
problem, we can build custom formatter to change the compute_tools log
format to be the same as Postgres's, like it was before this commit,
or we can change the Postgres log format to match tracing_formatter's,
or we can start printing compute_tool's log output to a different
destination than Postgres
2023-01-23 17:46:12 +02:00
Kirill Bulatov
ffca97bc1e Enable logs in unit tests 2023-01-23 17:46:12 +02:00
Kirill Bulatov
cb356f3259 Use actual temporary dir for pageserver unit tests 2023-01-23 17:46:12 +02:00
Vadim Kharitonov
c85374295f Change SENTRY_ENVIRONMENT from "development" to "staging" 2023-01-23 17:46:12 +02:00
Anastasia Lubennikova
4992160677 Fix metric_collection_endpoint for prod.
It was incorrectly set to staging url
2023-01-23 17:46:12 +02:00
Heikki Linnakangas
bd535b3371 If an error happens while checking for core dumps, don't panic.
If we panic, we skip the 30s wait in 'main', and don't give the
console a chance to observe the error. Which is not nice.

Spotted by @ololobus at
https://github.com/neondatabase/neon/pull/3352#discussion_r1072806981
2023-01-23 17:46:12 +02:00
Kirill Bulatov
d90c5a03af Add more io::Error context when fail to operate on a path (#3254)
I have a test failure that shows 

```
Caused by:
    0: Failed to reconstruct a page image:
    1: Directory not empty (os error 39)
```

but does not really show where exactly that happens.

https://neon-github-public-dev.s3.amazonaws.com/reports/pr-3227/release/3823785365/index.html#categories/c0057473fc9ec8fb70876fd29a171ce8/7088dab272f2c7b7/?attachment=60fe6ed2add4d82d

The PR aims to add more context in debugging that issue.
2023-01-23 17:46:12 +02:00
Anastasia Lubennikova
2d02cc9079 Merge pull request #3365 from neondatabase/main
Release 2023-01-17
2023-01-17 16:41:34 +02:00
Christian Schwarz
49ad94b99f Merge pull request #3301 from neondatabase/release-2023-01-10
Release 2023-01-10
2023-01-10 16:42:26 +01:00
Christian Schwarz
948a217398 Merge commit '95bf19b85a06b27a7fc3118dee03d48648efab15' into release-2023-01-10
Conflicts:
        .github/helm-values/neon-stress.proxy-scram.yaml
        .github/helm-values/neon-stress.proxy.yaml
        .github/helm-values/staging.proxy-scram.yaml
        .github/helm-values/staging.proxy.yaml
        All of the above were deleted in `main` after we hotfixed them
        in `release. Deleting them here
        storage_broker/src/bin/storage_broker.rs
        Hotfix toned down logging, but `main` has sinced implemented
        a proper fix. Taken `main`'s side, see
        https://neondb.slack.com/archives/C033RQ5SPDH/p1673354385387479?thread_ts=1673354306.474729&cid=C033RQ5SPDH

closes https://github.com/neondatabase/neon/issues/3287
2023-01-10 15:40:14 +01:00
Dmitry Rodionov
125381eae7 Merge pull request #3236 from neondatabase/dkr/retrofit-sk4-sk4-change
Move zenith-1-sk-3 to zenith-1-sk-4 (#3164)
2022-12-30 14:13:50 +03:00
Arthur Petukhovsky
cd01bbc715 Move zenith-1-sk-3 to zenith-1-sk-4 (#3164) 2022-12-30 12:32:52 +02:00
Dmitry Rodionov
d8b5e3b88d Merge pull request #3229 from neondatabase/dkr/add-pageserver-for-release
add pageserver to new region see https://github.com/neondatabase/aws/pull/116

decrease log volume for pageserver
2022-12-30 12:34:04 +03:00
Dmitry Rodionov
06d25f2186 switch to debug from info to produce less noise 2022-12-29 17:48:47 +02:00
Dmitry Rodionov
f759b561f3 add pageserver to new region see https://github.com/neondatabase/aws/pull/116 2022-12-29 17:17:35 +02:00
Sergey Melnikov
ece0555600 Push proxy metrics to Victoria Metrics (#3106) 2022-12-16 14:44:49 +02:00
Joonas Koivunen
73ea0a0b01 fix(remote_storage): use cached credentials (#3128)
IMDSv2 has limits, and if we query it on every s3 interaction we are
going to go over those limits. Changes the s3_bucket client
configuration to use:
- ChainCredentialsProvider to handle env variables or imds usage
- LazyCachingCredentialsProvider to actually cache any credentials

Related: https://github.com/awslabs/aws-sdk-rust/issues/629
Possibly related: https://github.com/neondatabase/neon/issues/3118
2022-12-16 14:44:49 +02:00
Arseny Sher
d8f6d6fd6f Merge pull request #3126 from neondatabase/broker-lb-release
Deploy broker with L4 LB in new env.
2022-12-16 01:25:28 +03:00
Arseny Sher
d24de169a7 Deploy broker with L4 LB in new env.
Seems to be fixing issue with missing keepalives.
2022-12-16 01:45:32 +04:00
Arseny Sher
0816168296 Hotfix: terminate subscription if channel is full.
Might help as a hotfix, but need to understand root better.
2022-12-15 12:23:56 +03:00
Dmitry Rodionov
277b44d57a Merge pull request #3102 from neondatabase/main
Hotfix. See commits for details
2022-12-14 19:38:43 +03:00
MMeent
68c2c3880e Merge pull request #3038 from neondatabase/main
Release 22-12-14
2022-12-14 14:35:47 +01:00
Arthur Petukhovsky
49da498f65 Merge pull request #2833 from neondatabase/main
Release 2022-11-16
2022-11-17 08:44:10 +01:00
Stas Kelvich
2c76ba3dd7 Merge pull request #2718 from neondatabase/main-rc-22-10-28
Release 22-10-28
2022-10-28 20:33:56 +03:00
Arseny Sher
dbe3dc69ad Merge branch 'main' into main-rc-22-10-28
Release 22-10-28.
2022-10-28 19:10:11 +04:00
Arseny Sher
8e5bb3ed49 Enable etcd compaction in neon_local. 2022-10-27 12:53:20 +03:00
Stas Kelvich
ab0be7b8da Avoid debian-testing packages in compute Dockerfiles
plv8 can only be built with a fairly new gold linker version. We used to install
it via binutils packages from testing, but it also updates libc and that causes
troubles in the resulting image as different extensions were built against
different libc versions. We could either use libc from debian-testing everywhere
or restrain from using testing packages and install necessary programs manually.
This patch uses the latter approach: gold for plv8 and cmake for h3 are
installed manually.

In a passing declare h3_postgis as a safe extension (previous omission).
2022-10-27 12:53:20 +03:00
bojanserafimov
b4c55f5d24 Move pagestream api to libs/pageserver_api (#2698) 2022-10-27 12:53:20 +03:00
mikecaat
ede70d833c Add a docker-compose example file (#1943) (#2666)
Co-authored-by: Masahiro Ikeda <masahiro.ikeda.us@hco.ntt.co.jp>
2022-10-27 12:53:20 +03:00
Sergey Melnikov
70c3d18bb0 Do not release to new staging proxies on release (#2685) 2022-10-27 12:53:20 +03:00
bojanserafimov
7a491f52c4 Add draw_timeline binary (#2688) 2022-10-27 12:53:20 +03:00
Alexander Bayandin
323c4ecb4f Add data format backward compatibility tests (#2626) 2022-10-27 12:53:20 +03:00
Anastasia Lubennikova
3d2466607e Merge pull request #2692 from neondatabase/main-rc
Release 2022-10-25
2022-10-25 18:18:58 +03:00
Anastasia Lubennikova
ed478b39f4 Merge branch 'release' into main-rc 2022-10-25 17:06:33 +03:00
Stas Kelvich
91585a558d Merge pull request #2678 from neondatabase/stas/hotfix_schema
Hotfix to disable grant create on public schema
2022-10-22 02:54:31 +03:00
Stas Kelvich
93467eae1f Hotfix to disable grant create on public schema
`GRANT CREATE ON SCHEMA public` fails if there is no schema `public`.
Disable it in release for now and make a better fix later (it is
needed for v15 support).
2022-10-22 02:26:28 +03:00
Stas Kelvich
f3aac81d19 Merge pull request #2668 from neondatabase/main
Release 2022-10-21
2022-10-21 15:21:42 +03:00
Stas Kelvich
979ad60c19 Merge pull request #2581 from neondatabase/main
Release 2022-10-07
2022-10-07 16:50:55 +03:00
Stas Kelvich
9316cb1b1f Merge pull request #2573 from neondatabase/main
Release 2022-10-06
2022-10-07 11:07:06 +03:00
Anastasia Lubennikova
e7939a527a Merge pull request #2377 from neondatabase/main
Release 2022-09-01
2022-09-01 20:20:44 +03:00
Arthur Petukhovsky
36d26665e1 Merge pull request #2299 from neondatabase/main
* Check for entire range during sasl validation (#2281)

* Gen2 GH runner (#2128)

* Re-add rustup override

* Try s3 bucket

* Set git version

* Use v4 cache key to prevent problems

* Switch to v5 for key

* Add second rustup fix

* Rebase

* Add kaniko steps

* Fix typo and set compress level

* Disable global run default

* Specify shell for step

* Change approach with kaniko

* Try less verbose shell spec

* Add submodule pull

* Add promote step

* Adjust dependency chain

* Try default swap again

* Use env

* Don't override aws key

* Make kaniko build conditional

* Specify runs on

* Try without dependency link

* Try soft fail

* Use image with git

* Try passing to next step

* Fix duplicate

* Try other approach

* Try other approach

* Fix typo

* Try other syntax

* Set env

* Adjust setup

* Try step 1

* Add link

* Try global env

* Fix mistake

* Debug

* Try other syntax

* Try other approach

* Change order

* Move output one step down

* Put output up one level

* Try other syntax

* Skip build

* Try output

* Re-enable build

* Try other syntax

* Skip middle step

* Update check

* Try first step of dockerhub push

* Update needs dependency

* Try explicit dir

* Add missing package

* Try other approach

* Try other approach

* Specify region

* Use with

* Try other approach

* Add debug

* Try other approach

* Set region

* Follow AWS example

* Try github approach

* Skip Qemu

* Try stdin

* Missing steps

* Add missing close

* Add echo debug

* Try v2 endpoint

* Use v1 endpoint

* Try without quotes

* Revert

* Try crane

* Add debug

* Split steps

* Fix duplicate

* Add shell step

* Conform to options

* Add verbose flag

* Try single step

* Try workaround

* First request fails hunch

* Try bullseye image

* Try other approach

* Adjust verbose level

* Try previous step

* Add more debug

* Remove debug step

* Remove rogue indent

* Try with larger image

* Add build tag step

* Update workflow for testing

* Add tag step for test

* Remove unused

* Update dependency chain

* Add ownership fix

* Use matrix for promote

* Force update

* Force build

* Remove unused

* Add new image

* Add missing argument

* Update dockerfile copy

* Update Dockerfile

* Update clone

* Update dockerfile

* Go to correct folder

* Use correct format

* Update dockerfile

* Remove cd

* Debug find where we are

* Add debug on first step

* Changedir to postgres

* Set workdir

* Use v1 approach

* Use other dependency

* Try other approach

* Try other approach

* Update dockerfile

* Update approach

* Update dockerfile

* Update approach

* Update dockerfile

* Update dockerfile

* Add workspace hack

* Update Dockerfile

* Update Dockerfile

* Update Dockerfile

* Change last step

* Cleanup pull in prep for review

* Force build images

* Add condition for latest tagging

* Use pinned version

* Try without name value

* Remove more names

* Shorten names

* Add kaniko comments

* Pin kaniko

* Pin crane and ecr helper

* Up one level

* Switch to pinned tag for rust image

* Force update for test

Co-authored-by: Rory de Zoete <rdezoete@RorysMacStudio.fritz.box>
Co-authored-by: Rory de Zoete <rdezoete@b04468bf-cdf4-41eb-9c94-aff4ca55e4bf.fritz.box>
Co-authored-by: Rory de Zoete <rdezoete@Rorys-Mac-Studio.fritz.box>
Co-authored-by: Rory de Zoete <rdezoete@4795e9ee-4f32-401f-85f3-f316263b62b8.fritz.box>
Co-authored-by: Rory de Zoete <rdezoete@2f8bc4e5-4ec2-4ea2-adb1-65d863c4a558.fritz.box>
Co-authored-by: Rory de Zoete <rdezoete@27565b2b-72d5-4742-9898-a26c9033e6f9.fritz.box>
Co-authored-by: Rory de Zoete <rdezoete@ecc96c26-c6c4-4664-be6e-34f7c3f89a3c.fritz.box>
Co-authored-by: Rory de Zoete <rdezoete@7caff3a5-bf03-4202-bd0e-f1a93c86bdae.fritz.box>

* Add missing step output, revert one deploy step (#2285)

* Add missing step output, revert one deploy step

* Conform to syntax

* Update approach

* Add missing value

* Add missing needs

Co-authored-by: Rory de Zoete <rdezoete@RorysMacStudio.fritz.box>

* Error for fatal not git repo (#2286)

Co-authored-by: Rory de Zoete <rdezoete@RorysMacStudio.fritz.box>

* Use main, not branch for ref check (#2288)

* Use main, not branch for ref check

* Add more debug

* Count main, not head

* Try new approach

* Conform to syntax

* Update approach

* Get full history

* Skip checkout

* Cleanup debug

* Remove more debug

Co-authored-by: Rory de Zoete <rdezoete@RorysMacStudio.fritz.box>

* Fix docker zombie process issue (#2289)

* Fix docker zombie process issue

* Init everywhere

Co-authored-by: Rory de Zoete <rdezoete@RorysMacStudio.fritz.box>

* Fix 1.63 clippy lints (#2282)

* split out timeline metrics, track layer map loading and size calculation

* reset rust cache for clippy run to avoid an ICE

additionally remove trailing whitespaces

* Rename pg_control_ffi.h to bindgen_deps.h, for clarity.

The pg_control_ffi.h name implies that it only includes stuff related to
pg_control.h. That's mostly true currently, but really the point of the
file is to include everything that we need to generate Rust definitions
from.

* Make local mypy behave like CI mypy (#2291)

* Fix flaky pageserver restarts in tests (#2261)

* Remove extra type aliases (#2280)

* Update cachepot endpoint (#2290)

* Update cachepot endpoint

* Update dockerfile & remove env

* Update image building process

* Cannot use metadata endpoint for this

* Update workflow

* Conform to kaniko syntax

* Update syntax

* Update approach

* Update dockerfiles

* Force update

* Update dockerfiles

* Update dockerfile

* Cleanup dockerfiles

* Update s3 test location

* Revert s3 experiment

* Add more debug

* Specify aws region

* Remove debug, add prefix

* Remove one more debug

Co-authored-by: Rory de Zoete <rdezoete@RorysMacStudio.fritz.box>

* workflows/benchmarking: increase timeout (#2294)

* Rework `init` in pageserver CLI  (#2272)

* Do not create initial tenant and timeline (adjust Python tests for that)
* Rework config handling during init, add --update-config to manage local config updates

* Fix: Always build images (#2296)

* Always build images

* Remove unused

Co-authored-by: Rory de Zoete <rdezoete@RorysMacStudio.fritz.box>

* Move auto-generated 'bindings' to a separate inner module.

Re-export only things that are used by other modules.

In the future, I'm imagining that we run bindgen twice, for Postgres
v14 and v15. The two sets of bindings would go into separate
'bindings_v14' and 'bindings_v15' modules.

Rearrange postgres_ffi modules.

Move function, to avoid Postgres version dependency in timelines.rs
Move function to generate a logical-message WAL record to postgres_ffi.

* fix cargo test

* Fix walreceiver and safekeeper bugs (#2295)

- There was an issue with zero commit_lsn `reason: LaggingWal { current_commit_lsn: 0/0, new_commit_lsn: 1/6FD90D38, threshold: 10485760 } }`. The problem was in `send_wal.rs`, where we initialized `end_pos = Lsn(0)` and in some cases sent it to the pageserver.
- IDENTIFY_SYSTEM previously returned `flush_lsn` as a physical end of WAL. Now it returns `flush_lsn` (as it was) to walproposer and `commit_lsn` to everyone else including pageserver.
- There was an issue with backoff where connection was cancelled right after initialization: `connected!` -> `safekeeper_handle_db: Connection cancelled` -> `Backoff: waiting 3 seconds`. The problem was in sleeping before establishing the connection. This is fixed by reworking retry logic.
- There was an issue with getting `NoKeepAlives` reason in a loop. The issue is probably the same as the previous.
- There was an issue with filtering safekeepers based on retry attempts, which could filter some safekeepers indefinetely. This is fixed by using retry cooldown duration instead of retry attempts.
- Some `send_wal.rs` connections failed with errors without context. This is fixed by adding a timeline to safekeepers errors.

New retry logic works like this:
- Every candidate has a `next_retry_at` timestamp and is not considered for connection until that moment
- When walreceiver connection is closed, we update `next_retry_at` using exponential backoff, increasing the cooldown on every disconnect.
- When `last_record_lsn` was advanced using the WAL from the safekeeper, we reset the retry cooldown and exponential backoff, allowing walreceiver to reconnect to the same safekeeper instantly.

* on safekeeper registration pass availability zone param (#2292)

Co-authored-by: Kirill Bulatov <kirill@neon.tech>
Co-authored-by: Rory de Zoete <33318916+zoete@users.noreply.github.com>
Co-authored-by: Rory de Zoete <rdezoete@RorysMacStudio.fritz.box>
Co-authored-by: Rory de Zoete <rdezoete@b04468bf-cdf4-41eb-9c94-aff4ca55e4bf.fritz.box>
Co-authored-by: Rory de Zoete <rdezoete@Rorys-Mac-Studio.fritz.box>
Co-authored-by: Rory de Zoete <rdezoete@4795e9ee-4f32-401f-85f3-f316263b62b8.fritz.box>
Co-authored-by: Rory de Zoete <rdezoete@2f8bc4e5-4ec2-4ea2-adb1-65d863c4a558.fritz.box>
Co-authored-by: Rory de Zoete <rdezoete@27565b2b-72d5-4742-9898-a26c9033e6f9.fritz.box>
Co-authored-by: Rory de Zoete <rdezoete@ecc96c26-c6c4-4664-be6e-34f7c3f89a3c.fritz.box>
Co-authored-by: Rory de Zoete <rdezoete@7caff3a5-bf03-4202-bd0e-f1a93c86bdae.fritz.box>
Co-authored-by: Dmitry Rodionov <dmitry@neon.tech>
Co-authored-by: Heikki Linnakangas <heikki@neon.tech>
Co-authored-by: bojanserafimov <bojan.serafimov7@gmail.com>
Co-authored-by: Alexander Bayandin <alexander@neon.tech>
Co-authored-by: Anastasia Lubennikova <anastasia@neon.tech>
Co-authored-by: Anton Galitsyn <agalitsyn@users.noreply.github.com>
2022-08-18 15:32:33 +03:00
Arthur Petukhovsky
873347f977 Merge pull request #2275 from neondatabase/main
* github/workflows: Fix git dubious ownership (#2223)

* Move relation size cache from WalIngest to DatadirTimeline (#2094)

* Move relation sie cache to layered timeline

* Fix obtaining current LSN for relation size cache

* Resolve merge conflicts

* Resolve merge conflicts

* Reestore 'lsn' field in DatadirModification

* adjust DatadirModification lsn in ingest_record

* Fix formatting

* Pass lsn to get_relsize

* Fix merge conflict

* Update pageserver/src/pgdatadir_mapping.rs

Co-authored-by: Heikki Linnakangas <heikki@zenith.tech>

* Update pageserver/src/pgdatadir_mapping.rs

Co-authored-by: Heikki Linnakangas <heikki@zenith.tech>

Co-authored-by: Heikki Linnakangas <heikki@zenith.tech>

* refactor: replace lazy-static with once-cell (#2195)

- Replacing all the occurrences of lazy-static with `once-cell::sync::Lazy`
- fixes #1147

Signed-off-by: Ankur Srivastava <best.ankur@gmail.com>

* Add more buckets to pageserver latency metrics (#2225)

* ignore record property warning to fix benchmarks

* increase statement timeout

* use event so it fires only if workload thread successfully finished

* remove debug log

* increase timeout to pass test with real s3

* avoid duplicate parameter, increase timeout

* Major migration script (#2073)

This script can be used to migrate a tenant across breaking storage versions, or (in the future) upgrading postgres versions. See the comment at the top for an overview.

Co-authored-by: Anastasia Lubennikova <anastasia@neon.tech>

* Fix etcd typos

* Fix links to safekeeper protocol docs. (#2188)

safekeeper/README_PROTO.md was moved to docs/safekeeper-protocol.md in
commit 0b14fdb078, as part of reorganizing the docs into 'mdbook' format.

Fixes issue #1475. Thanks to @banks for spotting the outdated references.

In addition to fixing the above issue, this patch also fixes other broken links as a result of 0b14fdb078. See https://github.com/neondatabase/neon/pull/2188#pullrequestreview-1055918480.

Co-authored-by: Heikki Linnakangas <heikki@neon.tech>
Co-authored-by: Thang Pham <thang@neon.tech>

* Update CONTRIBUTING.md

* Update CONTRIBUTING.md

* support node id and remote storage params in docker_entrypoint.sh

* Safe truncate (#2218)

* Move relation sie cache to layered timeline

* Fix obtaining current LSN for relation size cache

* Resolve merge conflicts

* Resolve merge conflicts

* Reestore 'lsn' field in DatadirModification

* adjust DatadirModification lsn in ingest_record

* Fix formatting

* Pass lsn to get_relsize

* Fix merge conflict

* Update pageserver/src/pgdatadir_mapping.rs

Co-authored-by: Heikki Linnakangas <heikki@zenith.tech>

* Update pageserver/src/pgdatadir_mapping.rs

Co-authored-by: Heikki Linnakangas <heikki@zenith.tech>

* Check if relation exists before trying to truncat it

refer #1932

* Add test reporducing FSM truncate problem

Co-authored-by: Heikki Linnakangas <heikki@zenith.tech>

* Fix exponential backoff values

* Update back `vendor/postgres` back; it was changed accidentally. (#2251)

Commit 4227cfc96e accidentally reverted vendor/postgres to an older
version. Update it back.

* Add pageserver checkpoint_timeout option.

To flush inmemory layer eventually when no new data arrives, which helps
safekeepers to suspend activity (stop pushing to the broker). Default 10m should
be ok.

* Share exponential backoff code and fix logic for delete task failure (#2252)

* Fix bug when import large (>1GB) relations (#2172)

Resolves #2097 

- use timeline modification's `lsn` and timeline's `last_record_lsn` to determine the corresponding LSN to query data in `DatadirModification::get`
- update `test_import_from_pageserver`. Split the test into 2 variants: `small` and `multisegment`. 
  + `small` is the old test
  + `multisegment` is to simulate #2097 by using a larger number of inserted rows to create multiple segment files of a relation. `multisegment` is configured to only run with a `release` build

* Fix timeline physical size flaky tests (#2244)

Resolves #2212.

- use `wait_for_last_flush_lsn` in `test_timeline_physical_size_*` tests

## Context
Need to wait for the pageserver to catch up with the compute's last flush LSN because during the timeline physical size API call, it's possible that there are running `LayerFlushThread` threads. These threads flush new layers into disk and hence update the physical size. This results in a mismatch between the physical size reported by the API and the actual physical size on disk.

### Note
The `LayerFlushThread` threads are processed **concurrently**, so it's possible that the above error still persists even with this patch. However, making the tests wait to finish processing all the WALs (not flushing) before calculating the physical size should help reduce the "flakiness" significantly

* postgres_ffi/waldecoder: validate more header fields

* postgres_ffi/waldecoder: remove unused startlsn

* postgres_ffi/waldecoder: introduce explicit `enum State`

Previously it was emulated with a combination of nullable fields.
This change should make the logic more readable.

* disable `test_import_from_pageserver_multisegment` (#2258)

This test failed consistently on `main` now. It's better to temporarily disable it to avoid blocking others' PRs while investigating the root cause for the test failure.

See: #2255, #2256

* get_binaries uses DOCKER_TAG taken from docker image build step (#2260)

* [proxy] Rework wire format of the password hack and some errors (#2236)

The new format has a few benefits: it's shorter, simpler and
human-readable as well. We don't use base64 anymore, since
url encoding got us covered.

We also show a better error in case we couldn't parse the
payload; the users should know it's all about passing the
correct project name.

* test_runner/pg_clients: collect docker logs (#2259)

* get_binaries script fix (#2263)

* get_binaries uses DOCKER_TAG taken from docker image build step

* remove docker tag discovery at all and fix get_binaries for version variable

* Better storage sync logs (#2268)

* Find end of WAL on safekeepers using WalStreamDecoder.

We could make it inside wal_storage.rs, but taking into account that
 - wal_storage.rs reading is async
 - we don't need s3 here
 - error handling is different; error during decoding is normal
I decided to put it separately.

Test
cargo test test_find_end_of_wal_last_crossing_segment
prepared earlier by @yeputons passes now.

Fixes https://github.com/neondatabase/neon/issues/544
      https://github.com/neondatabase/cloud/issues/2004
Supersedes https://github.com/neondatabase/neon/pull/2066

* Improve walreceiver logic (#2253)

This patch makes walreceiver logic more complicated, but it should work better in most cases. Added `test_wal_lagging` to test scenarios where alive safekeepers can lag behind other alive safekeepers.

- There was a bug which looks like `etcd_info.timeline.commit_lsn > Some(self.local_timeline.get_last_record_lsn())` filtered all safekeepers in some strange cases. I removed this filter, it should probably help with #2237
- Now walreceiver_connection reports status, including commit_lsn. This allows keeping safekeeper connection even when etcd is down.
- Safekeeper connection now fails if pageserver doesn't receive safekeeper messages for some time. Usually safekeeper sends messages at least once per second.
- `LaggingWal` check now uses `commit_lsn` directly from safekeeper. This fixes the issue with often reconnects, when compute generates WAL really fast.
- `NoWalTimeout` is rewritten to trigger only when we know about the new WAL and the connected safekeeper doesn't stream any WAL. This allows setting a small `lagging_wal_timeout` because it will trigger only when we observe that the connected safekeeper has stuck.

* increase timeout in wait_for_upload to avoid spurious failures when testing with real s3

* Bump vendor/postgres to include XLP_FIRST_IS_CONTRECORD fix. (#2274)

* Set up a workflow to run pgbench against captest (#2077)

Signed-off-by: Ankur Srivastava <best.ankur@gmail.com>
Co-authored-by: Alexander Bayandin <alexander@neon.tech>
Co-authored-by: Konstantin Knizhnik <knizhnik@garret.ru>
Co-authored-by: Heikki Linnakangas <heikki@zenith.tech>
Co-authored-by: Ankur Srivastava <ansrivas@users.noreply.github.com>
Co-authored-by: bojanserafimov <bojan.serafimov7@gmail.com>
Co-authored-by: Dmitry Rodionov <dmitry@neon.tech>
Co-authored-by: Anastasia Lubennikova <anastasia@neon.tech>
Co-authored-by: Kirill Bulatov <kirill@neon.tech>
Co-authored-by: Heikki Linnakangas <heikki@neon.tech>
Co-authored-by: Thang Pham <thang@neon.tech>
Co-authored-by: Stas Kelvich <stas.kelvich@gmail.com>
Co-authored-by: Arseny Sher <sher-ars@yandex.ru>
Co-authored-by: Egor Suvorov <egor@neon.tech>
Co-authored-by: Andrey Taranik <andrey@cicd.team>
Co-authored-by: Dmitry Ivanov <ivadmi5@gmail.com>
2022-08-15 21:30:45 +03:00
Arthur Petukhovsky
e814ac16f9 Merge pull request #2219 from neondatabase/main
Release 2022-08-04
2022-08-04 20:06:34 +03:00
Heikki Linnakangas
ad3055d386 Merge pull request #2203 from neondatabase/release-uuid-ossp
Deploy new storage and compute version to production

Release 2022-08-02
2022-08-02 15:08:14 +03:00
Heikki Linnakangas
94e03eb452 Merge remote-tracking branch 'origin/main' into 'release'
Release 2022-08-01
2022-08-02 12:43:49 +03:00
Sergey Melnikov
380f26ef79 Merge pull request #2170 from neondatabase/main (Release 2022-07-28)
Release 2022-07-28
2022-07-28 14:16:52 +03:00
Arthur Petukhovsky
3c5b7f59d7 Merge pull request #2119 from neondatabase/main
Release 2022-07-19
2022-07-19 11:58:48 +03:00
Arthur Petukhovsky
fee89f80b5 Merge pull request #2115 from neondatabase/main-2022-07-18
Release 2022-07-18
2022-07-18 19:21:11 +03:00
Arthur Petukhovsky
41cce8eaf1 Merge remote-tracking branch 'origin/release' into main-2022-07-18 2022-07-18 18:21:20 +03:00
Alexey Kondratov
f88fe0218d Merge pull request #1842 from neondatabase/release-deploy-hotfix
[HOTFIX] Release deploy fix

This PR uses this branch neondatabase/postgres#171 and several required commits from the main to use only locally built compute-tools. This should allow us to rollout safekeepers sync issue fix on prod
2022-06-01 11:04:30 +03:00
Alexey Kondratov
cc856eca85 Install missing openssl packages in the Github Actions workflow 2022-05-31 21:31:31 +02:00
Alexey Kondratov
cf350c6002 Use :local compute-tools tag to build compute-node image 2022-05-31 21:31:16 +02:00
Arseny Sher
0ce6b6a0a3 Merge pull request #1836 from neondatabase/release-hotfix-basebackup-lsn-page-boundary
Bump vendor/postgres to hotfix basebackup LSN comparison.
2022-05-31 16:54:03 +04:00
Arseny Sher
73f247d537 Bump vendor/postgres to hotfix basebackup LSN comparison. 2022-05-31 16:00:50 +04:00
Andrey Taranik
960be82183 Merge pull request #1792 from neondatabase/main
Release 2202-05-25 (second)
2022-05-25 16:37:57 +03:00
Andrey Taranik
806e5a6c19 Merge pull request #1787 from neondatabase/main
Release 2022-05-25
2022-05-25 13:34:11 +03:00
Alexey Kondratov
8d5df07cce Merge pull request #1385 from zenithdb/main
Release main 2022-03-22
2022-03-22 05:04:34 -05:00
Andrey Taranik
df7a9d1407 release fix 2022-03-16 (#1375) 2022-03-17 00:43:28 +03:00
302 changed files with 3808 additions and 21064 deletions

View File

@@ -46,9 +46,6 @@ workspace-members = [
"utils",
"wal_craft",
"walproposer",
"postgres-protocol2",
"postgres-types2",
"tokio-postgres2",
]
# Write out exact versions rather than a semver range. (Defaults to false.)

View File

@@ -43,8 +43,7 @@ runs:
PR_NUMBER=$(jq --raw-output .pull_request.number "$GITHUB_EVENT_PATH" || true)
if [ "${PR_NUMBER}" != "null" ]; then
BRANCH_OR_PR=pr-${PR_NUMBER}
elif [ "${GITHUB_REF_NAME}" = "main" ] || [ "${GITHUB_REF_NAME}" = "release" ] || \
[ "${GITHUB_REF_NAME}" = "release-proxy" ] || [ "${GITHUB_REF_NAME}" = "release-compute" ]; then
elif [ "${GITHUB_REF_NAME}" = "main" ] || [ "${GITHUB_REF_NAME}" = "release" ] || [ "${GITHUB_REF_NAME}" = "release-proxy" ]; then
# Shortcut for special branches
BRANCH_OR_PR=${GITHUB_REF_NAME}
else

View File

@@ -23,8 +23,7 @@ runs:
PR_NUMBER=$(jq --raw-output .pull_request.number "$GITHUB_EVENT_PATH" || true)
if [ "${PR_NUMBER}" != "null" ]; then
BRANCH_OR_PR=pr-${PR_NUMBER}
elif [ "${GITHUB_REF_NAME}" = "main" ] || [ "${GITHUB_REF_NAME}" = "release" ] || \
[ "${GITHUB_REF_NAME}" = "release-proxy" ] || [ "${GITHUB_REF_NAME}" = "release-compute" ]; then
elif [ "${GITHUB_REF_NAME}" = "main" ] || [ "${GITHUB_REF_NAME}" = "release" ] || [ "${GITHUB_REF_NAME}" = "release-proxy" ]; then
# Shortcut for special branches
BRANCH_OR_PR=${GITHUB_REF_NAME}
else

View File

@@ -36,8 +36,8 @@ inputs:
description: 'Region name for real s3 tests'
required: false
default: ''
rerun_failed:
description: 'Whether to rerun failed tests'
rerun_flaky:
description: 'Whether to rerun flaky tests'
required: false
default: 'false'
pg_version:
@@ -108,7 +108,7 @@ runs:
COMPATIBILITY_SNAPSHOT_DIR: /tmp/compatibility_snapshot_pg${{ inputs.pg_version }}
ALLOW_BACKWARD_COMPATIBILITY_BREAKAGE: contains(github.event.pull_request.labels.*.name, 'backward compatibility breakage')
ALLOW_FORWARD_COMPATIBILITY_BREAKAGE: contains(github.event.pull_request.labels.*.name, 'forward compatibility breakage')
RERUN_FAILED: ${{ inputs.rerun_failed }}
RERUN_FLAKY: ${{ inputs.rerun_flaky }}
PG_VERSION: ${{ inputs.pg_version }}
shell: bash -euxo pipefail {0}
run: |
@@ -154,8 +154,15 @@ runs:
EXTRA_PARAMS="--out-dir $PERF_REPORT_DIR $EXTRA_PARAMS"
fi
if [ "${RERUN_FAILED}" == "true" ]; then
EXTRA_PARAMS="--reruns 2 $EXTRA_PARAMS"
if [ "${RERUN_FLAKY}" == "true" ]; then
mkdir -p $TEST_OUTPUT
poetry run ./scripts/flaky_tests.py "${TEST_RESULT_CONNSTR}" \
--days 7 \
--output "$TEST_OUTPUT/flaky.json" \
--pg-version "${DEFAULT_PG_VERSION}" \
--build-type "${BUILD_TYPE}"
EXTRA_PARAMS="--flaky-tests-json $TEST_OUTPUT/flaky.json $EXTRA_PARAMS"
fi
# We use pytest-split plugin to run benchmarks in parallel on different CI runners

View File

@@ -293,7 +293,7 @@ jobs:
run_with_real_s3: true
real_s3_bucket: neon-github-ci-tests
real_s3_region: eu-central-1
rerun_failed: true
rerun_flaky: true
pg_version: ${{ matrix.pg_version }}
env:
TEST_RESULT_CONNSTR: ${{ secrets.REGRESS_TEST_RESULT_CONNSTR_NEW }}

View File

@@ -21,7 +21,7 @@ defaults:
shell: bash -euo pipefail {0}
jobs:
create-release-branch:
create-storage-release-branch:
runs-on: ubuntu-22.04
permissions:

View File

@@ -249,7 +249,7 @@ jobs:
# Post both success and failure to the Slack channel
- name: Post to a Slack channel
if: ${{ github.event.schedule && !cancelled() }}
if: ${{ github.event.schedule }}
uses: slackapi/slack-github-action@v1
with:
channel-id: "C06T9AMNDQQ" # on-call-compute-staging-stream

View File

@@ -6,7 +6,6 @@ on:
- main
- release
- release-proxy
- release-compute
pull_request:
defaults:
@@ -71,10 +70,8 @@ jobs:
echo "tag=release-$(git rev-list --count HEAD)" >> $GITHUB_OUTPUT
elif [[ "$GITHUB_REF_NAME" == "release-proxy" ]]; then
echo "tag=release-proxy-$(git rev-list --count HEAD)" >> $GITHUB_OUTPUT
elif [[ "$GITHUB_REF_NAME" == "release-compute" ]]; then
echo "tag=release-compute-$(git rev-list --count HEAD)" >> $GITHUB_OUTPUT
else
echo "GITHUB_REF_NAME (value '$GITHUB_REF_NAME') is not set to either 'main' or 'release', 'release-proxy', 'release-compute'"
echo "GITHUB_REF_NAME (value '$GITHUB_REF_NAME') is not set to either 'main' or 'release'"
echo "tag=$GITHUB_RUN_ID" >> $GITHUB_OUTPUT
fi
shell: bash
@@ -516,7 +513,7 @@ jobs:
})
trigger-e2e-tests:
if: ${{ !github.event.pull_request.draft || contains( github.event.pull_request.labels.*.name, 'run-e2e-tests-in-draft') || github.ref_name == 'main' || github.ref_name == 'release' || github.ref_name == 'release-proxy' || github.ref_name == 'release-compute' }}
if: ${{ !github.event.pull_request.draft || contains( github.event.pull_request.labels.*.name, 'run-e2e-tests-in-draft') || github.ref_name == 'main' || github.ref_name == 'release' || github.ref_name == 'release-proxy' }}
needs: [ check-permissions, promote-images, tag ]
uses: ./.github/workflows/trigger-e2e-tests.yml
secrets: inherit
@@ -672,7 +669,7 @@ jobs:
neondatabase/compute-node-${{ matrix.version.pg }}:${{ needs.tag.outputs.build-tag }}-${{ matrix.version.debian }}-${{ matrix.arch }}
- name: Build neon extensions test image
if: matrix.version.pg >= 'v16'
if: matrix.version.pg == 'v16'
uses: docker/build-push-action@v6
with:
context: .
@@ -687,7 +684,8 @@ jobs:
pull: true
file: compute/compute-node.Dockerfile
target: neon-pg-ext-test
cache-from: type=registry,ref=cache.neon.build/compute-node-${{ matrix.version.pg }}:cache-${{ matrix.version.debian }}-${{ matrix.arch }}
cache-from: type=registry,ref=cache.neon.build/neon-test-extensions-${{ matrix.version.pg }}:cache-${{ matrix.version.debian }}-${{ matrix.arch }}
cache-to: ${{ github.ref_name == 'main' && format('type=registry,ref=cache.neon.build/neon-test-extensions-{0}:cache-{1}-{2},mode=max', matrix.version.pg, matrix.version.debian, matrix.arch) || '' }}
tags: |
neondatabase/neon-test-extensions-${{ matrix.version.pg }}:${{needs.tag.outputs.build-tag}}-${{ matrix.version.debian }}-${{ matrix.arch }}
@@ -710,7 +708,7 @@ jobs:
push: true
pull: true
file: compute/compute-node.Dockerfile
cache-from: type=registry,ref=cache.neon.build/compute-node-${{ matrix.version.pg }}:cache-${{ matrix.version.debian }}-${{ matrix.arch }}
cache-from: type=registry,ref=cache.neon.build/neon-test-extensions-${{ matrix.version.pg }}:cache-${{ matrix.version.debian }}-${{ matrix.arch }}
cache-to: ${{ github.ref_name == 'main' && format('type=registry,ref=cache.neon.build/compute-tools-{0}:cache-{1}-{2},mode=max', matrix.version.pg, matrix.version.debian, matrix.arch) || '' }}
tags: |
neondatabase/compute-tools:${{ needs.tag.outputs.build-tag }}-${{ matrix.version.debian }}-${{ matrix.arch }}
@@ -746,7 +744,7 @@ jobs:
neondatabase/compute-node-${{ matrix.version.pg }}:${{ needs.tag.outputs.build-tag }}-${{ matrix.version.debian }}-arm64
- name: Create multi-arch neon-test-extensions image
if: matrix.version.pg >= 'v16'
if: matrix.version.pg == 'v16'
run: |
docker buildx imagetools create -t neondatabase/neon-test-extensions-${{ matrix.version.pg }}:${{ needs.tag.outputs.build-tag }} \
-t neondatabase/neon-test-extensions-${{ matrix.version.pg }}:${{ needs.tag.outputs.build-tag }}-${{ matrix.version.debian }} \
@@ -835,7 +833,6 @@ jobs:
fail-fast: false
matrix:
arch: [ x64, arm64 ]
pg_version: [v16, v17]
runs-on: ${{ fromJson(format('["self-hosted", "{0}"]', matrix.arch == 'arm64' && 'small-arm64' || 'small')) }}
@@ -874,10 +871,7 @@ jobs:
- name: Verify docker-compose example and test extensions
timeout-minutes: 20
env:
TAG: ${{needs.tag.outputs.build-tag}}
TEST_VERSION_ONLY: ${{ matrix.pg_version }}
run: ./docker-compose/docker_compose_test.sh
run: env TAG=${{needs.tag.outputs.build-tag}} ./docker-compose/docker_compose_test.sh
- name: Print logs and clean up
if: always()
@@ -937,7 +931,7 @@ jobs:
neondatabase/neon-test-extensions-v16:${{ needs.tag.outputs.build-tag }}
- name: Configure AWS-prod credentials
if: github.ref_name == 'release'|| github.ref_name == 'release-proxy' || github.ref_name == 'release-compute'
if: github.ref_name == 'release'|| github.ref_name == 'release-proxy'
uses: aws-actions/configure-aws-credentials@v4
with:
aws-region: eu-central-1
@@ -946,12 +940,12 @@ jobs:
- name: Login to prod ECR
uses: docker/login-action@v3
if: github.ref_name == 'release'|| github.ref_name == 'release-proxy' || github.ref_name == 'release-compute'
if: github.ref_name == 'release'|| github.ref_name == 'release-proxy'
with:
registry: 093970136003.dkr.ecr.eu-central-1.amazonaws.com
- name: Copy all images to prod ECR
if: github.ref_name == 'release' || github.ref_name == 'release-proxy' || github.ref_name == 'release-compute'
if: github.ref_name == 'release'|| github.ref_name == 'release-proxy'
run: |
for image in neon compute-tools {vm-,}compute-node-{v14,v15,v16,v17}; do
docker buildx imagetools create -t 093970136003.dkr.ecr.eu-central-1.amazonaws.com/${image}:${{ needs.tag.outputs.build-tag }} \
@@ -971,7 +965,7 @@ jobs:
tenant_id: ${{ vars.AZURE_TENANT_ID }}
push-to-acr-prod:
if: github.ref_name == 'release' || github.ref_name == 'release-proxy' || github.ref_name == 'release-compute'
if: github.ref_name == 'release'|| github.ref_name == 'release-proxy'
needs: [ tag, promote-images ]
uses: ./.github/workflows/_push-to-acr.yml
with:
@@ -1059,7 +1053,7 @@ jobs:
deploy:
needs: [ check-permissions, promote-images, tag, build-and-test-locally, trigger-custom-extensions-build-and-wait, push-to-acr-dev, push-to-acr-prod ]
# `!failure() && !cancelled()` is required because the workflow depends on the job that can be skipped: `push-to-acr-dev` and `push-to-acr-prod`
if: (github.ref_name == 'main' || github.ref_name == 'release' || github.ref_name == 'release-proxy' || github.ref_name == 'release-compute') && !failure() && !cancelled()
if: (github.ref_name == 'main' || github.ref_name == 'release' || github.ref_name == 'release-proxy') && !failure() && !cancelled()
runs-on: [ self-hosted, small ]
container: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/ansible:latest
@@ -1108,15 +1102,13 @@ jobs:
-f deployProxyAuthBroker=true \
-f branch=main \
-f dockerTag=${{needs.tag.outputs.build-tag}}
elif [[ "$GITHUB_REF_NAME" == "release-compute" ]]; then
gh workflow --repo neondatabase/infra run deploy-compute-dev.yml --ref main -f dockerTag=${{needs.tag.outputs.build-tag}}
else
echo "GITHUB_REF_NAME (value '$GITHUB_REF_NAME') is not set to either 'main', 'release', 'release-proxy' or 'release-compute'"
echo "GITHUB_REF_NAME (value '$GITHUB_REF_NAME') is not set to either 'main' or 'release'"
exit 1
fi
- name: Create git tag
if: github.ref_name == 'release' || github.ref_name == 'release-proxy' || github.ref_name == 'release-compute'
if: github.ref_name == 'release' || github.ref_name == 'release-proxy'
uses: actions/github-script@v7
with:
# Retry script for 5XX server errors: https://github.com/actions/github-script#retries

View File

@@ -26,7 +26,6 @@ concurrency:
jobs:
ingest:
strategy:
fail-fast: false # allow other variants to continue even if one fails
matrix:
target_project: [new_empty_project, large_existing_project]
permissions:

View File

@@ -15,10 +15,6 @@ on:
type: boolean
description: 'Create Proxy release PR'
required: false
create-compute-release-branch:
type: boolean
description: 'Create Compute release PR'
required: false
# No permission for GITHUB_TOKEN by default; the **minimal required** set of permissions should be granted in each job.
permissions: {}
@@ -29,20 +25,20 @@ defaults:
jobs:
create-storage-release-branch:
if: ${{ github.event.schedule == '0 6 * * MON' || inputs.create-storage-release-branch }}
if: ${{ github.event.schedule == '0 6 * * MON' || format('{0}', inputs.create-storage-release-branch) == 'true' }}
permissions:
contents: write
uses: ./.github/workflows/_create-release-pr.yml
with:
component-name: 'Storage'
component-name: 'Storage & Compute'
release-branch: 'release'
secrets:
ci-access-token: ${{ secrets.CI_ACCESS_TOKEN }}
create-proxy-release-branch:
if: ${{ github.event.schedule == '0 6 * * THU' || inputs.create-proxy-release-branch }}
if: ${{ github.event.schedule == '0 6 * * THU' || format('{0}', inputs.create-proxy-release-branch) == 'true' }}
permissions:
contents: write
@@ -53,16 +49,3 @@ jobs:
release-branch: 'release-proxy'
secrets:
ci-access-token: ${{ secrets.CI_ACCESS_TOKEN }}
create-compute-release-branch:
if: inputs.create-compute-release-branch
permissions:
contents: write
uses: ./.github/workflows/_create-release-pr.yml
with:
component-name: 'Compute'
release-branch: 'release-compute'
secrets:
ci-access-token: ${{ secrets.CI_ACCESS_TOKEN }}

View File

@@ -51,8 +51,6 @@ jobs:
echo "tag=release-$(git rev-list --count HEAD)" | tee -a $GITHUB_OUTPUT
elif [[ "$GITHUB_REF_NAME" == "release-proxy" ]]; then
echo "tag=release-proxy-$(git rev-list --count HEAD)" >> $GITHUB_OUTPUT
elif [[ "$GITHUB_REF_NAME" == "release-compute" ]]; then
echo "tag=release-compute-$(git rev-list --count HEAD)" >> $GITHUB_OUTPUT
else
echo "GITHUB_REF_NAME (value '$GITHUB_REF_NAME') is not set to either 'main' or 'release'"
BUILD_AND_TEST_RUN_ID=$(gh run list -b $CURRENT_BRANCH -c $CURRENT_SHA -w 'Build and Test' -L 1 --json databaseId --jq '.[].databaseId')

View File

@@ -2,7 +2,6 @@
/compute_tools/ @neondatabase/control-plane @neondatabase/compute
/libs/pageserver_api/ @neondatabase/storage
/libs/postgres_ffi/ @neondatabase/compute @neondatabase/storage
/libs/proxy/ @neondatabase/proxy
/libs/remote_storage/ @neondatabase/storage
/libs/safekeeper_api/ @neondatabase/storage
/libs/vm_monitor/ @neondatabase/autoscaling

390
Cargo.lock generated

File diff suppressed because it is too large Load Diff

View File

@@ -35,9 +35,6 @@ members = [
"libs/walproposer",
"libs/wal_decoder",
"libs/postgres_initdb",
"libs/proxy/postgres-protocol2",
"libs/proxy/postgres-types2",
"libs/proxy/tokio-postgres2",
]
[workspace.package]
@@ -74,7 +71,7 @@ bindgen = "0.70"
bit_field = "0.10.2"
bstr = "1.0"
byteorder = "1.4"
bytes = "1.9"
bytes = "1.0"
camino = "1.1.6"
cfg-if = "1.0.0"
chrono = { version = "0.4", default-features = false, features = ["clock"] }
@@ -83,7 +80,6 @@ comfy-table = "7.1"
const_format = "0.2"
crc32c = "0.6"
dashmap = { version = "5.5.0", features = ["raw-api"] }
diatomic-waker = { version = "0.2.3" }
either = "1.8"
enum-map = "2.4.2"
enumset = "1.0.12"
@@ -115,7 +111,6 @@ indoc = "2"
ipnet = "2.10.0"
itertools = "0.10"
itoa = "1.0.11"
jemalloc_pprof = "0.6"
jsonwebtoken = "9"
lasso = "0.7"
libc = "0.2"
@@ -128,10 +123,10 @@ notify = "6.0.0"
num_cpus = "1.15"
num-traits = "0.2.15"
once_cell = "1.13"
opentelemetry = "0.26"
opentelemetry_sdk = "0.26"
opentelemetry-otlp = { version = "0.26", default-features=false, features = ["http-proto", "trace", "http", "reqwest-client"] }
opentelemetry-semantic-conventions = "0.26"
opentelemetry = "0.24"
opentelemetry_sdk = "0.24"
opentelemetry-otlp = { version = "0.17", default-features=false, features = ["http-proto", "trace", "http", "reqwest-client"] }
opentelemetry-semantic-conventions = "0.16"
parking_lot = "0.12"
parquet = { version = "53", default-features = false, features = ["zstd"] }
parquet_derive = "53"
@@ -145,9 +140,9 @@ rand = "0.8"
redis = { version = "0.25.2", features = ["tokio-rustls-comp", "keep-alive"] }
regex = "1.10.2"
reqwest = { version = "0.12", default-features = false, features = ["rustls-tls"] }
reqwest-tracing = { version = "0.5", features = ["opentelemetry_0_26"] }
reqwest-middleware = "0.4"
reqwest-retry = "0.7"
reqwest-tracing = { version = "0.5", features = ["opentelemetry_0_24"] }
reqwest-middleware = "0.3.0"
reqwest-retry = "0.5"
routerify = "3"
rpds = "0.13"
rustc-hash = "1.1.0"
@@ -176,7 +171,7 @@ sync_wrapper = "0.1.2"
tar = "0.4"
test-context = "0.3"
thiserror = "1.0"
tikv-jemallocator = { version = "0.6", features = ["profiling", "stats", "unprefixed_malloc_on_supported_platforms"] }
tikv-jemallocator = { version = "0.6", features = ["stats"] }
tikv-jemalloc-ctl = { version = "0.6", features = ["stats"] }
tokio = { version = "1.17", features = ["macros"] }
tokio-epoll-uring = { git = "https://github.com/neondatabase/tokio-epoll-uring.git" , branch = "main" }
@@ -192,7 +187,7 @@ tonic = {version = "0.12.3", features = ["tls", "tls-roots"]}
tower-service = "0.3.2"
tracing = "0.1"
tracing-error = "0.2"
tracing-opentelemetry = "0.27"
tracing-opentelemetry = "0.25"
tracing-subscriber = { version = "0.3", default-features = false, features = ["smallvec", "fmt", "tracing-log", "std", "env-filter", "json"] }
try-lock = "0.2.5"
twox-hash = { version = "1.6.3", default-features = false }

View File

@@ -147,8 +147,6 @@ postgres-%: postgres-configure-% \
$(MAKE) -C $(POSTGRES_INSTALL_DIR)/build/$*/contrib/pg_prewarm install
+@echo "Compiling pg_buffercache $*"
$(MAKE) -C $(POSTGRES_INSTALL_DIR)/build/$*/contrib/pg_buffercache install
+@echo "Compiling pg_visibility $*"
$(MAKE) -C $(POSTGRES_INSTALL_DIR)/build/$*/contrib/pg_visibility install
+@echo "Compiling pageinspect $*"
$(MAKE) -C $(POSTGRES_INSTALL_DIR)/build/$*/contrib/pageinspect install
+@echo "Compiling amcheck $*"

View File

@@ -57,9 +57,9 @@ RUN mkdir -p /pgcopydb/bin && \
mkdir -p /pgcopydb/lib && \
chmod -R 755 /pgcopydb && \
chown -R nonroot:nonroot /pgcopydb
COPY --from=pgcopydb_builder /usr/lib/postgresql/16/bin/pgcopydb /pgcopydb/bin/pgcopydb
COPY --from=pgcopydb_builder /pgcopydb/lib/libpq.so.5 /pgcopydb/lib/libpq.so.5
COPY --from=pgcopydb_builder /usr/lib/postgresql/16/bin/pgcopydb /pgcopydb/bin/pgcopydb
COPY --from=pgcopydb_builder /pgcopydb/lib/libpq.so.5 /pgcopydb/lib/libpq.so.5
# System deps
#
@@ -258,14 +258,14 @@ WORKDIR /home/nonroot
# Rust
# Please keep the version of llvm (installed above) in sync with rust llvm (`rustc --version --verbose | grep LLVM`)
ENV RUSTC_VERSION=1.83.0
ENV RUSTC_VERSION=1.82.0
ENV RUSTUP_HOME="/home/nonroot/.rustup"
ENV PATH="/home/nonroot/.cargo/bin:${PATH}"
ARG RUSTFILT_VERSION=0.2.1
ARG CARGO_HAKARI_VERSION=0.9.33
ARG CARGO_DENY_VERSION=0.16.2
ARG CARGO_HACK_VERSION=0.6.33
ARG CARGO_NEXTEST_VERSION=0.9.85
ARG CARGO_HAKARI_VERSION=0.9.30
ARG CARGO_DENY_VERSION=0.16.1
ARG CARGO_HACK_VERSION=0.6.31
ARG CARGO_NEXTEST_VERSION=0.9.72
RUN curl -sSO https://static.rust-lang.org/rustup/dist/$(uname -m)-unknown-linux-gnu/rustup-init && whoami && \
chmod +x rustup-init && \
./rustup-init -y --default-toolchain ${RUSTC_VERSION} && \
@@ -289,7 +289,7 @@ RUN whoami \
&& cargo --version --verbose \
&& rustup --version --verbose \
&& rustc --version --verbose \
&& clang --version
&& clang --version
RUN if [ "${DEBIAN_VERSION}" = "bookworm" ]; then \
LD_LIBRARY_PATH=/pgcopydb/lib /pgcopydb/bin/pgcopydb --version; \

View File

@@ -14,9 +14,6 @@ ARG DEBIAN_FLAVOR=${DEBIAN_VERSION}-slim
FROM debian:$DEBIAN_FLAVOR AS build-deps
ARG DEBIAN_VERSION
# Use strict mode for bash to catch errors early
SHELL ["/bin/bash", "-euo", "pipefail", "-c"]
RUN case $DEBIAN_VERSION in \
# Version-specific installs for Bullseye (PG14-PG16):
# The h3_pg extension needs a cmake 3.20+, but Debian bullseye has 3.18.
@@ -109,7 +106,6 @@ RUN cd postgres && \
#
#########################################################################################
FROM build-deps AS postgis-build
ARG DEBIAN_VERSION
ARG PG_VERSION
COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
RUN apt update && \
@@ -126,12 +122,12 @@ RUN apt update && \
# and also we must check backward compatibility with older versions of PostGIS.
#
# Use new version only for v17
RUN case "${DEBIAN_VERSION}" in \
"bookworm") \
RUN case "${PG_VERSION}" in \
"v17") \
export SFCGAL_VERSION=1.4.1 \
export SFCGAL_CHECKSUM=1800c8a26241588f11cddcf433049e9b9aea902e923414d2ecef33a3295626c3 \
;; \
"bullseye") \
"v14" | "v15" | "v16") \
export SFCGAL_VERSION=1.3.10 \
export SFCGAL_CHECKSUM=4e39b3b2adada6254a7bdba6d297bb28e1a9835a9f879b74f37e2dab70203232 \
;; \
@@ -232,8 +228,6 @@ FROM build-deps AS plv8-build
ARG PG_VERSION
COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
COPY compute/patches/plv8-3.1.10.patch /plv8-3.1.10.patch
RUN apt update && \
apt install --no-install-recommends -y ninja-build python3-dev libncurses5 binutils clang
@@ -245,6 +239,8 @@ RUN apt update && \
#
# Use new version only for v17
# because since v3.2, plv8 doesn't include plcoffee and plls extensions
ENV PLV8_TAG=v3.2.3
RUN case "${PG_VERSION}" in \
"v17") \
export PLV8_TAG=v3.2.3 \
@@ -259,9 +255,8 @@ RUN case "${PG_VERSION}" in \
git clone --recurse-submodules --depth 1 --branch ${PLV8_TAG} https://github.com/plv8/plv8.git plv8-src && \
tar -czf plv8.tar.gz --exclude .git plv8-src && \
cd plv8-src && \
if [[ "${PG_VERSION}" < "v17" ]]; then patch -p1 < /plv8-3.1.10.patch; fi && \
# generate and copy upgrade scripts
mkdir -p upgrade && ./generate_upgrade.sh ${PLV8_TAG#v} && \
mkdir -p upgrade && ./generate_upgrade.sh 3.1.10 && \
cp upgrade/* /usr/local/pgsql/share/extension/ && \
export PATH="/usr/local/pgsql/bin:$PATH" && \
make DOCKER=1 -j $(getconf _NPROCESSORS_ONLN) install && \
@@ -358,10 +353,10 @@ COPY compute/patches/pgvector.patch /pgvector.patch
# because we build the images on different machines than where we run them.
# Pass OPTFLAGS="" to remove it.
#
# vector >0.7.4 supports v17
# last release v0.8.0 - Oct 30, 2024
RUN wget https://github.com/pgvector/pgvector/archive/refs/tags/v0.8.0.tar.gz -O pgvector.tar.gz && \
echo "867a2c328d4928a5a9d6f052cd3bc78c7d60228a9b914ad32aa3db88e9de27b0 pgvector.tar.gz" | sha256sum --check && \
# vector 0.7.4 supports v17
# last release v0.7.4 - Aug 5, 2024
RUN wget https://github.com/pgvector/pgvector/archive/refs/tags/v0.7.4.tar.gz -O pgvector.tar.gz && \
echo "0341edf89b1924ae0d552f617e14fb7f8867c0194ed775bcc44fa40288642583 pgvector.tar.gz" | sha256sum --check && \
mkdir pgvector-src && cd pgvector-src && tar xzf ../pgvector.tar.gz --strip-components=1 -C . && \
patch -p1 < /pgvector.patch && \
make -j $(getconf _NPROCESSORS_ONLN) OPTFLAGS="" PG_CONFIG=/usr/local/pgsql/bin/pg_config && \
@@ -1367,12 +1362,15 @@ RUN make PG_VERSION="${PG_VERSION}" -C compute
FROM neon-pg-ext-build AS neon-pg-ext-test
ARG PG_VERSION
RUN mkdir /ext-src
RUN case "${PG_VERSION}" in "v17") \
echo "v17 extensions are not supported yet. Quit" && exit 0;; \
esac && \
mkdir /ext-src
#COPY --from=postgis-build /postgis.tar.gz /ext-src/
#COPY --from=postgis-build /sfcgal/* /usr
COPY --from=plv8-build /plv8.tar.gz /ext-src/
#COPY --from=h3-pg-build /h3-pg.tar.gz /ext-src/
COPY --from=h3-pg-build /h3-pg.tar.gz /ext-src/
COPY --from=unit-pg-build /postgresql-unit.tar.gz /ext-src/
COPY --from=vector-pg-build /pgvector.tar.gz /ext-src/
COPY --from=vector-pg-build /pgvector.patch /ext-src/
@@ -1392,7 +1390,7 @@ COPY --from=hll-pg-build /hll.tar.gz /ext-src
COPY --from=plpgsql-check-pg-build /plpgsql_check.tar.gz /ext-src
#COPY --from=timescaledb-pg-build /timescaledb.tar.gz /ext-src
COPY --from=pg-hint-plan-pg-build /pg_hint_plan.tar.gz /ext-src
COPY compute/patches/pg_hint_plan_${PG_VERSION}.patch /ext-src
COPY compute/patches/pg_hint_plan.patch /ext-src
COPY --from=pg-cron-pg-build /pg_cron.tar.gz /ext-src
COPY compute/patches/pg_cron.patch /ext-src
#COPY --from=pg-pgx-ulid-build /home/nonroot/pgx_ulid.tar.gz /ext-src
@@ -1402,23 +1400,38 @@ COPY --from=pg-roaringbitmap-pg-build /pg_roaringbitmap.tar.gz /ext-src
COPY --from=pg-semver-pg-build /pg_semver.tar.gz /ext-src
#COPY --from=pg-embedding-pg-build /home/nonroot/pg_embedding-src/ /ext-src
#COPY --from=wal2json-pg-build /wal2json_2_5.tar.gz /ext-src
#pg_anon is not supported yet for pg v17 so, don't fail if nothing found
COPY --from=pg-anon-pg-build /pg_anon.tar.g? /ext-src
COPY --from=pg-anon-pg-build /pg_anon.tar.gz /ext-src
COPY compute/patches/pg_anon.patch /ext-src
COPY --from=pg-ivm-build /pg_ivm.tar.gz /ext-src
COPY --from=pg-partman-build /pg_partman.tar.gz /ext-src
RUN cd /ext-src/ && for f in *.tar.gz; \
RUN case "${PG_VERSION}" in "v17") \
echo "v17 extensions are not supported yet. Quit" && exit 0;; \
esac && \
cd /ext-src/ && for f in *.tar.gz; \
do echo $f; dname=$(echo $f | sed 's/\.tar.*//')-src; \
rm -rf $dname; mkdir $dname; tar xzf $f --strip-components=1 -C $dname \
|| exit 1; rm -f $f; done
RUN cd /ext-src/rum-src && patch -p1 <../rum.patch
RUN cd /ext-src/pgvector-src && patch -p1 <../pgvector.patch
RUN cd /ext-src/pg_hint_plan-src && patch -p1 < /ext-src/pg_hint_plan_${PG_VERSION}.patch
RUN case "${PG_VERSION}" in "v17") \
echo "v17 extensions are not supported yet. Quit" && exit 0;; \
esac && \
cd /ext-src/rum-src && patch -p1 <../rum.patch
RUN case "${PG_VERSION}" in "v17") \
echo "v17 extensions are not supported yet. Quit" && exit 0;; \
esac && \
cd /ext-src/pgvector-src && patch -p1 <../pgvector.patch
RUN case "${PG_VERSION}" in "v17") \
echo "v17 extensions are not supported yet. Quit" && exit 0;; \
esac && \
cd /ext-src/pg_hint_plan-src && patch -p1 < /ext-src/pg_hint_plan.patch
COPY --chmod=755 docker-compose/run-tests.sh /run-tests.sh
RUN case "${PG_VERSION}" in "v17") \
echo "postgresql_anonymizer does not yet support PG17" && exit 0;; \
esac && patch -p1 </ext-src/pg_anon.patch
RUN patch -p1 </ext-src/pg_cron.patch
echo "v17 extensions are not supported yet. Quit" && exit 0;; \
esac && \
patch -p1 </ext-src/pg_anon.patch
RUN case "${PG_VERSION}" in "v17") \
echo "v17 extensions are not supported yet. Quit" && exit 0;; \
esac && \
patch -p1 </ext-src/pg_cron.patch
ENV PATH=/usr/local/pgsql/bin:$PATH
ENV PGHOST=compute
ENV PGPORT=55433

View File

@@ -6,7 +6,6 @@
import 'sql_exporter/compute_backpressure_throttling_seconds.libsonnet',
import 'sql_exporter/compute_current_lsn.libsonnet',
import 'sql_exporter/compute_logical_snapshot_files.libsonnet',
import 'sql_exporter/compute_logical_snapshots_bytes.libsonnet',
import 'sql_exporter/compute_max_connections.libsonnet',
import 'sql_exporter/compute_receive_lsn.libsonnet',
import 'sql_exporter/compute_subscriptions_count.libsonnet',

View File

@@ -1,9 +1,5 @@
[databases]
;; pgbouncer propagates application_name (if it's specified) to the server, but some
;; clients don't set it. We set default application_name=pgbouncer to make it
;; easier to identify pgbouncer connections in Postgres. If client sets
;; application_name, it will be used instead.
*=host=localhost port=5432 auth_user=cloud_admin application_name=pgbouncer
*=host=localhost port=5432 auth_user=cloud_admin
[pgbouncer]
listen_port=6432
listen_addr=0.0.0.0

View File

@@ -1,7 +0,0 @@
SELECT
(SELECT current_setting('neon.timeline_id')) AS timeline_id,
-- Postgres creates temporary snapshot files of the form %X-%X.snap.%d.tmp.
-- These temporary snapshot files are renamed to the actual snapshot files
-- after they are completely built. We only WAL-log the completely built
-- snapshot files
(SELECT COALESCE(sum(size), 0) FROM pg_ls_logicalsnapdir() WHERE name LIKE '%.snap') AS logical_snapshots_bytes;

View File

@@ -1,17 +0,0 @@
local neon = import 'neon.libsonnet';
local pg_ls_logicalsnapdir = importstr 'sql_exporter/compute_logical_snapshots_bytes.15.sql';
local pg_ls_dir = importstr 'sql_exporter/compute_logical_snapshots_bytes.sql';
{
metric_name: 'compute_logical_snapshots_bytes',
type: 'gauge',
help: 'Size of the pg_logical/snapshots directory, not including temporary files',
key_labels: [
'timeline_id',
],
values: [
'logical_snapshots_bytes',
],
query: if neon.PG_MAJORVERSION_NUM < 15 then pg_ls_dir else pg_ls_logicalsnapdir,
}

View File

@@ -1,9 +0,0 @@
SELECT
(SELECT setting FROM pg_settings WHERE name = 'neon.timeline_id') AS timeline_id,
-- Postgres creates temporary snapshot files of the form %X-%X.snap.%d.tmp.
-- These temporary snapshot files are renamed to the actual snapshot files
-- after they are completely built. We only WAL-log the completely built
-- snapshot files
(SELECT COALESCE(sum((pg_stat_file('pg_logical/snapshots/' || name, missing_ok => true)).size), 0)
FROM (SELECT * FROM pg_ls_dir('pg_logical/snapshots') WHERE pg_ls_dir LIKE '%.snap') AS name
) AS logical_snapshots_bytes;

View File

@@ -1,174 +0,0 @@
diff --git a/expected/ut-A.out b/expected/ut-A.out
index e7d68a1..65a056c 100644
--- a/expected/ut-A.out
+++ b/expected/ut-A.out
@@ -9,13 +9,16 @@ SET search_path TO public;
----
-- No.A-1-1-3
CREATE EXTENSION pg_hint_plan;
+LOG: Sending request to compute_ctl: http://localhost:3080/extension_server/pg_hint_plan
-- No.A-1-2-3
DROP EXTENSION pg_hint_plan;
-- No.A-1-1-4
CREATE SCHEMA other_schema;
CREATE EXTENSION pg_hint_plan SCHEMA other_schema;
+LOG: Sending request to compute_ctl: http://localhost:3080/extension_server/pg_hint_plan
ERROR: extension "pg_hint_plan" must be installed in schema "hint_plan"
CREATE EXTENSION pg_hint_plan;
+LOG: Sending request to compute_ctl: http://localhost:3080/extension_server/pg_hint_plan
DROP SCHEMA other_schema;
----
---- No. A-5-1 comment pattern
diff --git a/expected/ut-J.out b/expected/ut-J.out
index 2fa3c70..314e929 100644
--- a/expected/ut-J.out
+++ b/expected/ut-J.out
@@ -789,38 +789,6 @@ NestLoop(st1 st2)
MergeJoin(t1 t2)
not used hint:
duplication hint:
-error hint:
-
-LOG: pg_hint_plan:
-used hint:
-not used hint:
-NestLoop(st1 st2)
-MergeJoin(t1 t2)
-duplication hint:
-error hint:
-
-LOG: pg_hint_plan:
-used hint:
-not used hint:
-NestLoop(st1 st2)
-MergeJoin(t1 t2)
-duplication hint:
-error hint:
-
-LOG: pg_hint_plan:
-used hint:
-not used hint:
-NestLoop(st1 st2)
-MergeJoin(t1 t2)
-duplication hint:
-error hint:
-
-LOG: pg_hint_plan:
-used hint:
-not used hint:
-NestLoop(st1 st2)
-MergeJoin(t1 t2)
-duplication hint:
error hint:
explain_filter
diff --git a/expected/ut-S.out b/expected/ut-S.out
index 0bfcfb8..e75f581 100644
--- a/expected/ut-S.out
+++ b/expected/ut-S.out
@@ -4415,34 +4415,6 @@ used hint:
IndexScan(ti1 ti1_pred)
not used hint:
duplication hint:
-error hint:
-
-LOG: pg_hint_plan:
-used hint:
-not used hint:
-IndexScan(ti1 ti1_pred)
-duplication hint:
-error hint:
-
-LOG: pg_hint_plan:
-used hint:
-not used hint:
-IndexScan(ti1 ti1_pred)
-duplication hint:
-error hint:
-
-LOG: pg_hint_plan:
-used hint:
-not used hint:
-IndexScan(ti1 ti1_pred)
-duplication hint:
-error hint:
-
-LOG: pg_hint_plan:
-used hint:
-not used hint:
-IndexScan(ti1 ti1_pred)
-duplication hint:
error hint:
explain_filter
diff --git a/expected/ut-W.out b/expected/ut-W.out
index a09bd34..0ad227c 100644
--- a/expected/ut-W.out
+++ b/expected/ut-W.out
@@ -1341,54 +1341,6 @@ IndexScan(ft1)
IndexScan(t)
Parallel(s1 3 hard)
duplication hint:
-error hint:
-
-LOG: pg_hint_plan:
-used hint:
-not used hint:
-IndexScan(*VALUES*)
-SeqScan(cte1)
-IndexScan(ft1)
-IndexScan(t)
-Parallel(p1 5 hard)
-Parallel(s1 3 hard)
-duplication hint:
-error hint:
-
-LOG: pg_hint_plan:
-used hint:
-not used hint:
-IndexScan(*VALUES*)
-SeqScan(cte1)
-IndexScan(ft1)
-IndexScan(t)
-Parallel(p1 5 hard)
-Parallel(s1 3 hard)
-duplication hint:
-error hint:
-
-LOG: pg_hint_plan:
-used hint:
-not used hint:
-IndexScan(*VALUES*)
-SeqScan(cte1)
-IndexScan(ft1)
-IndexScan(t)
-Parallel(p1 5 hard)
-Parallel(s1 3 hard)
-duplication hint:
-error hint:
-
-LOG: pg_hint_plan:
-used hint:
-not used hint:
-IndexScan(*VALUES*)
-SeqScan(cte1)
-IndexScan(ft1)
-IndexScan(t)
-Parallel(p1 5 hard)
-Parallel(s1 3 hard)
-duplication hint:
error hint:
explain_filter
diff --git a/expected/ut-fdw.out b/expected/ut-fdw.out
index 017fa4b..98d989b 100644
--- a/expected/ut-fdw.out
+++ b/expected/ut-fdw.out
@@ -7,6 +7,7 @@ SET pg_hint_plan.debug_print TO on;
SET client_min_messages TO LOG;
SET pg_hint_plan.enable_hint TO on;
CREATE EXTENSION file_fdw;
+LOG: Sending request to compute_ctl: http://localhost:3080/extension_server/file_fdw
CREATE SERVER file_server FOREIGN DATA WRAPPER file_fdw;
CREATE USER MAPPING FOR PUBLIC SERVER file_server;
CREATE FOREIGN TABLE ft1 (id int, val int) SERVER file_server OPTIONS (format 'csv', filename :'filename');

View File

@@ -1,42 +0,0 @@
commit 46b38d3e46f9cd6c70d9b189dd6ff4abaa17cf5e
Author: Alexander Bayandin <alexander@neon.tech>
Date: Sat Nov 30 18:29:32 2024 +0000
Fix v8 9.7.37 compilation on Debian 12
diff --git a/patches/code/84cf3230a9680aac3b73c410c2b758760b6d3066.patch b/patches/code/84cf3230a9680aac3b73c410c2b758760b6d3066.patch
new file mode 100644
index 0000000..f0a5dc7
--- /dev/null
+++ b/patches/code/84cf3230a9680aac3b73c410c2b758760b6d3066.patch
@@ -0,0 +1,30 @@
+From 84cf3230a9680aac3b73c410c2b758760b6d3066 Mon Sep 17 00:00:00 2001
+From: Michael Lippautz <mlippautz@chromium.org>
+Date: Thu, 27 Jan 2022 14:14:11 +0100
+Subject: [PATCH] cppgc: Fix include
+
+Add <utility> to cover for std::exchange.
+
+Bug: v8:12585
+Change-Id: Ida65144e93e466be8914527d0e646f348c136bcb
+Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3420309
+Auto-Submit: Michael Lippautz <mlippautz@chromium.org>
+Reviewed-by: Omer Katz <omerkatz@chromium.org>
+Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
+Cr-Commit-Position: refs/heads/main@{#78820}
+---
+ src/heap/cppgc/prefinalizer-handler.h | 1 +
+ 1 file changed, 1 insertion(+)
+
+diff --git a/src/heap/cppgc/prefinalizer-handler.h b/src/heap/cppgc/prefinalizer-handler.h
+index bc17c99b1838..c82c91ff5a45 100644
+--- a/src/heap/cppgc/prefinalizer-handler.h
++++ b/src/heap/cppgc/prefinalizer-handler.h
+@@ -5,6 +5,7 @@
+ #ifndef V8_HEAP_CPPGC_PREFINALIZER_HANDLER_H_
+ #define V8_HEAP_CPPGC_PREFINALIZER_HANDLER_H_
+
++#include <utility>
+ #include <vector>
+
+ #include "include/cppgc/prefinalizer.h"

View File

@@ -37,7 +37,6 @@ use std::collections::HashMap;
use std::fs::File;
use std::path::Path;
use std::process::exit;
use std::str::FromStr;
use std::sync::atomic::Ordering;
use std::sync::{mpsc, Arc, Condvar, Mutex, RwLock};
use std::{thread, time::Duration};
@@ -323,19 +322,11 @@ fn wait_spec(
} else {
spec_set = false;
}
let connstr = Url::parse(connstr).context("cannot parse connstr as a URL")?;
let conn_conf = postgres::config::Config::from_str(connstr.as_str())
.context("cannot build postgres config from connstr")?;
let tokio_conn_conf = tokio_postgres::config::Config::from_str(connstr.as_str())
.context("cannot build tokio postgres config from connstr")?;
let compute_node = ComputeNode {
connstr,
conn_conf,
tokio_conn_conf,
connstr: Url::parse(connstr).context("cannot parse connstr as a URL")?,
pgdata: pgdata.to_string(),
pgbin: pgbin.to_string(),
pgversion: get_pg_version_string(pgbin),
http_port,
live_config_allowed,
state: Mutex::new(new_state),
state_changed: Condvar::new(),
@@ -390,6 +381,7 @@ fn wait_spec(
Ok(WaitSpecResult {
compute,
http_port,
resize_swap_on_bind,
set_disk_quota_for_fs: set_disk_quota_for_fs.cloned(),
})
@@ -397,6 +389,8 @@ fn wait_spec(
struct WaitSpecResult {
compute: Arc<ComputeNode>,
// passed through from ProcessCliResult
http_port: u16,
resize_swap_on_bind: bool,
set_disk_quota_for_fs: Option<String>,
}
@@ -406,6 +400,7 @@ fn start_postgres(
#[allow(unused_variables)] matches: &clap::ArgMatches,
WaitSpecResult {
compute,
http_port,
resize_swap_on_bind,
set_disk_quota_for_fs,
}: WaitSpecResult,
@@ -478,10 +473,12 @@ fn start_postgres(
}
}
let extension_server_port: u16 = http_port;
// Start Postgres
let mut pg = None;
if !prestartup_failed {
pg = match compute.start_compute() {
pg = match compute.start_compute(extension_server_port) {
Ok(pg) => Some(pg),
Err(err) => {
error!("could not start the compute node: {:#}", err);

View File

@@ -21,7 +21,7 @@
//! - Build the image with the following command:
//!
//! ```bash
//! docker buildx build --platform linux/amd64 --build-arg DEBIAN_VERSION=bullseye --build-arg GIT_VERSION=local --build-arg PG_VERSION=v14 --build-arg BUILD_TAG="$(date --iso-8601=s -u)" -t localhost:3030/localregistry/compute-node-v14:latest -f compute/compute-node.Dockerfile .
//! docker buildx build --build-arg DEBIAN_FLAVOR=bullseye-slim --build-arg GIT_VERSION=local --build-arg PG_VERSION=v14 --build-arg BUILD_TAG="$(date --iso-8601=s -u)" -t localhost:3030/localregistry/compute-node-v14:latest -f compute/Dockerfile.com
//! docker push localhost:3030/localregistry/compute-node-v14:latest
//! ```
@@ -132,8 +132,7 @@ pub(crate) async fn main() -> anyhow::Result<()> {
//
// Initialize pgdata
//
let pgbin = pg_bin_dir.join("postgres");
let pg_version = match get_pg_version(pgbin.as_ref()) {
let pg_version = match get_pg_version(pg_bin_dir.as_str()) {
PostgresMajorVersion::V14 => 14,
PostgresMajorVersion::V15 => 15,
PostgresMajorVersion::V16 => 16,
@@ -156,7 +155,7 @@ pub(crate) async fn main() -> anyhow::Result<()> {
//
// Launch postgres process
//
let mut postgres_proc = tokio::process::Command::new(pgbin)
let mut postgres_proc = tokio::process::Command::new(pg_bin_dir.join("postgres"))
.arg("-D")
.arg(&pgdata_dir)
.args(["-c", "wal_level=minimal"])

View File

@@ -1,3 +1,4 @@
use compute_api::responses::CatalogObjects;
use futures::Stream;
use postgres::NoTls;
use std::{path::Path, process::Stdio, result::Result, sync::Arc};
@@ -6,17 +7,19 @@ use tokio::{
process::Command,
spawn,
};
use tokio_postgres::connect;
use tokio_stream::{self as stream, StreamExt};
use tokio_util::codec::{BytesCodec, FramedRead};
use tracing::warn;
use crate::compute::ComputeNode;
use crate::pg_helpers::{get_existing_dbs_async, get_existing_roles_async, postgres_conf_for_db};
use compute_api::responses::CatalogObjects;
use crate::pg_helpers::{get_existing_dbs_async, get_existing_roles_async};
pub async fn get_dbs_and_roles(compute: &Arc<ComputeNode>) -> anyhow::Result<CatalogObjects> {
let conf = compute.get_tokio_conn_conf(Some("compute_ctl:get_dbs_and_roles"));
let (client, connection): (tokio_postgres::Client, _) = conf.connect(NoTls).await?;
let connstr = compute.connstr.clone();
let (client, connection): (tokio_postgres::Client, _) =
connect(connstr.as_str(), NoTls).await?;
spawn(async move {
if let Err(e) = connection.await {
@@ -40,8 +43,6 @@ pub enum SchemaDumpError {
DatabaseDoesNotExist,
#[error("Failed to execute pg_dump.")]
IO(#[from] std::io::Error),
#[error("Unexpected error.")]
Unexpected,
}
// It uses the pg_dump utility to dump the schema of the specified database.
@@ -59,38 +60,11 @@ pub async fn get_database_schema(
let pgbin = &compute.pgbin;
let basepath = Path::new(pgbin).parent().unwrap();
let pgdump = basepath.join("pg_dump");
// Replace the DB in the connection string and disable it to parts.
// This is the only option to handle DBs with special characters.
let conf =
postgres_conf_for_db(&compute.connstr, dbname).map_err(|_| SchemaDumpError::Unexpected)?;
let host = conf
.get_hosts()
.first()
.ok_or(SchemaDumpError::Unexpected)?;
let host = match host {
tokio_postgres::config::Host::Tcp(ip) => ip.to_string(),
#[cfg(unix)]
tokio_postgres::config::Host::Unix(path) => path.to_string_lossy().to_string(),
};
let port = conf
.get_ports()
.first()
.ok_or(SchemaDumpError::Unexpected)?;
let user = conf.get_user().ok_or(SchemaDumpError::Unexpected)?;
let dbname = conf.get_dbname().ok_or(SchemaDumpError::Unexpected)?;
let mut connstr = compute.connstr.clone();
connstr.set_path(dbname);
let mut cmd = Command::new(pgdump)
// XXX: this seems to be the only option to deal with DBs with `=` in the name
// See <https://www.postgresql.org/message-id/flat/20151023003445.931.91267%40wrigleys.postgresql.org>
.env("PGDATABASE", dbname)
.arg("--host")
.arg(host)
.arg("--port")
.arg(port.to_string())
.arg("--username")
.arg(user)
.arg("--schema-only")
.arg(connstr.as_str())
.stdout(Stdio::piped())
.stderr(Stdio::piped())
.kill_on_drop(true)

View File

@@ -9,8 +9,7 @@ use crate::compute::ComputeNode;
#[instrument(skip_all)]
pub async fn check_writability(compute: &ComputeNode) -> Result<()> {
// Connect to the database.
let conf = compute.get_tokio_conn_conf(Some("compute_ctl:availability_checker"));
let (client, connection) = conf.connect(NoTls).await?;
let (client, connection) = tokio_postgres::connect(compute.connstr.as_str(), NoTls).await?;
if client.is_closed() {
return Err(anyhow!("connection to postgres closed"));
}

View File

@@ -20,9 +20,8 @@ use futures::future::join_all;
use futures::stream::FuturesUnordered;
use futures::StreamExt;
use nix::unistd::Pid;
use postgres;
use postgres::error::SqlState;
use postgres::NoTls;
use postgres::{Client, NoTls};
use tracing::{debug, error, info, instrument, warn};
use utils::id::{TenantId, TimelineId};
use utils::lsn::Lsn;
@@ -35,8 +34,9 @@ use utils::measured_stream::MeasuredReader;
use nix::sys::signal::{kill, Signal};
use remote_storage::{DownloadError, RemotePath};
use tokio::spawn;
use url::Url;
use crate::installed_extensions::get_installed_extensions;
use crate::installed_extensions::get_installed_extensions_sync;
use crate::local_proxy;
use crate::pg_helpers::*;
use crate::spec::*;
@@ -59,10 +59,6 @@ pub static PG_PID: AtomicU32 = AtomicU32::new(0);
pub struct ComputeNode {
// Url type maintains proper escaping
pub connstr: url::Url,
// We connect to Postgres from many different places, so build configs once
// and reuse them where needed.
pub conn_conf: postgres::config::Config,
pub tokio_conn_conf: tokio_postgres::config::Config,
pub pgdata: String,
pub pgbin: String,
pub pgversion: String,
@@ -79,8 +75,6 @@ pub struct ComputeNode {
/// - we push spec and it does configuration
/// - but then it is restarted without any spec again
pub live_config_allowed: bool,
/// The port that the compute's HTTP server listens on
pub http_port: u16,
/// Volatile part of the `ComputeNode`, which should be used under `Mutex`.
/// To allow HTTP API server to serving status requests, while configuration
/// is in progress, lock should be held only for short periods of time to do
@@ -613,7 +607,11 @@ impl ComputeNode {
/// Do all the preparations like PGDATA directory creation, configuration,
/// safekeepers sync, basebackup, etc.
#[instrument(skip_all)]
pub fn prepare_pgdata(&self, compute_state: &ComputeState) -> Result<()> {
pub fn prepare_pgdata(
&self,
compute_state: &ComputeState,
extension_server_port: u16,
) -> Result<()> {
let pspec = compute_state.pspec.as_ref().expect("spec must be set");
let spec = &pspec.spec;
let pgdata_path = Path::new(&self.pgdata);
@@ -623,7 +621,7 @@ impl ComputeNode {
config::write_postgres_conf(
&pgdata_path.join("postgresql.conf"),
&pspec.spec,
self.http_port,
Some(extension_server_port),
)?;
// Syncing safekeepers is only safe with primary nodes: if a primary
@@ -803,10 +801,10 @@ impl ComputeNode {
/// version. In the future, it may upgrade all 3rd-party extensions.
#[instrument(skip_all)]
pub fn post_apply_config(&self) -> Result<()> {
let conf = self.get_conn_conf(Some("compute_ctl:post_apply_config"));
let connstr = self.connstr.clone();
thread::spawn(move || {
let func = || {
let mut client = conf.connect(NoTls)?;
let mut client = Client::connect(connstr.as_str(), NoTls)?;
handle_neon_extension_upgrade(&mut client)
.context("handle_neon_extension_upgrade")?;
Ok::<_, anyhow::Error>(())
@@ -818,48 +816,30 @@ impl ComputeNode {
Ok(())
}
pub fn get_conn_conf(&self, application_name: Option<&str>) -> postgres::Config {
let mut conf = self.conn_conf.clone();
if let Some(application_name) = application_name {
conf.application_name(application_name);
}
conf
}
async fn get_maintenance_client(url: &Url) -> Result<tokio_postgres::Client> {
let mut connstr = url.clone();
pub fn get_tokio_conn_conf(&self, application_name: Option<&str>) -> tokio_postgres::Config {
let mut conf = self.tokio_conn_conf.clone();
if let Some(application_name) = application_name {
conf.application_name(application_name);
}
conf
}
connstr
.query_pairs_mut()
.append_pair("application_name", "apply_config");
async fn get_maintenance_client(
conf: &tokio_postgres::Config,
) -> Result<tokio_postgres::Client> {
let mut conf = conf.clone();
conf.application_name("compute_ctl:apply_config");
let (client, conn) = match conf.connect(NoTls).await {
// If connection fails, it may be the old node with `zenith_admin` superuser.
//
// In this case we need to connect with old `zenith_admin` name
// and create new user. We cannot simply rename connected user,
// but we can create a new one and grant it all privileges.
let (client, conn) = match tokio_postgres::connect(connstr.as_str(), NoTls).await {
Err(e) => match e.code() {
Some(&SqlState::INVALID_PASSWORD)
| Some(&SqlState::INVALID_AUTHORIZATION_SPECIFICATION) => {
// Connect with zenith_admin if cloud_admin could not authenticate
// connect with zenith_admin if cloud_admin could not authenticate
info!(
"cannot connect to postgres: {}, retrying with `zenith_admin` username",
e
);
let mut zenith_admin_conf = postgres::config::Config::from(conf.clone());
zenith_admin_conf.application_name("compute_ctl:apply_config");
zenith_admin_conf.user("zenith_admin");
let mut zenith_admin_connstr = connstr.clone();
zenith_admin_connstr
.set_username("zenith_admin")
.map_err(|_| anyhow::anyhow!("invalid connstr"))?;
let mut client =
zenith_admin_conf.connect(NoTls)
Client::connect(zenith_admin_connstr.as_str(), NoTls)
.context("broken cloud_admin credential: tried connecting with cloud_admin but could not authenticate, and zenith_admin does not work either")?;
// Disable forwarding so that users don't get a cloud_admin role
@@ -873,8 +853,8 @@ impl ComputeNode {
drop(client);
// Reconnect with connstring with expected name
conf.connect(NoTls).await?
// reconnect with connstring with expected name
tokio_postgres::connect(connstr.as_str(), NoTls).await?
}
_ => return Err(e.into()),
},
@@ -905,7 +885,7 @@ impl ComputeNode {
pub fn apply_spec_sql(
&self,
spec: Arc<ComputeSpec>,
conf: Arc<tokio_postgres::Config>,
url: Arc<Url>,
concurrency: usize,
) -> Result<()> {
let rt = tokio::runtime::Builder::new_multi_thread()
@@ -917,7 +897,7 @@ impl ComputeNode {
rt.block_on(async {
// Proceed with post-startup configuration. Note, that order of operations is important.
let client = Self::get_maintenance_client(&conf).await?;
let client = Self::get_maintenance_client(&url).await?;
let spec = spec.clone();
let databases = get_existing_dbs_async(&client).await?;
@@ -951,7 +931,7 @@ impl ComputeNode {
RenameAndDeleteDatabases,
CreateAndAlterDatabases,
] {
info!("Applying phase {:?}", &phase);
debug!("Applying phase {:?}", &phase);
apply_operations(
spec.clone(),
ctx.clone(),
@@ -962,7 +942,6 @@ impl ComputeNode {
.await?;
}
info!("Applying RunInEachDatabase phase");
let concurrency_token = Arc::new(tokio::sync::Semaphore::new(concurrency));
let db_processes = spec
@@ -976,7 +955,7 @@ impl ComputeNode {
let spec = spec.clone();
let ctx = ctx.clone();
let jwks_roles = jwks_roles.clone();
let mut conf = conf.as_ref().clone();
let mut url = url.as_ref().clone();
let concurrency_token = concurrency_token.clone();
let db = db.clone();
@@ -985,14 +964,14 @@ impl ComputeNode {
match &db {
DB::SystemDB => {}
DB::UserDB(db) => {
conf.dbname(db.name.as_str());
url.set_path(db.name.as_str());
}
}
let conf = Arc::new(conf);
let url = Arc::new(url);
let fut = Self::apply_spec_sql_db(
spec.clone(),
conf,
url,
ctx.clone(),
jwks_roles.clone(),
concurrency_token.clone(),
@@ -1038,7 +1017,7 @@ impl ComputeNode {
/// semaphore. The caller has to make sure the semaphore isn't exhausted.
async fn apply_spec_sql_db(
spec: Arc<ComputeSpec>,
conf: Arc<tokio_postgres::Config>,
url: Arc<Url>,
ctx: Arc<tokio::sync::RwLock<MutableApplyContext>>,
jwks_roles: Arc<HashSet<String>>,
concurrency_token: Arc<tokio::sync::Semaphore>,
@@ -1067,7 +1046,7 @@ impl ComputeNode {
// that database.
|| async {
if client_conn.is_none() {
let db_client = Self::get_maintenance_client(&conf).await?;
let db_client = Self::get_maintenance_client(&url).await?;
client_conn.replace(db_client);
}
let client = client_conn.as_ref().unwrap();
@@ -1082,16 +1061,34 @@ impl ComputeNode {
Ok::<(), anyhow::Error>(())
}
/// Choose how many concurrent connections to use for applying the spec changes.
pub fn max_service_connections(
&self,
compute_state: &ComputeState,
spec: &ComputeSpec,
) -> usize {
// If the cluster is in Init state we don't have to deal with user connections,
/// Do initial configuration of the already started Postgres.
#[instrument(skip_all)]
pub fn apply_config(&self, compute_state: &ComputeState) -> Result<()> {
// If connection fails,
// it may be the old node with `zenith_admin` superuser.
//
// In this case we need to connect with old `zenith_admin` name
// and create new user. We cannot simply rename connected user,
// but we can create a new one and grant it all privileges.
let mut url = self.connstr.clone();
url.query_pairs_mut()
.append_pair("application_name", "apply_config");
let url = Arc::new(url);
let spec = Arc::new(
compute_state
.pspec
.as_ref()
.expect("spec must be set")
.spec
.clone(),
);
// Choose how many concurrent connections to use for applying the spec changes.
// If the cluster is not currently Running we don't have to deal with user connections,
// and can thus use all `max_connections` connection slots. However, that's generally not
// very efficient, so we generally still limit it to a smaller number.
if compute_state.status == ComputeStatus::Init {
let max_concurrent_connections = if compute_state.status != ComputeStatus::Running {
// If the settings contain 'max_connections', use that as template
if let Some(config) = spec.cluster.settings.find("max_connections") {
config.parse::<usize>().ok()
@@ -1147,28 +1144,10 @@ impl ComputeNode {
.map(|val| if val > 1 { val - 1 } else { 1 })
.last()
.unwrap_or(3)
}
}
/// Do initial configuration of the already started Postgres.
#[instrument(skip_all)]
pub fn apply_config(&self, compute_state: &ComputeState) -> Result<()> {
let conf = self.get_tokio_conn_conf(Some("compute_ctl:apply_config"));
let conf = Arc::new(conf);
let spec = Arc::new(
compute_state
.pspec
.as_ref()
.expect("spec must be set")
.spec
.clone(),
);
let max_concurrent_connections = self.max_service_connections(compute_state, &spec);
};
// Merge-apply spec & changes to PostgreSQL state.
self.apply_spec_sql(spec.clone(), conf.clone(), max_concurrent_connections)?;
self.apply_spec_sql(spec.clone(), url.clone(), max_concurrent_connections)?;
if let Some(ref local_proxy) = &spec.clone().local_proxy_config {
info!("configuring local_proxy");
@@ -1177,11 +1156,12 @@ impl ComputeNode {
// Run migrations separately to not hold up cold starts
thread::spawn(move || {
let conf = conf.as_ref().clone();
let mut conf = postgres::config::Config::from(conf);
conf.application_name("compute_ctl:migrations");
let mut connstr = url.as_ref().clone();
connstr
.query_pairs_mut()
.append_pair("application_name", "migrations");
let mut client = conf.connect(NoTls)?;
let mut client = Client::connect(connstr.as_str(), NoTls)?;
handle_migrations(&mut client).context("apply_config handle_migrations")
});
@@ -1241,24 +1221,22 @@ impl ComputeNode {
// Write new config
let pgdata_path = Path::new(&self.pgdata);
let postgresql_conf_path = pgdata_path.join("postgresql.conf");
config::write_postgres_conf(&postgresql_conf_path, &spec, self.http_port)?;
let max_concurrent_connections = spec.reconfigure_concurrency;
// Temporarily reset max_cluster_size in config
config::write_postgres_conf(&postgresql_conf_path, &spec, None)?;
// temporarily reset max_cluster_size in config
// to avoid the possibility of hitting the limit, while we are reconfiguring:
// creating new extensions, roles, etc.
// creating new extensions, roles, etc...
config::with_compute_ctl_tmp_override(pgdata_path, "neon.max_cluster_size=-1", || {
self.pg_reload_conf()?;
if spec.mode == ComputeMode::Primary {
let mut conf = tokio_postgres::Config::from_str(self.connstr.as_str()).unwrap();
conf.application_name("apply_config");
let conf = Arc::new(conf);
let mut url = self.connstr.clone();
url.query_pairs_mut()
.append_pair("application_name", "apply_config");
let url = Arc::new(url);
let spec = Arc::new(spec.clone());
self.apply_spec_sql(spec, conf, max_concurrent_connections)?;
self.apply_spec_sql(spec, url, 1)?;
}
Ok(())
@@ -1277,7 +1255,10 @@ impl ComputeNode {
}
#[instrument(skip_all)]
pub fn start_compute(&self) -> Result<(std::process::Child, std::thread::JoinHandle<()>)> {
pub fn start_compute(
&self,
extension_server_port: u16,
) -> Result<(std::process::Child, std::thread::JoinHandle<()>)> {
let compute_state = self.state.lock().unwrap().clone();
let pspec = compute_state.pspec.as_ref().expect("spec must be set");
info!(
@@ -1352,7 +1333,7 @@ impl ComputeNode {
info!("{:?}", remote_ext_metrics);
}
self.prepare_pgdata(&compute_state)?;
self.prepare_pgdata(&compute_state, extension_server_port)?;
let start_time = Utc::now();
let pg_process = self.start_postgres(pspec.storage_auth_token.clone())?;
@@ -1379,19 +1360,9 @@ impl ComputeNode {
}
self.post_apply_config()?;
let conf = self.get_conn_conf(None);
let connstr = self.connstr.clone();
thread::spawn(move || {
let res = get_installed_extensions(conf);
match res {
Ok(extensions) => {
info!(
"[NEON_EXT_STAT] {}",
serde_json::to_string(&extensions)
.expect("failed to serialize extensions list")
);
}
Err(err) => error!("could not get installed extensions: {err:?}"),
}
get_installed_extensions_sync(connstr).context("get_installed_extensions")
});
}
@@ -1520,8 +1491,7 @@ impl ComputeNode {
/// Select `pg_stat_statements` data and return it as a stringified JSON
pub async fn collect_insights(&self) -> String {
let mut result_rows: Vec<String> = Vec::new();
let conf = self.get_tokio_conn_conf(Some("compute_ctl:collect_insights"));
let connect_result = conf.connect(NoTls).await;
let connect_result = tokio_postgres::connect(self.connstr.as_str(), NoTls).await;
let (client, connection) = connect_result.unwrap();
tokio::spawn(async move {
if let Err(e) = connection.await {
@@ -1647,9 +1617,10 @@ LIMIT 100",
privileges: &[Privilege],
role_name: &PgIdent,
) -> Result<()> {
use tokio_postgres::config::Config;
use tokio_postgres::NoTls;
let mut conf = self.get_tokio_conn_conf(Some("compute_ctl:set_role_grants"));
let mut conf = Config::from_str(self.connstr.as_str()).unwrap();
conf.dbname(db_name);
let (db_client, conn) = conf
@@ -1686,9 +1657,10 @@ LIMIT 100",
db_name: &PgIdent,
ext_version: ExtVersion,
) -> Result<ExtVersion> {
use tokio_postgres::config::Config;
use tokio_postgres::NoTls;
let mut conf = self.get_tokio_conn_conf(Some("compute_ctl:install_extension"));
let mut conf = Config::from_str(self.connstr.as_str()).unwrap();
conf.dbname(db_name);
let (db_client, conn) = conf

View File

@@ -37,7 +37,7 @@ pub fn line_in_file(path: &Path, line: &str) -> Result<bool> {
pub fn write_postgres_conf(
path: &Path,
spec: &ComputeSpec,
extension_server_port: u16,
extension_server_port: Option<u16>,
) -> Result<()> {
// File::create() destroys the file content if it exists.
let mut file = File::create(path)?;
@@ -127,7 +127,9 @@ pub fn write_postgres_conf(
writeln!(file, "# Managed by compute_ctl: end")?;
}
writeln!(file, "neon.extension_server_port={}", extension_server_port)?;
if let Some(port) = extension_server_port {
writeln!(file, "neon.extension_server_port={}", port)?;
}
// This is essential to keep this line at the end of the file,
// because it is intended to override any settings above.

View File

@@ -295,12 +295,8 @@ async fn routes(req: Request<Body>, compute: &Arc<ComputeNode>) -> Response<Body
return Response::new(Body::from(msg));
}
let conf = compute.get_conn_conf(None);
let res =
task::spawn_blocking(move || installed_extensions::get_installed_extensions(conf))
.await
.unwrap();
let connstr = compute.connstr.clone();
let res = crate::installed_extensions::get_installed_extensions(connstr).await;
match res {
Ok(res) => render_json(Body::from(serde_json::to_string(&res).unwrap())),
Err(e) => render_json_error(

View File

@@ -2,9 +2,12 @@ use compute_api::responses::{InstalledExtension, InstalledExtensions};
use metrics::proto::MetricFamily;
use std::collections::HashMap;
use std::collections::HashSet;
use tracing::info;
use url::Url;
use anyhow::Result;
use postgres::{Client, NoTls};
use tokio::task;
use metrics::core::Collector;
use metrics::{register_uint_gauge_vec, UIntGaugeVec};
@@ -39,53 +42,75 @@ fn list_dbs(client: &mut Client) -> Result<Vec<String>> {
///
/// Same extension can be installed in multiple databases with different versions,
/// we only keep the highest and lowest version across all databases.
pub fn get_installed_extensions(mut conf: postgres::config::Config) -> Result<InstalledExtensions> {
conf.application_name("compute_ctl:get_installed_extensions");
let mut client = conf.connect(NoTls)?;
pub async fn get_installed_extensions(connstr: Url) -> Result<InstalledExtensions> {
let mut connstr = connstr.clone();
let databases: Vec<String> = list_dbs(&mut client)?;
task::spawn_blocking(move || {
let mut client = Client::connect(connstr.as_str(), NoTls)?;
let databases: Vec<String> = list_dbs(&mut client)?;
let mut extensions_map: HashMap<String, InstalledExtension> = HashMap::new();
for db in databases.iter() {
conf.dbname(db);
let mut db_client = conf.connect(NoTls)?;
let extensions: Vec<(String, String)> = db_client
.query(
"SELECT extname, extversion FROM pg_catalog.pg_extension;",
&[],
)?
.iter()
.map(|row| (row.get("extname"), row.get("extversion")))
.collect();
let mut extensions_map: HashMap<String, InstalledExtension> = HashMap::new();
for db in databases.iter() {
connstr.set_path(db);
let mut db_client = Client::connect(connstr.as_str(), NoTls)?;
let extensions: Vec<(String, String)> = db_client
.query(
"SELECT extname, extversion FROM pg_catalog.pg_extension;",
&[],
)?
.iter()
.map(|row| (row.get("extname"), row.get("extversion")))
.collect();
for (extname, v) in extensions.iter() {
let version = v.to_string();
for (extname, v) in extensions.iter() {
let version = v.to_string();
// increment the number of databases where the version of extension is installed
INSTALLED_EXTENSIONS
.with_label_values(&[extname, &version])
.inc();
// increment the number of databases where the version of extension is installed
INSTALLED_EXTENSIONS
.with_label_values(&[extname, &version])
.inc();
extensions_map
.entry(extname.to_string())
.and_modify(|e| {
e.versions.insert(version.clone());
// count the number of databases where the extension is installed
e.n_databases += 1;
})
.or_insert(InstalledExtension {
extname: extname.to_string(),
versions: HashSet::from([version.clone()]),
n_databases: 1,
});
extensions_map
.entry(extname.to_string())
.and_modify(|e| {
e.versions.insert(version.clone());
// count the number of databases where the extension is installed
e.n_databases += 1;
})
.or_insert(InstalledExtension {
extname: extname.to_string(),
versions: HashSet::from([version.clone()]),
n_databases: 1,
});
}
}
}
let res = InstalledExtensions {
extensions: extensions_map.into_values().collect(),
};
let res = InstalledExtensions {
extensions: extensions_map.values().cloned().collect(),
};
Ok(res)
Ok(res)
})
.await?
}
// Gather info about installed extensions
pub fn get_installed_extensions_sync(connstr: Url) -> Result<()> {
let rt = tokio::runtime::Builder::new_current_thread()
.enable_all()
.build()
.expect("failed to create runtime");
let result = rt
.block_on(crate::installed_extensions::get_installed_extensions(
connstr,
))
.expect("failed to get installed extensions");
info!(
"[NEON_EXT_STAT] {}",
serde_json::to_string(&result).expect("failed to serialize extensions list")
);
Ok(())
}
static INSTALLED_EXTENSIONS: Lazy<UIntGaugeVec> = Lazy::new(|| {

View File

@@ -17,8 +17,11 @@ const MONITOR_CHECK_INTERVAL: Duration = Duration::from_millis(500);
// should be handled gracefully.
fn watch_compute_activity(compute: &ComputeNode) {
// Suppose that `connstr` doesn't change
let connstr = compute.connstr.clone();
let conf = compute.get_conn_conf(Some("compute_ctl:activity_monitor"));
let mut connstr = compute.connstr.clone();
connstr
.query_pairs_mut()
.append_pair("application_name", "compute_activity_monitor");
let connstr = connstr.as_str();
// During startup and configuration we connect to every Postgres database,
// but we don't want to count this as some user activity. So wait until
@@ -26,7 +29,7 @@ fn watch_compute_activity(compute: &ComputeNode) {
wait_for_postgres_start(compute);
// Define `client` outside of the loop to reuse existing connection if it's active.
let mut client = conf.connect(NoTls);
let mut client = Client::connect(connstr, NoTls);
let mut sleep = false;
let mut prev_active_time: Option<f64> = None;
@@ -54,7 +57,7 @@ fn watch_compute_activity(compute: &ComputeNode) {
info!("connection to Postgres is closed, trying to reconnect");
// Connection is closed, reconnect and try again.
client = conf.connect(NoTls);
client = Client::connect(connstr, NoTls);
continue;
}
@@ -193,7 +196,7 @@ fn watch_compute_activity(compute: &ComputeNode) {
debug!("could not connect to Postgres: {}, retrying", e);
// Establish a new connection and try again.
client = conf.connect(NoTls);
client = Client::connect(connstr, NoTls);
}
}
}

View File

@@ -6,7 +6,6 @@ use std::io::{BufRead, BufReader};
use std::os::unix::fs::PermissionsExt;
use std::path::Path;
use std::process::Child;
use std::str::FromStr;
use std::thread::JoinHandle;
use std::time::{Duration, Instant};
@@ -14,10 +13,8 @@ use anyhow::{bail, Result};
use futures::StreamExt;
use ini::Ini;
use notify::{RecursiveMode, Watcher};
use postgres::config::Config;
use tokio::io::AsyncBufReadExt;
use tokio::time::timeout;
use tokio_postgres;
use tokio_postgres::NoTls;
use tracing::{debug, error, info, instrument};
@@ -545,11 +542,3 @@ async fn handle_postgres_logs_async(stderr: tokio::process::ChildStderr) -> Resu
Ok(())
}
/// `Postgres::config::Config` handles database names with whitespaces
/// and special characters properly.
pub fn postgres_conf_for_db(connstr: &url::Url, dbname: &str) -> Result<Config> {
let mut conf = Config::from_str(connstr.as_str())?;
conf.dbname(dbname);
Ok(conf)
}

View File

@@ -53,7 +53,6 @@ use compute_api::spec::Role;
use nix::sys::signal::kill;
use nix::sys::signal::Signal;
use pageserver_api::shard::ShardStripeSize;
use reqwest::header::CONTENT_TYPE;
use serde::{Deserialize, Serialize};
use url::Host;
use utils::id::{NodeId, TenantId, TimelineId};
@@ -311,10 +310,6 @@ impl Endpoint {
conf.append("wal_log_hints", "off");
conf.append("max_replication_slots", "10");
conf.append("hot_standby", "on");
// Set to 1MB to both exercise getPage requests/LFC, and still have enough room for
// Postgres to operate. Everything smaller might be not enough for Postgres under load,
// and can cause errors like 'no unpinned buffers available', see
// <https://github.com/neondatabase/neon/issues/9956>
conf.append("shared_buffers", "1MB");
conf.append("fsync", "off");
conf.append("max_connections", "100");
@@ -619,7 +614,6 @@ impl Endpoint {
pgbouncer_settings: None,
shard_stripe_size: Some(shard_stripe_size),
local_proxy_config: None,
reconfigure_concurrency: 1,
};
let spec_path = self.endpoint_path().join("spec.json");
std::fs::write(spec_path, serde_json::to_string_pretty(&spec)?)?;
@@ -819,7 +813,6 @@ impl Endpoint {
self.http_address.ip(),
self.http_address.port()
))
.header(CONTENT_TYPE.as_str(), "application/json")
.body(format!(
"{{\"spec\":{}}}",
serde_json::to_string_pretty(&spec)?

View File

@@ -5,7 +5,6 @@
//! ```text
//! .neon/safekeepers/<safekeeper id>
//! ```
use std::error::Error as _;
use std::future::Future;
use std::io::Write;
use std::path::PathBuf;
@@ -27,7 +26,7 @@ use crate::{
#[derive(Error, Debug)]
pub enum SafekeeperHttpError {
#[error("request error: {0}{}", .0.source().map(|e| format!(": {e}")).unwrap_or_default())]
#[error("Reqwest error: {0}")]
Transport(#[from] reqwest::Error),
#[error("Error: {0}")]

View File

@@ -560,26 +560,14 @@ async fn main() -> anyhow::Result<()> {
.await?;
}
Command::TenantDescribe { tenant_id } => {
let TenantDescribeResponse {
tenant_id,
shards,
stripe_size,
policy,
config,
} = storcon_client
let describe_response = storcon_client
.dispatch::<(), TenantDescribeResponse>(
Method::GET,
format!("control/v1/tenant/{tenant_id}"),
None,
)
.await?;
println!("Tenant {tenant_id}");
let mut table = comfy_table::Table::new();
table.add_row(["Policy", &format!("{:?}", policy)]);
table.add_row(["Stripe size", &format!("{:?}", stripe_size)]);
table.add_row(["Config", &serde_json::to_string_pretty(&config).unwrap()]);
println!("{table}");
println!("Shards:");
let shards = describe_response.shards;
let mut table = comfy_table::Table::new();
table.set_header(["Shard", "Attached", "Secondary", "Last error", "status"]);
for shard in shards {

View File

@@ -4,16 +4,14 @@ ARG TAG=latest
FROM $REPOSITORY/${COMPUTE_IMAGE}:$TAG
ARG COMPUTE_IMAGE
USER root
RUN apt-get update && \
apt-get install -y curl \
jq \
python3-pip \
netcat-openbsd
netcat
#Faker is required for the pg_anon test
RUN case $COMPUTE_IMAGE in compute-node-v17) OPT="--break-system-packages";; *) OPT= ;; esac && pip3 install $OPT Faker
RUN pip3 install Faker
#This is required for the pg_hintplan test
RUN mkdir -p /ext-src/pg_hint_plan-src && chown postgres /ext-src/pg_hint_plan-src

View File

@@ -30,17 +30,10 @@ cleanup() {
docker compose --profile test-extensions -f $COMPOSE_FILE down
}
for pg_version in ${TEST_VERSION_ONLY-14 15 16 17}; do
pg_version=${pg_version/v/}
for pg_version in 14 15 16; do
echo "clean up containers if exists"
cleanup
PG_TEST_VERSION=$((pg_version < 16 ? 16 : pg_version))
# The support of pg_anon not yet added to PG17, so we have to remove the corresponding option
if [ $pg_version -eq 17 ]; then
SPEC_PATH="compute_wrapper/var/db/postgres/specs"
mv $SPEC_PATH/spec.json $SPEC_PATH/spec.bak
jq 'del(.cluster.settings[] | select (.name == "session_preload_libraries"))' $SPEC_PATH/spec.bak > $SPEC_PATH/spec.json
fi
PG_TEST_VERSION=$(($pg_version < 16 ? 16 : $pg_version))
PG_VERSION=$pg_version PG_TEST_VERSION=$PG_TEST_VERSION docker compose --profile test-extensions -f $COMPOSE_FILE up --build -d
echo "wait until the compute is ready. timeout after 60s. "
@@ -61,7 +54,8 @@ for pg_version in ${TEST_VERSION_ONLY-14 15 16 17}; do
fi
done
if [ $pg_version -ge 16 ]; then
if [ $pg_version -ge 16 ]
then
echo Enabling trust connection
docker exec $COMPUTE_CONTAINER_NAME bash -c "sed -i '\$d' /var/db/postgres/compute/pg_hba.conf && echo -e 'host\t all\t all\t all\t trust' >> /var/db/postgres/compute/pg_hba.conf && psql $PSQL_OPTION -c 'select pg_reload_conf()' "
echo Adding postgres role
@@ -74,13 +68,10 @@ for pg_version in ${TEST_VERSION_ONLY-14 15 16 17}; do
# The test assumes that it is running on the same host with the postgres engine.
# In our case it's not true, that's why we are copying files to the compute node
TMPDIR=$(mktemp -d)
# Add support for pg_anon for pg_v16
if [ $pg_version -ne 17 ]; then
docker cp $TEST_CONTAINER_NAME:/ext-src/pg_anon-src/data $TMPDIR/data
echo -e '1\t too \t many \t tabs' > $TMPDIR/data/bad.csv
docker cp $TMPDIR/data $COMPUTE_CONTAINER_NAME:/tmp/tmp_anon_alternate_data
docker cp $TEST_CONTAINER_NAME:/ext-src/pg_anon-src/data $TMPDIR/data
echo -e '1\t too \t many \t tabs' > $TMPDIR/data/bad.csv
docker cp $TMPDIR/data $COMPUTE_CONTAINER_NAME:/tmp/tmp_anon_alternate_data
rm -rf $TMPDIR
fi
TMPDIR=$(mktemp -d)
# The following block does the same for the pg_hintplan test
docker cp $TEST_CONTAINER_NAME:/ext-src/pg_hint_plan-src/data $TMPDIR/data
@@ -106,8 +97,4 @@ for pg_version in ${TEST_VERSION_ONLY-14 15 16 17}; do
fi
fi
cleanup
# The support of pg_anon not yet added to PG17, so we have to remove the corresponding option
if [ $pg_version -eq 17 ]; then
mv $SPEC_PATH/spec.bak $SPEC_PATH/spec.json
fi
done

View File

@@ -19,10 +19,6 @@ pub type PgIdent = String;
/// String type alias representing Postgres extension version
pub type ExtVersion = String;
fn default_reconfigure_concurrency() -> usize {
1
}
/// Cluster spec or configuration represented as an optional number of
/// delta operations + final cluster state description.
#[derive(Clone, Debug, Default, Deserialize, Serialize)]
@@ -71,7 +67,7 @@ pub struct ComputeSpec {
pub cluster: Cluster,
pub delta_operations: Option<Vec<DeltaOp>>,
/// An optional hint that can be passed to speed up startup time if we know
/// An optinal hint that can be passed to speed up startup time if we know
/// that no pg catalog mutations (like role creation, database creation,
/// extension creation) need to be done on the actual database to start.
#[serde(default)] // Default false
@@ -90,7 +86,9 @@ pub struct ComputeSpec {
// etc. GUCs in cluster.settings. TODO: Once the control plane has been
// updated to fill these fields, we can make these non optional.
pub tenant_id: Option<TenantId>,
pub timeline_id: Option<TimelineId>,
pub pageserver_connstring: Option<String>,
#[serde(default)]
@@ -115,20 +113,6 @@ pub struct ComputeSpec {
/// Local Proxy configuration used for JWT authentication
#[serde(default)]
pub local_proxy_config: Option<LocalProxySpec>,
/// Number of concurrent connections during the parallel RunInEachDatabase
/// phase of the apply config process.
///
/// We need a higher concurrency during reconfiguration in case of many DBs,
/// but instance is already running and used by client. We can easily get out of
/// `max_connections` limit, and the current code won't handle that.
///
/// Default is 1, but also allow control plane to override this value for specific
/// projects. It's also recommended to bump `superuser_reserved_connections` +=
/// `reconfigure_concurrency` for such projects to ensure that we always have
/// enough spare connections for reconfiguration process to succeed.
#[serde(default = "default_reconfigure_concurrency")]
pub reconfigure_concurrency: usize,
}
/// Feature flag to signal `compute_ctl` to enable certain experimental functionality.
@@ -331,9 +315,6 @@ mod tests {
// Features list defaults to empty vector.
assert!(spec.features.is_empty());
// Reconfigure concurrency defaults to 1.
assert_eq!(spec.reconfigure_concurrency, 1);
}
#[test]

View File

@@ -103,12 +103,11 @@ impl<'a> IdempotencyKey<'a> {
}
}
/// Split into chunks of 1000 metrics to avoid exceeding the max request size.
pub const CHUNK_SIZE: usize = 1000;
// Just a wrapper around a slice of events
// to serialize it as `{"events" : [ ] }
#[derive(Debug, serde::Serialize, serde::Deserialize, PartialEq)]
pub struct EventChunk<'a, T: Clone + PartialEq> {
#[derive(serde::Serialize, Deserialize)]
pub struct EventChunk<'a, T: Clone> {
pub events: std::borrow::Cow<'a, [T]>,
}

View File

@@ -118,8 +118,9 @@ pub struct ConfigToml {
pub virtual_file_io_mode: Option<crate::models::virtual_file::IoMode>,
#[serde(skip_serializing_if = "Option::is_none")]
pub no_sync: Option<bool>,
#[serde(with = "humantime_serde")]
pub server_side_batch_timeout: Option<Duration>,
pub wal_receiver_protocol: PostgresClientProtocol,
pub page_service_pipelining: PageServicePipeliningConfig,
}
#[derive(Debug, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
@@ -136,28 +137,6 @@ pub struct DiskUsageEvictionTaskConfig {
pub eviction_order: EvictionOrder,
}
#[derive(Debug, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
#[serde(tag = "mode", rename_all = "kebab-case")]
#[serde(deny_unknown_fields)]
pub enum PageServicePipeliningConfig {
Serial,
Pipelined(PageServicePipeliningConfigPipelined),
}
#[derive(Debug, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
#[serde(deny_unknown_fields)]
pub struct PageServicePipeliningConfigPipelined {
/// Causes runtime errors if larger than max get_vectored batch size.
pub max_batch_size: NonZeroUsize,
pub execution: PageServiceProtocolPipelinedExecutionStrategy,
}
#[derive(Debug, Clone, Copy, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
#[serde(rename_all = "kebab-case")]
pub enum PageServiceProtocolPipelinedExecutionStrategy {
ConcurrentFutures,
Tasks,
}
pub mod statvfs {
pub mod mock {
#[derive(Debug, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
@@ -353,6 +332,8 @@ pub mod defaults {
pub const DEFAULT_IO_BUFFER_ALIGNMENT: usize = 512;
pub const DEFAULT_SERVER_SIDE_BATCH_TIMEOUT: Option<&str> = None;
pub const DEFAULT_WAL_RECEIVER_PROTOCOL: utils::postgres_client::PostgresClientProtocol =
utils::postgres_client::PostgresClientProtocol::Vanilla;
}
@@ -439,17 +420,11 @@ impl Default for ConfigToml {
ephemeral_bytes_per_memory_kb: (DEFAULT_EPHEMERAL_BYTES_PER_MEMORY_KB),
l0_flush: None,
virtual_file_io_mode: None,
server_side_batch_timeout: DEFAULT_SERVER_SIDE_BATCH_TIMEOUT
.map(|duration| humantime::parse_duration(duration).unwrap()),
tenant_config: TenantConfigToml::default(),
no_sync: None,
wal_receiver_protocol: DEFAULT_WAL_RECEIVER_PROTOCOL,
page_service_pipelining: if !cfg!(test) {
PageServicePipeliningConfig::Serial
} else {
PageServicePipeliningConfig::Pipelined(PageServicePipeliningConfigPipelined {
max_batch_size: NonZeroUsize::new(32).unwrap(),
execution: PageServiceProtocolPipelinedExecutionStrategy::ConcurrentFutures,
})
},
}
}
}

View File

@@ -48,7 +48,7 @@ pub struct TenantCreateResponse {
pub shards: Vec<TenantCreateResponseShard>,
}
#[derive(Serialize, Deserialize, Debug, Clone)]
#[derive(Serialize, Deserialize)]
pub struct NodeRegisterRequest {
pub node_id: NodeId,
@@ -75,7 +75,7 @@ pub struct TenantPolicyRequest {
pub scheduling: Option<ShardSchedulingPolicy>,
}
#[derive(Clone, Serialize, Deserialize, PartialEq, Eq, Hash, Debug)]
#[derive(Clone, Serialize, Deserialize, PartialEq, Eq, Hash)]
pub struct AvailabilityZone(pub String);
impl Display for AvailabilityZone {

View File

@@ -770,11 +770,6 @@ impl Key {
&& self.field6 == 1
}
#[inline(always)]
pub fn is_aux_file_key(&self) -> bool {
self.field1 == AUX_KEY_PREFIX
}
/// Guaranteed to return `Ok()` if [`Self::is_rel_block_key`] returns `true` for `key`.
#[inline(always)]
pub fn to_rel_block(self) -> anyhow::Result<(RelTag, BlockNumber)> {

View File

@@ -501,9 +501,7 @@ pub struct EvictionPolicyLayerAccessThreshold {
#[derive(Debug, Serialize, Deserialize, Clone, PartialEq, Eq)]
pub struct ThrottleConfig {
/// See [`ThrottleConfigTaskKinds`] for why we do the serde `rename`.
#[serde(rename = "task_kinds")]
pub enabled: ThrottleConfigTaskKinds,
pub task_kinds: Vec<String>, // TaskKind
pub initial: u32,
#[serde(with = "humantime_serde")]
pub refill_interval: Duration,
@@ -511,38 +509,10 @@ pub struct ThrottleConfig {
pub max: u32,
}
/// Before <https://github.com/neondatabase/neon/pull/9962>
/// the throttle was a per `Timeline::get`/`Timeline::get_vectored` call.
/// The `task_kinds` field controlled which Pageserver "Task Kind"s
/// were subject to the throttle.
///
/// After that PR, the throttle is applied at pagestream request level
/// and the `task_kinds` field does not apply since the only task kind
/// that us subject to the throttle is that of the page service.
///
/// However, we don't want to make a breaking config change right now
/// because it means we have to migrate all the tenant configs.
/// This will be done in a future PR.
///
/// In the meantime, we use emptiness / non-emptsiness of the `task_kinds`
/// field to determine if the throttle is enabled or not.
#[derive(Serialize, Deserialize, Debug, Clone, PartialEq, Eq)]
#[serde(transparent)]
pub struct ThrottleConfigTaskKinds(Vec<String>);
impl ThrottleConfigTaskKinds {
pub fn disabled() -> Self {
Self(vec![])
}
pub fn is_enabled(&self) -> bool {
!self.0.is_empty()
}
}
impl ThrottleConfig {
pub fn disabled() -> Self {
Self {
enabled: ThrottleConfigTaskKinds::disabled(),
task_kinds: vec![], // effectively disables the throttle
// other values don't matter with emtpy `task_kinds`.
initial: 0,
refill_interval: Duration::from_millis(1),
@@ -556,30 +526,6 @@ impl ThrottleConfig {
}
}
#[cfg(test)]
mod throttle_config_tests {
use super::*;
#[test]
fn test_disabled_is_disabled() {
let config = ThrottleConfig::disabled();
assert!(!config.enabled.is_enabled());
}
#[test]
fn test_enabled_backwards_compat() {
let input = serde_json::json!({
"task_kinds": ["PageRequestHandler"],
"initial": 40000,
"refill_interval": "50ms",
"refill_amount": 1000,
"max": 40000,
"fair": true
});
let config: ThrottleConfig = serde_json::from_value(input).unwrap();
assert!(config.enabled.is_enabled());
}
}
/// A flattened analog of a `pagesever::tenant::LocationMode`, which
/// lists out all possible states (and the virtual "Detached" state)
/// in a flat form rather than using rust-style enums.

View File

@@ -158,8 +158,7 @@ impl ShardIdentity {
key_to_shard_number(self.count, self.stripe_size, key)
}
/// Return true if the key is stored only on this shard. This does not include
/// global keys, see is_key_global().
/// Return true if the key should be ingested by this shard
///
/// Shards must ingest _at least_ keys which return true from this check.
pub fn is_key_local(&self, key: &Key) -> bool {
@@ -171,37 +170,19 @@ impl ShardIdentity {
}
}
/// Return true if the key should be stored on all shards, not just one.
pub fn is_key_global(&self, key: &Key) -> bool {
if key.is_slru_block_key() || key.is_slru_segment_size_key() || key.is_aux_file_key() {
// Special keys that are only stored on shard 0
false
} else if key.is_rel_block_key() {
// Ordinary relation blocks are distributed across shards
false
} else if key.is_rel_size_key() {
// All shards maintain rel size keys (although only shard 0 is responsible for
// keeping it strictly accurate, other shards just reflect the highest block they've ingested)
true
} else {
// For everything else, we assume it must be kept everywhere, because ingest code
// might assume this -- this covers functionality where the ingest code has
// not (yet) been made fully shard aware.
true
}
}
/// Return true if the key should be discarded if found in this shard's
/// data store, e.g. during compaction after a split.
///
/// Shards _may_ drop keys which return false here, but are not obliged to.
pub fn is_key_disposable(&self, key: &Key) -> bool {
if self.count < ShardCount(2) {
// Fast path: unsharded tenant doesn't dispose of anything
return false;
}
if self.is_key_global(key) {
if key_is_shard0(key) {
// Q: Why can't we dispose of shard0 content if we're not shard 0?
// A1: because the WAL ingestion logic currently ingests some shard 0
// content on all shards, even though it's only read on shard 0. If we
// dropped it, then subsequent WAL ingest to these keys would encounter
// an error.
// A2: because key_is_shard0 also covers relation size keys, which are written
// on all shards even though they're only maintained accurately on shard 0.
false
} else {
!self.is_key_local(key)

View File

@@ -100,7 +100,7 @@ impl StartupMessageParamsBuilder {
#[derive(Debug, Clone, Default)]
pub struct StartupMessageParams {
pub params: Bytes,
params: Bytes,
}
impl StartupMessageParams {
@@ -565,8 +565,6 @@ pub enum BeMessage<'a> {
/// Batch of interpreted, shard filtered WAL records,
/// ready for the pageserver to ingest
InterpretedWalRecords(InterpretedWalRecordsBody<'a>),
Raw(u8, &'a [u8]),
}
/// Common shorthands.
@@ -756,10 +754,6 @@ impl BeMessage<'_> {
/// one more buffer.
pub fn write(buf: &mut BytesMut, message: &BeMessage) -> Result<(), ProtocolError> {
match message {
BeMessage::Raw(code, data) => {
buf.put_u8(*code);
write_body(buf, |b| b.put_slice(data))
}
BeMessage::AuthenticationOk => {
buf.put_u8(b'R');
write_body(buf, |buf| {

View File

@@ -1,6 +0,0 @@
This directory contains libraries that are specific for proxy.
Currently, it contains a signficant fork/refactoring of rust-postgres that no longer reflects the API
of the original library. Since it was so significant, it made sense to upgrade it to it's own set of libraries.
Proxy needs unique access to the protocol, which explains why such heavy modifications were necessary.

View File

@@ -1,20 +0,0 @@
[package]
name = "postgres-protocol2"
version = "0.1.0"
edition = "2018"
license = "MIT/Apache-2.0"
[dependencies]
base64 = "0.20"
byteorder.workspace = true
bytes.workspace = true
fallible-iterator.workspace = true
hmac.workspace = true
memchr = "2.0"
rand.workspace = true
sha2.workspace = true
stringprep = "0.1"
tokio = { workspace = true, features = ["rt"] }
[dev-dependencies]
tokio = { workspace = true, features = ["full"] }

View File

@@ -1,2 +0,0 @@
//! Authentication protocol support.
pub mod sasl;

View File

@@ -1,516 +0,0 @@
//! SASL-based authentication support.
use hmac::{Hmac, Mac};
use rand::{self, Rng};
use sha2::digest::FixedOutput;
use sha2::{Digest, Sha256};
use std::fmt::Write;
use std::io;
use std::iter;
use std::mem;
use std::str;
use tokio::task::yield_now;
const NONCE_LENGTH: usize = 24;
/// The identifier of the SCRAM-SHA-256 SASL authentication mechanism.
pub const SCRAM_SHA_256: &str = "SCRAM-SHA-256";
/// The identifier of the SCRAM-SHA-256-PLUS SASL authentication mechanism.
pub const SCRAM_SHA_256_PLUS: &str = "SCRAM-SHA-256-PLUS";
// since postgres passwords are not required to exclude saslprep-prohibited
// characters or even be valid UTF8, we run saslprep if possible and otherwise
// return the raw password.
fn normalize(pass: &[u8]) -> Vec<u8> {
let pass = match str::from_utf8(pass) {
Ok(pass) => pass,
Err(_) => return pass.to_vec(),
};
match stringprep::saslprep(pass) {
Ok(pass) => pass.into_owned().into_bytes(),
Err(_) => pass.as_bytes().to_vec(),
}
}
pub(crate) async fn hi(str: &[u8], salt: &[u8], iterations: u32) -> [u8; 32] {
let mut hmac =
Hmac::<Sha256>::new_from_slice(str).expect("HMAC is able to accept all key sizes");
hmac.update(salt);
hmac.update(&[0, 0, 0, 1]);
let mut prev = hmac.finalize().into_bytes();
let mut hi = prev;
for i in 1..iterations {
let mut hmac = Hmac::<Sha256>::new_from_slice(str).expect("already checked above");
hmac.update(&prev);
prev = hmac.finalize().into_bytes();
for (hi, prev) in hi.iter_mut().zip(prev) {
*hi ^= prev;
}
// yield every ~250us
// hopefully reduces tail latencies
if i % 1024 == 0 {
yield_now().await
}
}
hi.into()
}
enum ChannelBindingInner {
Unrequested,
Unsupported,
TlsServerEndPoint(Vec<u8>),
}
/// The channel binding configuration for a SCRAM authentication exchange.
pub struct ChannelBinding(ChannelBindingInner);
impl ChannelBinding {
/// The server did not request channel binding.
pub fn unrequested() -> ChannelBinding {
ChannelBinding(ChannelBindingInner::Unrequested)
}
/// The server requested channel binding but the client is unable to provide it.
pub fn unsupported() -> ChannelBinding {
ChannelBinding(ChannelBindingInner::Unsupported)
}
/// The server requested channel binding and the client will use the `tls-server-end-point`
/// method.
pub fn tls_server_end_point(signature: Vec<u8>) -> ChannelBinding {
ChannelBinding(ChannelBindingInner::TlsServerEndPoint(signature))
}
fn gs2_header(&self) -> &'static str {
match self.0 {
ChannelBindingInner::Unrequested => "y,,",
ChannelBindingInner::Unsupported => "n,,",
ChannelBindingInner::TlsServerEndPoint(_) => "p=tls-server-end-point,,",
}
}
fn cbind_data(&self) -> &[u8] {
match self.0 {
ChannelBindingInner::Unrequested | ChannelBindingInner::Unsupported => &[],
ChannelBindingInner::TlsServerEndPoint(ref buf) => buf,
}
}
}
/// A pair of keys for the SCRAM-SHA-256 mechanism.
/// See <https://datatracker.ietf.org/doc/html/rfc5802#section-3> for details.
#[derive(Debug, Clone, Copy, PartialEq, Eq)]
pub struct ScramKeys<const N: usize> {
/// Used by server to authenticate client.
pub client_key: [u8; N],
/// Used by client to verify server's signature.
pub server_key: [u8; N],
}
/// Password or keys which were derived from it.
enum Credentials<const N: usize> {
/// A regular password as a vector of bytes.
Password(Vec<u8>),
/// A precomputed pair of keys.
Keys(ScramKeys<N>),
}
enum State {
Update {
nonce: String,
password: Credentials<32>,
channel_binding: ChannelBinding,
},
Finish {
server_key: [u8; 32],
auth_message: String,
},
Done,
}
/// A type which handles the client side of the SCRAM-SHA-256/SCRAM-SHA-256-PLUS authentication
/// process.
///
/// During the authentication process, if the backend sends an `AuthenticationSASL` message which
/// includes `SCRAM-SHA-256` as an authentication mechanism, this type can be used.
///
/// After a `ScramSha256` is constructed, the buffer returned by the `message()` method should be
/// sent to the backend in a `SASLInitialResponse` message along with the mechanism name.
///
/// The server will reply with an `AuthenticationSASLContinue` message. Its contents should be
/// passed to the `update()` method, after which the buffer returned by the `message()` method
/// should be sent to the backend in a `SASLResponse` message.
///
/// The server will reply with an `AuthenticationSASLFinal` message. Its contents should be passed
/// to the `finish()` method, after which the authentication process is complete.
pub struct ScramSha256 {
message: String,
state: State,
}
fn nonce() -> String {
// rand 0.5's ThreadRng is cryptographically secure
let mut rng = rand::thread_rng();
(0..NONCE_LENGTH)
.map(|_| {
let mut v = rng.gen_range(0x21u8..0x7e);
if v == 0x2c {
v = 0x7e
}
v as char
})
.collect()
}
impl ScramSha256 {
/// Constructs a new instance which will use the provided password for authentication.
pub fn new(password: &[u8], channel_binding: ChannelBinding) -> ScramSha256 {
let password = Credentials::Password(normalize(password));
ScramSha256::new_inner(password, channel_binding, nonce())
}
/// Constructs a new instance which will use the provided key pair for authentication.
pub fn new_with_keys(keys: ScramKeys<32>, channel_binding: ChannelBinding) -> ScramSha256 {
let password = Credentials::Keys(keys);
ScramSha256::new_inner(password, channel_binding, nonce())
}
fn new_inner(
password: Credentials<32>,
channel_binding: ChannelBinding,
nonce: String,
) -> ScramSha256 {
ScramSha256 {
message: format!("{}n=,r={}", channel_binding.gs2_header(), nonce),
state: State::Update {
nonce,
password,
channel_binding,
},
}
}
/// Returns the message which should be sent to the backend in an `SASLResponse` message.
pub fn message(&self) -> &[u8] {
if let State::Done = self.state {
panic!("invalid SCRAM state");
}
self.message.as_bytes()
}
/// Updates the state machine with the response from the backend.
///
/// This should be called when an `AuthenticationSASLContinue` message is received.
pub async fn update(&mut self, message: &[u8]) -> io::Result<()> {
let (client_nonce, password, channel_binding) =
match mem::replace(&mut self.state, State::Done) {
State::Update {
nonce,
password,
channel_binding,
} => (nonce, password, channel_binding),
_ => return Err(io::Error::new(io::ErrorKind::Other, "invalid SCRAM state")),
};
let message =
str::from_utf8(message).map_err(|e| io::Error::new(io::ErrorKind::InvalidInput, e))?;
let parsed = Parser::new(message).server_first_message()?;
if !parsed.nonce.starts_with(&client_nonce) {
return Err(io::Error::new(io::ErrorKind::InvalidInput, "invalid nonce"));
}
let (client_key, server_key) = match password {
Credentials::Password(password) => {
let salt = match base64::decode(parsed.salt) {
Ok(salt) => salt,
Err(e) => return Err(io::Error::new(io::ErrorKind::InvalidInput, e)),
};
let salted_password = hi(&password, &salt, parsed.iteration_count).await;
let make_key = |name| {
let mut hmac = Hmac::<Sha256>::new_from_slice(&salted_password)
.expect("HMAC is able to accept all key sizes");
hmac.update(name);
let mut key = [0u8; 32];
key.copy_from_slice(hmac.finalize().into_bytes().as_slice());
key
};
(make_key(b"Client Key"), make_key(b"Server Key"))
}
Credentials::Keys(keys) => (keys.client_key, keys.server_key),
};
let mut hash = Sha256::default();
hash.update(client_key);
let stored_key = hash.finalize_fixed();
let mut cbind_input = vec![];
cbind_input.extend(channel_binding.gs2_header().as_bytes());
cbind_input.extend(channel_binding.cbind_data());
let cbind_input = base64::encode(&cbind_input);
self.message.clear();
write!(&mut self.message, "c={},r={}", cbind_input, parsed.nonce).unwrap();
let auth_message = format!("n=,r={},{},{}", client_nonce, message, self.message);
let mut hmac = Hmac::<Sha256>::new_from_slice(&stored_key)
.expect("HMAC is able to accept all key sizes");
hmac.update(auth_message.as_bytes());
let client_signature = hmac.finalize().into_bytes();
let mut client_proof = client_key;
for (proof, signature) in client_proof.iter_mut().zip(client_signature) {
*proof ^= signature;
}
write!(&mut self.message, ",p={}", base64::encode(client_proof)).unwrap();
self.state = State::Finish {
server_key,
auth_message,
};
Ok(())
}
/// Finalizes the authentication process.
///
/// This should be called when the backend sends an `AuthenticationSASLFinal` message.
/// Authentication has only succeeded if this method returns `Ok(())`.
pub fn finish(&mut self, message: &[u8]) -> io::Result<()> {
let (server_key, auth_message) = match mem::replace(&mut self.state, State::Done) {
State::Finish {
server_key,
auth_message,
} => (server_key, auth_message),
_ => return Err(io::Error::new(io::ErrorKind::Other, "invalid SCRAM state")),
};
let message =
str::from_utf8(message).map_err(|e| io::Error::new(io::ErrorKind::InvalidInput, e))?;
let parsed = Parser::new(message).server_final_message()?;
let verifier = match parsed {
ServerFinalMessage::Error(e) => {
return Err(io::Error::new(
io::ErrorKind::Other,
format!("SCRAM error: {}", e),
));
}
ServerFinalMessage::Verifier(verifier) => verifier,
};
let verifier = match base64::decode(verifier) {
Ok(verifier) => verifier,
Err(e) => return Err(io::Error::new(io::ErrorKind::InvalidInput, e)),
};
let mut hmac = Hmac::<Sha256>::new_from_slice(&server_key)
.expect("HMAC is able to accept all key sizes");
hmac.update(auth_message.as_bytes());
hmac.verify_slice(&verifier)
.map_err(|_| io::Error::new(io::ErrorKind::InvalidInput, "SCRAM verification error"))
}
}
struct Parser<'a> {
s: &'a str,
it: iter::Peekable<str::CharIndices<'a>>,
}
impl<'a> Parser<'a> {
fn new(s: &'a str) -> Parser<'a> {
Parser {
s,
it: s.char_indices().peekable(),
}
}
fn eat(&mut self, target: char) -> io::Result<()> {
match self.it.next() {
Some((_, c)) if c == target => Ok(()),
Some((i, c)) => {
let m = format!(
"unexpected character at byte {}: expected `{}` but got `{}",
i, target, c
);
Err(io::Error::new(io::ErrorKind::InvalidInput, m))
}
None => Err(io::Error::new(
io::ErrorKind::UnexpectedEof,
"unexpected EOF",
)),
}
}
fn take_while<F>(&mut self, f: F) -> io::Result<&'a str>
where
F: Fn(char) -> bool,
{
let start = match self.it.peek() {
Some(&(i, _)) => i,
None => return Ok(""),
};
loop {
match self.it.peek() {
Some(&(_, c)) if f(c) => {
self.it.next();
}
Some(&(i, _)) => return Ok(&self.s[start..i]),
None => return Ok(&self.s[start..]),
}
}
}
fn printable(&mut self) -> io::Result<&'a str> {
self.take_while(|c| matches!(c, '\x21'..='\x2b' | '\x2d'..='\x7e'))
}
fn nonce(&mut self) -> io::Result<&'a str> {
self.eat('r')?;
self.eat('=')?;
self.printable()
}
fn base64(&mut self) -> io::Result<&'a str> {
self.take_while(|c| matches!(c, 'a'..='z' | 'A'..='Z' | '0'..='9' | '/' | '+' | '='))
}
fn salt(&mut self) -> io::Result<&'a str> {
self.eat('s')?;
self.eat('=')?;
self.base64()
}
fn posit_number(&mut self) -> io::Result<u32> {
let n = self.take_while(|c| c.is_ascii_digit())?;
n.parse()
.map_err(|e| io::Error::new(io::ErrorKind::InvalidInput, e))
}
fn iteration_count(&mut self) -> io::Result<u32> {
self.eat('i')?;
self.eat('=')?;
self.posit_number()
}
fn eof(&mut self) -> io::Result<()> {
match self.it.peek() {
Some(&(i, _)) => Err(io::Error::new(
io::ErrorKind::InvalidInput,
format!("unexpected trailing data at byte {}", i),
)),
None => Ok(()),
}
}
fn server_first_message(&mut self) -> io::Result<ServerFirstMessage<'a>> {
let nonce = self.nonce()?;
self.eat(',')?;
let salt = self.salt()?;
self.eat(',')?;
let iteration_count = self.iteration_count()?;
self.eof()?;
Ok(ServerFirstMessage {
nonce,
salt,
iteration_count,
})
}
fn value(&mut self) -> io::Result<&'a str> {
self.take_while(|c| matches!(c, '\0' | '=' | ','))
}
fn server_error(&mut self) -> io::Result<Option<&'a str>> {
match self.it.peek() {
Some(&(_, 'e')) => {}
_ => return Ok(None),
}
self.eat('e')?;
self.eat('=')?;
self.value().map(Some)
}
fn verifier(&mut self) -> io::Result<&'a str> {
self.eat('v')?;
self.eat('=')?;
self.base64()
}
fn server_final_message(&mut self) -> io::Result<ServerFinalMessage<'a>> {
let message = match self.server_error()? {
Some(error) => ServerFinalMessage::Error(error),
None => ServerFinalMessage::Verifier(self.verifier()?),
};
self.eof()?;
Ok(message)
}
}
struct ServerFirstMessage<'a> {
nonce: &'a str,
salt: &'a str,
iteration_count: u32,
}
enum ServerFinalMessage<'a> {
Error(&'a str),
Verifier(&'a str),
}
#[cfg(test)]
mod test {
use super::*;
#[test]
fn parse_server_first_message() {
let message = "r=fyko+d2lbbFgONRv9qkxdawL3rfcNHYJY1ZVvWVs7j,s=QSXCR+Q6sek8bf92,i=4096";
let message = Parser::new(message).server_first_message().unwrap();
assert_eq!(message.nonce, "fyko+d2lbbFgONRv9qkxdawL3rfcNHYJY1ZVvWVs7j");
assert_eq!(message.salt, "QSXCR+Q6sek8bf92");
assert_eq!(message.iteration_count, 4096);
}
// recorded auth exchange from psql
#[tokio::test]
async fn exchange() {
let password = "foobar";
let nonce = "9IZ2O01zb9IgiIZ1WJ/zgpJB";
let client_first = "n,,n=,r=9IZ2O01zb9IgiIZ1WJ/zgpJB";
let server_first =
"r=9IZ2O01zb9IgiIZ1WJ/zgpJBjx/oIRLs02gGSHcw1KEty3eY,s=fs3IXBy7U7+IvVjZ,i\
=4096";
let client_final =
"c=biws,r=9IZ2O01zb9IgiIZ1WJ/zgpJBjx/oIRLs02gGSHcw1KEty3eY,p=AmNKosjJzS3\
1NTlQYNs5BTeQjdHdk7lOflDo5re2an8=";
let server_final = "v=U+ppxD5XUKtradnv8e2MkeupiA8FU87Sg8CXzXHDAzw=";
let mut scram = ScramSha256::new_inner(
Credentials::Password(normalize(password.as_bytes())),
ChannelBinding::unsupported(),
nonce.to_string(),
);
assert_eq!(str::from_utf8(scram.message()).unwrap(), client_first);
scram.update(server_first.as_bytes()).await.unwrap();
assert_eq!(str::from_utf8(scram.message()).unwrap(), client_final);
scram.finish(server_final.as_bytes()).unwrap();
}
}

View File

@@ -1,93 +0,0 @@
//! Provides functions for escaping literals and identifiers for use
//! in SQL queries.
//!
//! Prefer parameterized queries where possible. Do not escape
//! parameters in a parameterized query.
#[cfg(test)]
mod test;
/// Escape a literal and surround result with single quotes. Not
/// recommended in most cases.
///
/// If input contains backslashes, result will be of the form `
/// E'...'` so it is safe to use regardless of the setting of
/// standard_conforming_strings.
pub fn escape_literal(input: &str) -> String {
escape_internal(input, false)
}
/// Escape an identifier and surround result with double quotes.
pub fn escape_identifier(input: &str) -> String {
escape_internal(input, true)
}
// Translation of PostgreSQL libpq's PQescapeInternal(). Does not
// require a connection because input string is known to be valid
// UTF-8.
//
// Escape arbitrary strings. If as_ident is true, we escape the
// result as an identifier; if false, as a literal. The result is
// returned in a newly allocated buffer. If we fail due to an
// encoding violation or out of memory condition, we return NULL,
// storing an error message into conn.
fn escape_internal(input: &str, as_ident: bool) -> String {
let mut num_backslashes = 0;
let mut num_quotes = 0;
let quote_char = if as_ident { '"' } else { '\'' };
// Scan the string for characters that must be escaped.
for ch in input.chars() {
if ch == quote_char {
num_quotes += 1;
} else if ch == '\\' {
num_backslashes += 1;
}
}
// Allocate output String.
let mut result_size = input.len() + num_quotes + 3; // two quotes, plus a NUL
if !as_ident && num_backslashes > 0 {
result_size += num_backslashes + 2;
}
let mut output = String::with_capacity(result_size);
// If we are escaping a literal that contains backslashes, we use
// the escape string syntax so that the result is correct under
// either value of standard_conforming_strings. We also emit a
// leading space in this case, to guard against the possibility
// that the result might be interpolated immediately following an
// identifier.
if !as_ident && num_backslashes > 0 {
output.push(' ');
output.push('E');
}
// Opening quote.
output.push(quote_char);
// Use fast path if possible.
//
// We've already verified that the input string is well-formed in
// the current encoding. If it contains no quotes and, in the
// case of literal-escaping, no backslashes, then we can just copy
// it directly to the output buffer, adding the necessary quotes.
//
// If not, we must rescan the input and process each character
// individually.
if num_quotes == 0 && (num_backslashes == 0 || as_ident) {
output.push_str(input);
} else {
for ch in input.chars() {
if ch == quote_char || (!as_ident && ch == '\\') {
output.push(ch);
}
output.push(ch);
}
}
output.push(quote_char);
output
}

View File

@@ -1,17 +0,0 @@
use crate::escape::{escape_identifier, escape_literal};
#[test]
fn test_escape_idenifier() {
assert_eq!(escape_identifier("foo"), String::from("\"foo\""));
assert_eq!(escape_identifier("f\\oo"), String::from("\"f\\oo\""));
assert_eq!(escape_identifier("f'oo"), String::from("\"f'oo\""));
assert_eq!(escape_identifier("f\"oo"), String::from("\"f\"\"oo\""));
}
#[test]
fn test_escape_literal() {
assert_eq!(escape_literal("foo"), String::from("'foo'"));
assert_eq!(escape_literal("f\\oo"), String::from(" E'f\\\\oo'"));
assert_eq!(escape_literal("f'oo"), String::from("'f''oo'"));
assert_eq!(escape_literal("f\"oo"), String::from("'f\"oo'"));
}

View File

@@ -1,78 +0,0 @@
//! Low level Postgres protocol APIs.
//!
//! This crate implements the low level components of Postgres's communication
//! protocol, including message and value serialization and deserialization.
//! It is designed to be used as a building block by higher level APIs such as
//! `rust-postgres`, and should not typically be used directly.
//!
//! # Note
//!
//! This library assumes that the `client_encoding` backend parameter has been
//! set to `UTF8`. It will most likely not behave properly if that is not the case.
#![doc(html_root_url = "https://docs.rs/postgres-protocol/0.6")]
#![warn(missing_docs, rust_2018_idioms, clippy::all)]
use byteorder::{BigEndian, ByteOrder};
use bytes::{BufMut, BytesMut};
use std::io;
pub mod authentication;
pub mod escape;
pub mod message;
pub mod password;
pub mod types;
/// A Postgres OID.
pub type Oid = u32;
/// A Postgres Log Sequence Number (LSN).
pub type Lsn = u64;
/// An enum indicating if a value is `NULL` or not.
pub enum IsNull {
/// The value is `NULL`.
Yes,
/// The value is not `NULL`.
No,
}
fn write_nullable<F, E>(serializer: F, buf: &mut BytesMut) -> Result<(), E>
where
F: FnOnce(&mut BytesMut) -> Result<IsNull, E>,
E: From<io::Error>,
{
let base = buf.len();
buf.put_i32(0);
let size = match serializer(buf)? {
IsNull::No => i32::from_usize(buf.len() - base - 4)?,
IsNull::Yes => -1,
};
BigEndian::write_i32(&mut buf[base..], size);
Ok(())
}
trait FromUsize: Sized {
fn from_usize(x: usize) -> Result<Self, io::Error>;
}
macro_rules! from_usize {
($t:ty) => {
impl FromUsize for $t {
#[inline]
fn from_usize(x: usize) -> io::Result<$t> {
if x > <$t>::MAX as usize {
Err(io::Error::new(
io::ErrorKind::InvalidInput,
"value too large to transmit",
))
} else {
Ok(x as $t)
}
}
}
};
}
from_usize!(i16);
from_usize!(i32);

View File

@@ -1,766 +0,0 @@
#![allow(missing_docs)]
use byteorder::{BigEndian, ByteOrder, ReadBytesExt};
use bytes::{Bytes, BytesMut};
use fallible_iterator::FallibleIterator;
use memchr::memchr;
use std::cmp;
use std::io::{self, Read};
use std::ops::Range;
use std::str;
use crate::Oid;
// top-level message tags
const PARSE_COMPLETE_TAG: u8 = b'1';
const BIND_COMPLETE_TAG: u8 = b'2';
const CLOSE_COMPLETE_TAG: u8 = b'3';
pub const NOTIFICATION_RESPONSE_TAG: u8 = b'A';
const COPY_DONE_TAG: u8 = b'c';
const COMMAND_COMPLETE_TAG: u8 = b'C';
const COPY_DATA_TAG: u8 = b'd';
const DATA_ROW_TAG: u8 = b'D';
const ERROR_RESPONSE_TAG: u8 = b'E';
const COPY_IN_RESPONSE_TAG: u8 = b'G';
const COPY_OUT_RESPONSE_TAG: u8 = b'H';
const COPY_BOTH_RESPONSE_TAG: u8 = b'W';
const EMPTY_QUERY_RESPONSE_TAG: u8 = b'I';
const BACKEND_KEY_DATA_TAG: u8 = b'K';
pub const NO_DATA_TAG: u8 = b'n';
pub const NOTICE_RESPONSE_TAG: u8 = b'N';
const AUTHENTICATION_TAG: u8 = b'R';
const PORTAL_SUSPENDED_TAG: u8 = b's';
pub const PARAMETER_STATUS_TAG: u8 = b'S';
const PARAMETER_DESCRIPTION_TAG: u8 = b't';
const ROW_DESCRIPTION_TAG: u8 = b'T';
pub const READY_FOR_QUERY_TAG: u8 = b'Z';
#[derive(Debug, Copy, Clone)]
pub struct Header {
tag: u8,
len: i32,
}
#[allow(clippy::len_without_is_empty)]
impl Header {
#[inline]
pub fn parse(buf: &[u8]) -> io::Result<Option<Header>> {
if buf.len() < 5 {
return Ok(None);
}
let tag = buf[0];
let len = BigEndian::read_i32(&buf[1..]);
if len < 4 {
return Err(io::Error::new(
io::ErrorKind::InvalidData,
"invalid message length: header length < 4",
));
}
Ok(Some(Header { tag, len }))
}
#[inline]
pub fn tag(self) -> u8 {
self.tag
}
#[inline]
pub fn len(self) -> i32 {
self.len
}
}
/// An enum representing Postgres backend messages.
#[non_exhaustive]
pub enum Message {
AuthenticationCleartextPassword,
AuthenticationGss,
AuthenticationKerberosV5,
AuthenticationMd5Password,
AuthenticationOk,
AuthenticationScmCredential,
AuthenticationSspi,
AuthenticationGssContinue,
AuthenticationSasl(AuthenticationSaslBody),
AuthenticationSaslContinue(AuthenticationSaslContinueBody),
AuthenticationSaslFinal(AuthenticationSaslFinalBody),
BackendKeyData(BackendKeyDataBody),
BindComplete,
CloseComplete,
CommandComplete(CommandCompleteBody),
CopyData,
CopyDone,
CopyInResponse,
CopyOutResponse,
CopyBothResponse,
DataRow(DataRowBody),
EmptyQueryResponse,
ErrorResponse(ErrorResponseBody),
NoData,
NoticeResponse(NoticeResponseBody),
NotificationResponse(NotificationResponseBody),
ParameterDescription(ParameterDescriptionBody),
ParameterStatus(ParameterStatusBody),
ParseComplete,
PortalSuspended,
ReadyForQuery(ReadyForQueryBody),
RowDescription(RowDescriptionBody),
}
impl Message {
#[inline]
pub fn parse(buf: &mut BytesMut) -> io::Result<Option<Message>> {
if buf.len() < 5 {
let to_read = 5 - buf.len();
buf.reserve(to_read);
return Ok(None);
}
let tag = buf[0];
let len = (&buf[1..5]).read_u32::<BigEndian>().unwrap();
if len < 4 {
return Err(io::Error::new(
io::ErrorKind::InvalidInput,
"invalid message length: parsing u32",
));
}
let total_len = len as usize + 1;
if buf.len() < total_len {
let to_read = total_len - buf.len();
buf.reserve(to_read);
return Ok(None);
}
let mut buf = Buffer {
bytes: buf.split_to(total_len).freeze(),
idx: 5,
};
let message = match tag {
PARSE_COMPLETE_TAG => Message::ParseComplete,
BIND_COMPLETE_TAG => Message::BindComplete,
CLOSE_COMPLETE_TAG => Message::CloseComplete,
NOTIFICATION_RESPONSE_TAG => {
let process_id = buf.read_i32::<BigEndian>()?;
let channel = buf.read_cstr()?;
let message = buf.read_cstr()?;
Message::NotificationResponse(NotificationResponseBody {
process_id,
channel,
message,
})
}
COPY_DONE_TAG => Message::CopyDone,
COMMAND_COMPLETE_TAG => {
let tag = buf.read_cstr()?;
Message::CommandComplete(CommandCompleteBody { tag })
}
COPY_DATA_TAG => Message::CopyData,
DATA_ROW_TAG => {
let len = buf.read_u16::<BigEndian>()?;
let storage = buf.read_all();
Message::DataRow(DataRowBody { storage, len })
}
ERROR_RESPONSE_TAG => {
let storage = buf.read_all();
Message::ErrorResponse(ErrorResponseBody { storage })
}
COPY_IN_RESPONSE_TAG => Message::CopyInResponse,
COPY_OUT_RESPONSE_TAG => Message::CopyOutResponse,
COPY_BOTH_RESPONSE_TAG => Message::CopyBothResponse,
EMPTY_QUERY_RESPONSE_TAG => Message::EmptyQueryResponse,
BACKEND_KEY_DATA_TAG => {
let process_id = buf.read_i32::<BigEndian>()?;
let secret_key = buf.read_i32::<BigEndian>()?;
Message::BackendKeyData(BackendKeyDataBody {
process_id,
secret_key,
})
}
NO_DATA_TAG => Message::NoData,
NOTICE_RESPONSE_TAG => {
let storage = buf.read_all();
Message::NoticeResponse(NoticeResponseBody { storage })
}
AUTHENTICATION_TAG => match buf.read_i32::<BigEndian>()? {
0 => Message::AuthenticationOk,
2 => Message::AuthenticationKerberosV5,
3 => Message::AuthenticationCleartextPassword,
5 => Message::AuthenticationMd5Password,
6 => Message::AuthenticationScmCredential,
7 => Message::AuthenticationGss,
8 => Message::AuthenticationGssContinue,
9 => Message::AuthenticationSspi,
10 => {
let storage = buf.read_all();
Message::AuthenticationSasl(AuthenticationSaslBody(storage))
}
11 => {
let storage = buf.read_all();
Message::AuthenticationSaslContinue(AuthenticationSaslContinueBody(storage))
}
12 => {
let storage = buf.read_all();
Message::AuthenticationSaslFinal(AuthenticationSaslFinalBody(storage))
}
tag => {
return Err(io::Error::new(
io::ErrorKind::InvalidInput,
format!("unknown authentication tag `{}`", tag),
));
}
},
PORTAL_SUSPENDED_TAG => Message::PortalSuspended,
PARAMETER_STATUS_TAG => {
let name = buf.read_cstr()?;
let value = buf.read_cstr()?;
Message::ParameterStatus(ParameterStatusBody { name, value })
}
PARAMETER_DESCRIPTION_TAG => {
let len = buf.read_u16::<BigEndian>()?;
let storage = buf.read_all();
Message::ParameterDescription(ParameterDescriptionBody { storage, len })
}
ROW_DESCRIPTION_TAG => {
let len = buf.read_u16::<BigEndian>()?;
let storage = buf.read_all();
Message::RowDescription(RowDescriptionBody { storage, len })
}
READY_FOR_QUERY_TAG => {
let status = buf.read_u8()?;
Message::ReadyForQuery(ReadyForQueryBody { status })
}
tag => {
return Err(io::Error::new(
io::ErrorKind::InvalidInput,
format!("unknown message tag `{}`", tag),
));
}
};
if !buf.is_empty() {
return Err(io::Error::new(
io::ErrorKind::InvalidInput,
"invalid message length: expected buffer to be empty",
));
}
Ok(Some(message))
}
}
struct Buffer {
bytes: Bytes,
idx: usize,
}
impl Buffer {
#[inline]
fn slice(&self) -> &[u8] {
&self.bytes[self.idx..]
}
#[inline]
fn is_empty(&self) -> bool {
self.slice().is_empty()
}
#[inline]
fn read_cstr(&mut self) -> io::Result<Bytes> {
match memchr(0, self.slice()) {
Some(pos) => {
let start = self.idx;
let end = start + pos;
let cstr = self.bytes.slice(start..end);
self.idx = end + 1;
Ok(cstr)
}
None => Err(io::Error::new(
io::ErrorKind::UnexpectedEof,
"unexpected EOF",
)),
}
}
#[inline]
fn read_all(&mut self) -> Bytes {
let buf = self.bytes.slice(self.idx..);
self.idx = self.bytes.len();
buf
}
}
impl Read for Buffer {
#[inline]
fn read(&mut self, buf: &mut [u8]) -> io::Result<usize> {
let len = {
let slice = self.slice();
let len = cmp::min(slice.len(), buf.len());
buf[..len].copy_from_slice(&slice[..len]);
len
};
self.idx += len;
Ok(len)
}
}
pub struct AuthenticationMd5PasswordBody {
salt: [u8; 4],
}
impl AuthenticationMd5PasswordBody {
#[inline]
pub fn salt(&self) -> [u8; 4] {
self.salt
}
}
pub struct AuthenticationSaslBody(Bytes);
impl AuthenticationSaslBody {
#[inline]
pub fn mechanisms(&self) -> SaslMechanisms<'_> {
SaslMechanisms(&self.0)
}
}
pub struct SaslMechanisms<'a>(&'a [u8]);
impl<'a> FallibleIterator for SaslMechanisms<'a> {
type Item = &'a str;
type Error = io::Error;
#[inline]
fn next(&mut self) -> io::Result<Option<&'a str>> {
let value_end = find_null(self.0, 0)?;
if value_end == 0 {
if self.0.len() != 1 {
return Err(io::Error::new(
io::ErrorKind::InvalidData,
"invalid message length: expected to be at end of iterator for sasl",
));
}
Ok(None)
} else {
let value = get_str(&self.0[..value_end])?;
self.0 = &self.0[value_end + 1..];
Ok(Some(value))
}
}
}
pub struct AuthenticationSaslContinueBody(Bytes);
impl AuthenticationSaslContinueBody {
#[inline]
pub fn data(&self) -> &[u8] {
&self.0
}
}
pub struct AuthenticationSaslFinalBody(Bytes);
impl AuthenticationSaslFinalBody {
#[inline]
pub fn data(&self) -> &[u8] {
&self.0
}
}
pub struct BackendKeyDataBody {
process_id: i32,
secret_key: i32,
}
impl BackendKeyDataBody {
#[inline]
pub fn process_id(&self) -> i32 {
self.process_id
}
#[inline]
pub fn secret_key(&self) -> i32 {
self.secret_key
}
}
pub struct CommandCompleteBody {
tag: Bytes,
}
impl CommandCompleteBody {
#[inline]
pub fn tag(&self) -> io::Result<&str> {
get_str(&self.tag)
}
}
#[derive(Debug)]
pub struct DataRowBody {
storage: Bytes,
len: u16,
}
impl DataRowBody {
#[inline]
pub fn ranges(&self) -> DataRowRanges<'_> {
DataRowRanges {
buf: &self.storage,
len: self.storage.len(),
remaining: self.len,
}
}
#[inline]
pub fn buffer(&self) -> &[u8] {
&self.storage
}
}
pub struct DataRowRanges<'a> {
buf: &'a [u8],
len: usize,
remaining: u16,
}
impl FallibleIterator for DataRowRanges<'_> {
type Item = Option<Range<usize>>;
type Error = io::Error;
#[inline]
fn next(&mut self) -> io::Result<Option<Option<Range<usize>>>> {
if self.remaining == 0 {
if self.buf.is_empty() {
return Ok(None);
} else {
return Err(io::Error::new(
io::ErrorKind::InvalidInput,
"invalid message length: datarowrange is not empty",
));
}
}
self.remaining -= 1;
let len = self.buf.read_i32::<BigEndian>()?;
if len < 0 {
Ok(Some(None))
} else {
let len = len as usize;
if self.buf.len() < len {
return Err(io::Error::new(
io::ErrorKind::UnexpectedEof,
"unexpected EOF",
));
}
let base = self.len - self.buf.len();
self.buf = &self.buf[len..];
Ok(Some(Some(base..base + len)))
}
}
#[inline]
fn size_hint(&self) -> (usize, Option<usize>) {
let len = self.remaining as usize;
(len, Some(len))
}
}
pub struct ErrorResponseBody {
storage: Bytes,
}
impl ErrorResponseBody {
#[inline]
pub fn fields(&self) -> ErrorFields<'_> {
ErrorFields { buf: &self.storage }
}
}
pub struct ErrorFields<'a> {
buf: &'a [u8],
}
impl<'a> FallibleIterator for ErrorFields<'a> {
type Item = ErrorField<'a>;
type Error = io::Error;
#[inline]
fn next(&mut self) -> io::Result<Option<ErrorField<'a>>> {
let type_ = self.buf.read_u8()?;
if type_ == 0 {
if self.buf.is_empty() {
return Ok(None);
} else {
return Err(io::Error::new(
io::ErrorKind::InvalidInput,
"invalid message length: error fields is not drained",
));
}
}
let value_end = find_null(self.buf, 0)?;
let value = get_str(&self.buf[..value_end])?;
self.buf = &self.buf[value_end + 1..];
Ok(Some(ErrorField { type_, value }))
}
}
pub struct ErrorField<'a> {
type_: u8,
value: &'a str,
}
impl ErrorField<'_> {
#[inline]
pub fn type_(&self) -> u8 {
self.type_
}
#[inline]
pub fn value(&self) -> &str {
self.value
}
}
pub struct NoticeResponseBody {
storage: Bytes,
}
impl NoticeResponseBody {
#[inline]
pub fn fields(&self) -> ErrorFields<'_> {
ErrorFields { buf: &self.storage }
}
pub fn as_bytes(&self) -> &[u8] {
&self.storage
}
}
pub struct NotificationResponseBody {
process_id: i32,
channel: Bytes,
message: Bytes,
}
impl NotificationResponseBody {
#[inline]
pub fn process_id(&self) -> i32 {
self.process_id
}
#[inline]
pub fn channel(&self) -> io::Result<&str> {
get_str(&self.channel)
}
#[inline]
pub fn message(&self) -> io::Result<&str> {
get_str(&self.message)
}
}
pub struct ParameterDescriptionBody {
storage: Bytes,
len: u16,
}
impl ParameterDescriptionBody {
#[inline]
pub fn parameters(&self) -> Parameters<'_> {
Parameters {
buf: &self.storage,
remaining: self.len,
}
}
}
pub struct Parameters<'a> {
buf: &'a [u8],
remaining: u16,
}
impl FallibleIterator for Parameters<'_> {
type Item = Oid;
type Error = io::Error;
#[inline]
fn next(&mut self) -> io::Result<Option<Oid>> {
if self.remaining == 0 {
if self.buf.is_empty() {
return Ok(None);
} else {
return Err(io::Error::new(
io::ErrorKind::InvalidInput,
"invalid message length: parameters is not drained",
));
}
}
self.remaining -= 1;
self.buf.read_u32::<BigEndian>().map(Some)
}
#[inline]
fn size_hint(&self) -> (usize, Option<usize>) {
let len = self.remaining as usize;
(len, Some(len))
}
}
pub struct ParameterStatusBody {
name: Bytes,
value: Bytes,
}
impl ParameterStatusBody {
#[inline]
pub fn name(&self) -> io::Result<&str> {
get_str(&self.name)
}
#[inline]
pub fn value(&self) -> io::Result<&str> {
get_str(&self.value)
}
}
pub struct ReadyForQueryBody {
status: u8,
}
impl ReadyForQueryBody {
#[inline]
pub fn status(&self) -> u8 {
self.status
}
}
pub struct RowDescriptionBody {
storage: Bytes,
len: u16,
}
impl RowDescriptionBody {
#[inline]
pub fn fields(&self) -> Fields<'_> {
Fields {
buf: &self.storage,
remaining: self.len,
}
}
}
pub struct Fields<'a> {
buf: &'a [u8],
remaining: u16,
}
impl<'a> FallibleIterator for Fields<'a> {
type Item = Field<'a>;
type Error = io::Error;
#[inline]
fn next(&mut self) -> io::Result<Option<Field<'a>>> {
if self.remaining == 0 {
if self.buf.is_empty() {
return Ok(None);
} else {
return Err(io::Error::new(
io::ErrorKind::InvalidInput,
"invalid message length: field is not drained",
));
}
}
self.remaining -= 1;
let name_end = find_null(self.buf, 0)?;
let name = get_str(&self.buf[..name_end])?;
self.buf = &self.buf[name_end + 1..];
let table_oid = self.buf.read_u32::<BigEndian>()?;
let column_id = self.buf.read_i16::<BigEndian>()?;
let type_oid = self.buf.read_u32::<BigEndian>()?;
let type_size = self.buf.read_i16::<BigEndian>()?;
let type_modifier = self.buf.read_i32::<BigEndian>()?;
let format = self.buf.read_i16::<BigEndian>()?;
Ok(Some(Field {
name,
table_oid,
column_id,
type_oid,
type_size,
type_modifier,
format,
}))
}
}
pub struct Field<'a> {
name: &'a str,
table_oid: Oid,
column_id: i16,
type_oid: Oid,
type_size: i16,
type_modifier: i32,
format: i16,
}
impl<'a> Field<'a> {
#[inline]
pub fn name(&self) -> &'a str {
self.name
}
#[inline]
pub fn table_oid(&self) -> Oid {
self.table_oid
}
#[inline]
pub fn column_id(&self) -> i16 {
self.column_id
}
#[inline]
pub fn type_oid(&self) -> Oid {
self.type_oid
}
#[inline]
pub fn type_size(&self) -> i16 {
self.type_size
}
#[inline]
pub fn type_modifier(&self) -> i32 {
self.type_modifier
}
#[inline]
pub fn format(&self) -> i16 {
self.format
}
}
#[inline]
fn find_null(buf: &[u8], start: usize) -> io::Result<usize> {
match memchr(0, &buf[start..]) {
Some(pos) => Ok(pos + start),
None => Err(io::Error::new(
io::ErrorKind::UnexpectedEof,
"unexpected EOF",
)),
}
}
#[inline]
fn get_str(buf: &[u8]) -> io::Result<&str> {
str::from_utf8(buf).map_err(|e| io::Error::new(io::ErrorKind::InvalidInput, e))
}

View File

@@ -1,309 +0,0 @@
//! Frontend message serialization.
#![allow(missing_docs)]
use byteorder::{BigEndian, ByteOrder};
use bytes::{Buf, BufMut, BytesMut};
use std::convert::TryFrom;
use std::error::Error;
use std::io;
use std::marker;
use crate::{write_nullable, FromUsize, IsNull, Oid};
#[inline]
fn write_body<F, E>(buf: &mut BytesMut, f: F) -> Result<(), E>
where
F: FnOnce(&mut BytesMut) -> Result<(), E>,
E: From<io::Error>,
{
let base = buf.len();
buf.extend_from_slice(&[0; 4]);
f(buf)?;
let size = i32::from_usize(buf.len() - base)?;
BigEndian::write_i32(&mut buf[base..], size);
Ok(())
}
pub enum BindError {
Conversion(Box<dyn Error + marker::Sync + Send>),
Serialization(io::Error),
}
impl From<Box<dyn Error + marker::Sync + Send>> for BindError {
#[inline]
fn from(e: Box<dyn Error + marker::Sync + Send>) -> BindError {
BindError::Conversion(e)
}
}
impl From<io::Error> for BindError {
#[inline]
fn from(e: io::Error) -> BindError {
BindError::Serialization(e)
}
}
#[inline]
pub fn bind<I, J, F, T, K>(
portal: &str,
statement: &str,
formats: I,
values: J,
mut serializer: F,
result_formats: K,
buf: &mut BytesMut,
) -> Result<(), BindError>
where
I: IntoIterator<Item = i16>,
J: IntoIterator<Item = T>,
F: FnMut(T, &mut BytesMut) -> Result<IsNull, Box<dyn Error + marker::Sync + Send>>,
K: IntoIterator<Item = i16>,
{
buf.put_u8(b'B');
write_body(buf, |buf| {
write_cstr(portal.as_bytes(), buf)?;
write_cstr(statement.as_bytes(), buf)?;
write_counted(
formats,
|f, buf| {
buf.put_i16(f);
Ok::<_, io::Error>(())
},
buf,
)?;
write_counted(
values,
|v, buf| write_nullable(|buf| serializer(v, buf), buf),
buf,
)?;
write_counted(
result_formats,
|f, buf| {
buf.put_i16(f);
Ok::<_, io::Error>(())
},
buf,
)?;
Ok(())
})
}
#[inline]
fn write_counted<I, T, F, E>(items: I, mut serializer: F, buf: &mut BytesMut) -> Result<(), E>
where
I: IntoIterator<Item = T>,
F: FnMut(T, &mut BytesMut) -> Result<(), E>,
E: From<io::Error>,
{
let base = buf.len();
buf.extend_from_slice(&[0; 2]);
let mut count = 0;
for item in items {
serializer(item, buf)?;
count += 1;
}
let count = i16::from_usize(count)?;
BigEndian::write_i16(&mut buf[base..], count);
Ok(())
}
#[inline]
pub fn cancel_request(process_id: i32, secret_key: i32, buf: &mut BytesMut) {
write_body(buf, |buf| {
buf.put_i32(80_877_102);
buf.put_i32(process_id);
buf.put_i32(secret_key);
Ok::<_, io::Error>(())
})
.unwrap();
}
#[inline]
pub fn close(variant: u8, name: &str, buf: &mut BytesMut) -> io::Result<()> {
buf.put_u8(b'C');
write_body(buf, |buf| {
buf.put_u8(variant);
write_cstr(name.as_bytes(), buf)
})
}
pub struct CopyData<T> {
buf: T,
len: i32,
}
impl<T> CopyData<T>
where
T: Buf,
{
pub fn new(buf: T) -> io::Result<CopyData<T>> {
let len = buf
.remaining()
.checked_add(4)
.and_then(|l| i32::try_from(l).ok())
.ok_or_else(|| {
io::Error::new(io::ErrorKind::InvalidInput, "message length overflow")
})?;
Ok(CopyData { buf, len })
}
pub fn write(self, out: &mut BytesMut) {
out.put_u8(b'd');
out.put_i32(self.len);
out.put(self.buf);
}
}
#[inline]
pub fn copy_done(buf: &mut BytesMut) {
buf.put_u8(b'c');
write_body(buf, |_| Ok::<(), io::Error>(())).unwrap();
}
#[inline]
pub fn copy_fail(message: &str, buf: &mut BytesMut) -> io::Result<()> {
buf.put_u8(b'f');
write_body(buf, |buf| write_cstr(message.as_bytes(), buf))
}
#[inline]
pub fn describe(variant: u8, name: &str, buf: &mut BytesMut) -> io::Result<()> {
buf.put_u8(b'D');
write_body(buf, |buf| {
buf.put_u8(variant);
write_cstr(name.as_bytes(), buf)
})
}
#[inline]
pub fn execute(portal: &str, max_rows: i32, buf: &mut BytesMut) -> io::Result<()> {
buf.put_u8(b'E');
write_body(buf, |buf| {
write_cstr(portal.as_bytes(), buf)?;
buf.put_i32(max_rows);
Ok(())
})
}
#[inline]
pub fn parse<I>(name: &str, query: &str, param_types: I, buf: &mut BytesMut) -> io::Result<()>
where
I: IntoIterator<Item = Oid>,
{
buf.put_u8(b'P');
write_body(buf, |buf| {
write_cstr(name.as_bytes(), buf)?;
write_cstr(query.as_bytes(), buf)?;
write_counted(
param_types,
|t, buf| {
buf.put_u32(t);
Ok::<_, io::Error>(())
},
buf,
)?;
Ok(())
})
}
#[inline]
pub fn password_message(password: &[u8], buf: &mut BytesMut) -> io::Result<()> {
buf.put_u8(b'p');
write_body(buf, |buf| write_cstr(password, buf))
}
#[inline]
pub fn query(query: &str, buf: &mut BytesMut) -> io::Result<()> {
buf.put_u8(b'Q');
write_body(buf, |buf| write_cstr(query.as_bytes(), buf))
}
#[inline]
pub fn sasl_initial_response(mechanism: &str, data: &[u8], buf: &mut BytesMut) -> io::Result<()> {
buf.put_u8(b'p');
write_body(buf, |buf| {
write_cstr(mechanism.as_bytes(), buf)?;
let len = i32::from_usize(data.len())?;
buf.put_i32(len);
buf.put_slice(data);
Ok(())
})
}
#[inline]
pub fn sasl_response(data: &[u8], buf: &mut BytesMut) -> io::Result<()> {
buf.put_u8(b'p');
write_body(buf, |buf| {
buf.put_slice(data);
Ok(())
})
}
#[inline]
pub fn ssl_request(buf: &mut BytesMut) {
write_body(buf, |buf| {
buf.put_i32(80_877_103);
Ok::<_, io::Error>(())
})
.unwrap();
}
#[inline]
pub fn startup_message(parameters: &StartupMessageParams, buf: &mut BytesMut) -> io::Result<()> {
write_body(buf, |buf| {
// postgres protocol version 3.0(196608) in bigger-endian
buf.put_i32(0x00_03_00_00);
buf.put_slice(&parameters.params);
buf.put_u8(0);
Ok(())
})
}
#[derive(Debug, Clone, Default, PartialEq, Eq)]
pub struct StartupMessageParams {
pub params: BytesMut,
}
impl StartupMessageParams {
/// Set parameter's value by its name.
pub fn insert(&mut self, name: &str, value: &str) {
if name.contains('\0') || value.contains('\0') {
panic!("startup parameter name or value contained a null")
}
self.params.put_slice(name.as_bytes());
self.params.put_u8(0);
self.params.put_slice(value.as_bytes());
self.params.put_u8(0);
}
}
#[inline]
pub fn sync(buf: &mut BytesMut) {
buf.put_u8(b'S');
write_body(buf, |_| Ok::<(), io::Error>(())).unwrap();
}
#[inline]
pub fn terminate(buf: &mut BytesMut) {
buf.put_u8(b'X');
write_body(buf, |_| Ok::<(), io::Error>(())).unwrap();
}
#[inline]
fn write_cstr(s: &[u8], buf: &mut BytesMut) -> Result<(), io::Error> {
if s.contains(&0) {
return Err(io::Error::new(
io::ErrorKind::InvalidInput,
"string contains embedded null",
));
}
buf.put_slice(s);
buf.put_u8(0);
Ok(())
}

View File

@@ -1,8 +0,0 @@
//! Postgres message protocol support.
//!
//! See [Postgres's documentation][docs] for more information on message flow.
//!
//! [docs]: https://www.postgresql.org/docs/9.5/static/protocol-flow.html
pub mod backend;
pub mod frontend;

View File

@@ -1,89 +0,0 @@
//! Functions to encrypt a password in the client.
//!
//! This is intended to be used by client applications that wish to
//! send commands like `ALTER USER joe PASSWORD 'pwd'`. The password
//! need not be sent in cleartext if it is encrypted on the client
//! side. This is good because it ensures the cleartext password won't
//! end up in logs pg_stat displays, etc.
use crate::authentication::sasl;
use hmac::{Hmac, Mac};
use rand::RngCore;
use sha2::digest::FixedOutput;
use sha2::{Digest, Sha256};
#[cfg(test)]
mod test;
const SCRAM_DEFAULT_ITERATIONS: u32 = 4096;
const SCRAM_DEFAULT_SALT_LEN: usize = 16;
/// Hash password using SCRAM-SHA-256 with a randomly-generated
/// salt.
///
/// The client may assume the returned string doesn't contain any
/// special characters that would require escaping in an SQL command.
pub async fn scram_sha_256(password: &[u8]) -> String {
let mut salt: [u8; SCRAM_DEFAULT_SALT_LEN] = [0; SCRAM_DEFAULT_SALT_LEN];
let mut rng = rand::thread_rng();
rng.fill_bytes(&mut salt);
scram_sha_256_salt(password, salt).await
}
// Internal implementation of scram_sha_256 with a caller-provided
// salt. This is useful for testing.
pub(crate) async fn scram_sha_256_salt(
password: &[u8],
salt: [u8; SCRAM_DEFAULT_SALT_LEN],
) -> String {
// Prepare the password, per [RFC
// 4013](https://tools.ietf.org/html/rfc4013), if possible.
//
// Postgres treats passwords as byte strings (without embedded NUL
// bytes), but SASL expects passwords to be valid UTF-8.
//
// Follow the behavior of libpq's PQencryptPasswordConn(), and
// also the backend. If the password is not valid UTF-8, or if it
// contains prohibited characters (such as non-ASCII whitespace),
// just skip the SASLprep step and use the original byte
// sequence.
let prepared: Vec<u8> = match std::str::from_utf8(password) {
Ok(password_str) => {
match stringprep::saslprep(password_str) {
Ok(p) => p.into_owned().into_bytes(),
// contains invalid characters; skip saslprep
Err(_) => Vec::from(password),
}
}
// not valid UTF-8; skip saslprep
Err(_) => Vec::from(password),
};
// salt password
let salted_password = sasl::hi(&prepared, &salt, SCRAM_DEFAULT_ITERATIONS).await;
// client key
let mut hmac = Hmac::<Sha256>::new_from_slice(&salted_password)
.expect("HMAC is able to accept all key sizes");
hmac.update(b"Client Key");
let client_key = hmac.finalize().into_bytes();
// stored key
let mut hash = Sha256::default();
hash.update(client_key.as_slice());
let stored_key = hash.finalize_fixed();
// server key
let mut hmac = Hmac::<Sha256>::new_from_slice(&salted_password)
.expect("HMAC is able to accept all key sizes");
hmac.update(b"Server Key");
let server_key = hmac.finalize().into_bytes();
format!(
"SCRAM-SHA-256${}:{}${}:{}",
SCRAM_DEFAULT_ITERATIONS,
base64::encode(salt),
base64::encode(stored_key),
base64::encode(server_key)
)
}

View File

@@ -1,11 +0,0 @@
use crate::password;
#[tokio::test]
async fn test_encrypt_scram_sha_256() {
// Specify the salt to make the test deterministic. Any bytes will do.
let salt: [u8; 16] = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16];
assert_eq!(
password::scram_sha_256_salt(b"secret", salt).await,
"SCRAM-SHA-256$4096:AQIDBAUGBwgJCgsMDQ4PEA==$8rrDg00OqaiWXJ7p+sCgHEIaBSHY89ZJl3mfIsf32oY=:05L1f+yZbiN8O0AnO40Og85NNRhvzTS57naKRWCcsIA="
);
}

View File

@@ -1,294 +0,0 @@
//! Conversions to and from Postgres's binary format for various types.
use byteorder::{BigEndian, ReadBytesExt};
use bytes::{BufMut, BytesMut};
use fallible_iterator::FallibleIterator;
use std::boxed::Box as StdBox;
use std::error::Error;
use std::str;
use crate::Oid;
#[cfg(test)]
mod test;
/// Serializes a `TEXT`, `VARCHAR`, `CHAR(n)`, `NAME`, or `CITEXT` value.
#[inline]
pub fn text_to_sql(v: &str, buf: &mut BytesMut) {
buf.put_slice(v.as_bytes());
}
/// Deserializes a `TEXT`, `VARCHAR`, `CHAR(n)`, `NAME`, or `CITEXT` value.
#[inline]
pub fn text_from_sql(buf: &[u8]) -> Result<&str, StdBox<dyn Error + Sync + Send>> {
Ok(str::from_utf8(buf)?)
}
/// Deserializes a `"char"` value.
#[inline]
pub fn char_from_sql(mut buf: &[u8]) -> Result<i8, StdBox<dyn Error + Sync + Send>> {
let v = buf.read_i8()?;
if !buf.is_empty() {
return Err("invalid buffer size".into());
}
Ok(v)
}
/// Serializes an `OID` value.
#[inline]
pub fn oid_to_sql(v: Oid, buf: &mut BytesMut) {
buf.put_u32(v);
}
/// Deserializes an `OID` value.
#[inline]
pub fn oid_from_sql(mut buf: &[u8]) -> Result<Oid, StdBox<dyn Error + Sync + Send>> {
let v = buf.read_u32::<BigEndian>()?;
if !buf.is_empty() {
return Err("invalid buffer size".into());
}
Ok(v)
}
/// A fallible iterator over `HSTORE` entries.
pub struct HstoreEntries<'a> {
remaining: i32,
buf: &'a [u8],
}
impl<'a> FallibleIterator for HstoreEntries<'a> {
type Item = (&'a str, Option<&'a str>);
type Error = StdBox<dyn Error + Sync + Send>;
#[inline]
#[allow(clippy::type_complexity)]
fn next(
&mut self,
) -> Result<Option<(&'a str, Option<&'a str>)>, StdBox<dyn Error + Sync + Send>> {
if self.remaining == 0 {
if !self.buf.is_empty() {
return Err("invalid buffer size".into());
}
return Ok(None);
}
self.remaining -= 1;
let key_len = self.buf.read_i32::<BigEndian>()?;
if key_len < 0 {
return Err("invalid key length".into());
}
let (key, buf) = self.buf.split_at(key_len as usize);
let key = str::from_utf8(key)?;
self.buf = buf;
let value_len = self.buf.read_i32::<BigEndian>()?;
let value = if value_len < 0 {
None
} else {
let (value, buf) = self.buf.split_at(value_len as usize);
let value = str::from_utf8(value)?;
self.buf = buf;
Some(value)
};
Ok(Some((key, value)))
}
#[inline]
fn size_hint(&self) -> (usize, Option<usize>) {
let len = self.remaining as usize;
(len, Some(len))
}
}
/// Deserializes an array value.
#[inline]
pub fn array_from_sql(mut buf: &[u8]) -> Result<Array<'_>, StdBox<dyn Error + Sync + Send>> {
let dimensions = buf.read_i32::<BigEndian>()?;
if dimensions < 0 {
return Err("invalid dimension count".into());
}
let mut r = buf;
let mut elements = 1i32;
for _ in 0..dimensions {
let len = r.read_i32::<BigEndian>()?;
if len < 0 {
return Err("invalid dimension size".into());
}
let _lower_bound = r.read_i32::<BigEndian>()?;
elements = match elements.checked_mul(len) {
Some(elements) => elements,
None => return Err("too many array elements".into()),
};
}
if dimensions == 0 {
elements = 0;
}
Ok(Array {
dimensions,
elements,
buf,
})
}
/// A Postgres array.
pub struct Array<'a> {
dimensions: i32,
elements: i32,
buf: &'a [u8],
}
impl<'a> Array<'a> {
/// Returns an iterator over the dimensions of the array.
#[inline]
pub fn dimensions(&self) -> ArrayDimensions<'a> {
ArrayDimensions(&self.buf[..self.dimensions as usize * 8])
}
/// Returns an iterator over the values of the array.
#[inline]
pub fn values(&self) -> ArrayValues<'a> {
ArrayValues {
remaining: self.elements,
buf: &self.buf[self.dimensions as usize * 8..],
}
}
}
/// An iterator over the dimensions of an array.
pub struct ArrayDimensions<'a>(&'a [u8]);
impl FallibleIterator for ArrayDimensions<'_> {
type Item = ArrayDimension;
type Error = StdBox<dyn Error + Sync + Send>;
#[inline]
fn next(&mut self) -> Result<Option<ArrayDimension>, StdBox<dyn Error + Sync + Send>> {
if self.0.is_empty() {
return Ok(None);
}
let len = self.0.read_i32::<BigEndian>()?;
let lower_bound = self.0.read_i32::<BigEndian>()?;
Ok(Some(ArrayDimension { len, lower_bound }))
}
#[inline]
fn size_hint(&self) -> (usize, Option<usize>) {
let len = self.0.len() / 8;
(len, Some(len))
}
}
/// Information about a dimension of an array.
#[derive(Debug, Copy, Clone, PartialEq, Eq)]
pub struct ArrayDimension {
/// The length of this dimension.
pub len: i32,
/// The base value used to index into this dimension.
pub lower_bound: i32,
}
/// An iterator over the values of an array, in row-major order.
pub struct ArrayValues<'a> {
remaining: i32,
buf: &'a [u8],
}
impl<'a> FallibleIterator for ArrayValues<'a> {
type Item = Option<&'a [u8]>;
type Error = StdBox<dyn Error + Sync + Send>;
#[inline]
fn next(&mut self) -> Result<Option<Option<&'a [u8]>>, StdBox<dyn Error + Sync + Send>> {
if self.remaining == 0 {
if !self.buf.is_empty() {
return Err("invalid message length: arrayvalue not drained".into());
}
return Ok(None);
}
self.remaining -= 1;
let len = self.buf.read_i32::<BigEndian>()?;
let val = if len < 0 {
None
} else {
if self.buf.len() < len as usize {
return Err("invalid value length".into());
}
let (val, buf) = self.buf.split_at(len as usize);
self.buf = buf;
Some(val)
};
Ok(Some(val))
}
fn size_hint(&self) -> (usize, Option<usize>) {
let len = self.remaining as usize;
(len, Some(len))
}
}
/// Serializes a Postgres ltree string
#[inline]
pub fn ltree_to_sql(v: &str, buf: &mut BytesMut) {
// A version number is prepended to an ltree string per spec
buf.put_u8(1);
// Append the rest of the query
buf.put_slice(v.as_bytes());
}
/// Deserialize a Postgres ltree string
#[inline]
pub fn ltree_from_sql(buf: &[u8]) -> Result<&str, StdBox<dyn Error + Sync + Send>> {
match buf {
// Remove the version number from the front of the ltree per spec
[1u8, rest @ ..] => Ok(str::from_utf8(rest)?),
_ => Err("ltree version 1 only supported".into()),
}
}
/// Serializes a Postgres lquery string
#[inline]
pub fn lquery_to_sql(v: &str, buf: &mut BytesMut) {
// A version number is prepended to an lquery string per spec
buf.put_u8(1);
// Append the rest of the query
buf.put_slice(v.as_bytes());
}
/// Deserialize a Postgres lquery string
#[inline]
pub fn lquery_from_sql(buf: &[u8]) -> Result<&str, StdBox<dyn Error + Sync + Send>> {
match buf {
// Remove the version number from the front of the lquery per spec
[1u8, rest @ ..] => Ok(str::from_utf8(rest)?),
_ => Err("lquery version 1 only supported".into()),
}
}
/// Serializes a Postgres ltxtquery string
#[inline]
pub fn ltxtquery_to_sql(v: &str, buf: &mut BytesMut) {
// A version number is prepended to an ltxtquery string per spec
buf.put_u8(1);
// Append the rest of the query
buf.put_slice(v.as_bytes());
}
/// Deserialize a Postgres ltxtquery string
#[inline]
pub fn ltxtquery_from_sql(buf: &[u8]) -> Result<&str, StdBox<dyn Error + Sync + Send>> {
match buf {
// Remove the version number from the front of the ltxtquery per spec
[1u8, rest @ ..] => Ok(str::from_utf8(rest)?),
_ => Err("ltxtquery version 1 only supported".into()),
}
}

View File

@@ -1,87 +0,0 @@
use bytes::{Buf, BytesMut};
use super::*;
#[test]
fn ltree_sql() {
let mut query = vec![1u8];
query.extend_from_slice("A.B.C".as_bytes());
let mut buf = BytesMut::new();
ltree_to_sql("A.B.C", &mut buf);
assert_eq!(query.as_slice(), buf.chunk());
}
#[test]
fn ltree_str() {
let mut query = vec![1u8];
query.extend_from_slice("A.B.C".as_bytes());
assert!(ltree_from_sql(query.as_slice()).is_ok())
}
#[test]
fn ltree_wrong_version() {
let mut query = vec![2u8];
query.extend_from_slice("A.B.C".as_bytes());
assert!(ltree_from_sql(query.as_slice()).is_err())
}
#[test]
fn lquery_sql() {
let mut query = vec![1u8];
query.extend_from_slice("A.B.C".as_bytes());
let mut buf = BytesMut::new();
lquery_to_sql("A.B.C", &mut buf);
assert_eq!(query.as_slice(), buf.chunk());
}
#[test]
fn lquery_str() {
let mut query = vec![1u8];
query.extend_from_slice("A.B.C".as_bytes());
assert!(lquery_from_sql(query.as_slice()).is_ok())
}
#[test]
fn lquery_wrong_version() {
let mut query = vec![2u8];
query.extend_from_slice("A.B.C".as_bytes());
assert!(lquery_from_sql(query.as_slice()).is_err())
}
#[test]
fn ltxtquery_sql() {
let mut query = vec![1u8];
query.extend_from_slice("a & b*".as_bytes());
let mut buf = BytesMut::new();
ltree_to_sql("a & b*", &mut buf);
assert_eq!(query.as_slice(), buf.chunk());
}
#[test]
fn ltxtquery_str() {
let mut query = vec![1u8];
query.extend_from_slice("a & b*".as_bytes());
assert!(ltree_from_sql(query.as_slice()).is_ok())
}
#[test]
fn ltxtquery_wrong_version() {
let mut query = vec![2u8];
query.extend_from_slice("a & b*".as_bytes());
assert!(ltree_from_sql(query.as_slice()).is_err())
}

View File

@@ -1,10 +0,0 @@
[package]
name = "postgres-types2"
version = "0.1.0"
edition = "2018"
license = "MIT/Apache-2.0"
[dependencies]
bytes.workspace = true
fallible-iterator.workspace = true
postgres-protocol2 = { path = "../postgres-protocol2" }

View File

@@ -1,477 +0,0 @@
//! Conversions to and from Postgres types.
//!
//! This crate is used by the `tokio-postgres` and `postgres` crates. You normally don't need to depend directly on it
//! unless you want to define your own `ToSql` or `FromSql` definitions.
#![doc(html_root_url = "https://docs.rs/postgres-types/0.2")]
#![warn(clippy::all, rust_2018_idioms, missing_docs)]
use fallible_iterator::FallibleIterator;
use postgres_protocol2::types;
use std::any::type_name;
use std::error::Error;
use std::fmt;
use std::sync::Arc;
use crate::type_gen::{Inner, Other};
#[doc(inline)]
pub use postgres_protocol2::Oid;
use bytes::BytesMut;
/// Generates a simple implementation of `ToSql::accepts` which accepts the
/// types passed to it.
macro_rules! accepts {
($($expected:ident),+) => (
fn accepts(ty: &$crate::Type) -> bool {
matches!(*ty, $($crate::Type::$expected)|+)
}
)
}
/// Generates an implementation of `ToSql::to_sql_checked`.
///
/// All `ToSql` implementations should use this macro.
macro_rules! to_sql_checked {
() => {
fn to_sql_checked(
&self,
ty: &$crate::Type,
out: &mut $crate::private::BytesMut,
) -> ::std::result::Result<
$crate::IsNull,
Box<dyn ::std::error::Error + ::std::marker::Sync + ::std::marker::Send>,
> {
$crate::__to_sql_checked(self, ty, out)
}
};
}
// WARNING: this function is not considered part of this crate's public API.
// It is subject to change at any time.
#[doc(hidden)]
pub fn __to_sql_checked<T>(
v: &T,
ty: &Type,
out: &mut BytesMut,
) -> Result<IsNull, Box<dyn Error + Sync + Send>>
where
T: ToSql,
{
if !T::accepts(ty) {
return Err(Box::new(WrongType::new::<T>(ty.clone())));
}
v.to_sql(ty, out)
}
// mod pg_lsn;
#[doc(hidden)]
pub mod private;
// mod special;
mod type_gen;
/// A Postgres type.
#[derive(PartialEq, Eq, Clone, Hash)]
pub struct Type(Inner);
impl fmt::Debug for Type {
fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result {
fmt::Debug::fmt(&self.0, fmt)
}
}
impl fmt::Display for Type {
fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result {
match self.schema() {
"public" | "pg_catalog" => {}
schema => write!(fmt, "{}.", schema)?,
}
fmt.write_str(self.name())
}
}
impl Type {
/// Creates a new `Type`.
pub fn new(name: String, oid: Oid, kind: Kind, schema: String) -> Type {
Type(Inner::Other(Arc::new(Other {
name,
oid,
kind,
schema,
})))
}
/// Returns the `Type` corresponding to the provided `Oid` if it
/// corresponds to a built-in type.
pub fn from_oid(oid: Oid) -> Option<Type> {
Inner::from_oid(oid).map(Type)
}
/// Returns the OID of the `Type`.
pub fn oid(&self) -> Oid {
self.0.oid()
}
/// Returns the kind of this type.
pub fn kind(&self) -> &Kind {
self.0.kind()
}
/// Returns the schema of this type.
pub fn schema(&self) -> &str {
match self.0 {
Inner::Other(ref u) => &u.schema,
_ => "pg_catalog",
}
}
/// Returns the name of this type.
pub fn name(&self) -> &str {
self.0.name()
}
}
/// Represents the kind of a Postgres type.
#[derive(Debug, Clone, PartialEq, Eq, Hash)]
#[non_exhaustive]
pub enum Kind {
/// A simple type like `VARCHAR` or `INTEGER`.
Simple,
/// An enumerated type along with its variants.
Enum(Vec<String>),
/// A pseudo-type.
Pseudo,
/// An array type along with the type of its elements.
Array(Type),
/// A range type along with the type of its elements.
Range(Type),
/// A multirange type along with the type of its elements.
Multirange(Type),
/// A domain type along with its underlying type.
Domain(Type),
/// A composite type along with information about its fields.
Composite(Vec<Field>),
}
/// Information about a field of a composite type.
#[derive(Debug, Clone, PartialEq, Eq, Hash)]
pub struct Field {
name: String,
type_: Type,
}
impl Field {
/// Creates a new `Field`.
pub fn new(name: String, type_: Type) -> Field {
Field { name, type_ }
}
/// Returns the name of the field.
pub fn name(&self) -> &str {
&self.name
}
/// Returns the type of the field.
pub fn type_(&self) -> &Type {
&self.type_
}
}
/// An error indicating that a `NULL` Postgres value was passed to a `FromSql`
/// implementation that does not support `NULL` values.
#[derive(Debug, Clone, Copy)]
pub struct WasNull;
impl fmt::Display for WasNull {
fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result {
fmt.write_str("a Postgres value was `NULL`")
}
}
impl Error for WasNull {}
/// An error indicating that a conversion was attempted between incompatible
/// Rust and Postgres types.
#[derive(Debug)]
pub struct WrongType {
postgres: Type,
rust: &'static str,
}
impl fmt::Display for WrongType {
fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result {
write!(
fmt,
"cannot convert between the Rust type `{}` and the Postgres type `{}`",
self.rust, self.postgres,
)
}
}
impl Error for WrongType {}
impl WrongType {
/// Creates a new `WrongType` error.
pub fn new<T>(ty: Type) -> WrongType {
WrongType {
postgres: ty,
rust: type_name::<T>(),
}
}
}
/// An error indicating that a as_text conversion was attempted on a binary
/// result.
#[derive(Debug)]
pub struct WrongFormat {}
impl Error for WrongFormat {}
impl fmt::Display for WrongFormat {
fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result {
write!(
fmt,
"cannot read column as text while it is in binary format"
)
}
}
/// A trait for types that can be created from a Postgres value.
pub trait FromSql<'a>: Sized {
/// Creates a new value of this type from a buffer of data of the specified
/// Postgres `Type` in its binary format.
///
/// The caller of this method is responsible for ensuring that this type
/// is compatible with the Postgres `Type`.
fn from_sql(ty: &Type, raw: &'a [u8]) -> Result<Self, Box<dyn Error + Sync + Send>>;
/// Creates a new value of this type from a `NULL` SQL value.
///
/// The caller of this method is responsible for ensuring that this type
/// is compatible with the Postgres `Type`.
///
/// The default implementation returns `Err(Box::new(WasNull))`.
#[allow(unused_variables)]
fn from_sql_null(ty: &Type) -> Result<Self, Box<dyn Error + Sync + Send>> {
Err(Box::new(WasNull))
}
/// A convenience function that delegates to `from_sql` and `from_sql_null` depending on the
/// value of `raw`.
fn from_sql_nullable(
ty: &Type,
raw: Option<&'a [u8]>,
) -> Result<Self, Box<dyn Error + Sync + Send>> {
match raw {
Some(raw) => Self::from_sql(ty, raw),
None => Self::from_sql_null(ty),
}
}
/// Determines if a value of this type can be created from the specified
/// Postgres `Type`.
fn accepts(ty: &Type) -> bool;
}
/// A trait for types which can be created from a Postgres value without borrowing any data.
///
/// This is primarily useful for trait bounds on functions.
pub trait FromSqlOwned: for<'a> FromSql<'a> {}
impl<T> FromSqlOwned for T where T: for<'a> FromSql<'a> {}
impl<'a, T: FromSql<'a>> FromSql<'a> for Option<T> {
fn from_sql(ty: &Type, raw: &'a [u8]) -> Result<Option<T>, Box<dyn Error + Sync + Send>> {
<T as FromSql>::from_sql(ty, raw).map(Some)
}
fn from_sql_null(_: &Type) -> Result<Option<T>, Box<dyn Error + Sync + Send>> {
Ok(None)
}
fn accepts(ty: &Type) -> bool {
<T as FromSql>::accepts(ty)
}
}
impl<'a, T: FromSql<'a>> FromSql<'a> for Vec<T> {
fn from_sql(ty: &Type, raw: &'a [u8]) -> Result<Vec<T>, Box<dyn Error + Sync + Send>> {
let member_type = match *ty.kind() {
Kind::Array(ref member) => member,
_ => panic!("expected array type"),
};
let array = types::array_from_sql(raw)?;
if array.dimensions().count()? > 1 {
return Err("array contains too many dimensions".into());
}
array
.values()
.map(|v| T::from_sql_nullable(member_type, v))
.collect()
}
fn accepts(ty: &Type) -> bool {
match *ty.kind() {
Kind::Array(ref inner) => T::accepts(inner),
_ => false,
}
}
}
impl<'a> FromSql<'a> for String {
fn from_sql(ty: &Type, raw: &'a [u8]) -> Result<String, Box<dyn Error + Sync + Send>> {
<&str as FromSql>::from_sql(ty, raw).map(ToString::to_string)
}
fn accepts(ty: &Type) -> bool {
<&str as FromSql>::accepts(ty)
}
}
impl<'a> FromSql<'a> for &'a str {
fn from_sql(ty: &Type, raw: &'a [u8]) -> Result<&'a str, Box<dyn Error + Sync + Send>> {
match *ty {
ref ty if ty.name() == "ltree" => types::ltree_from_sql(raw),
ref ty if ty.name() == "lquery" => types::lquery_from_sql(raw),
ref ty if ty.name() == "ltxtquery" => types::ltxtquery_from_sql(raw),
_ => types::text_from_sql(raw),
}
}
fn accepts(ty: &Type) -> bool {
match *ty {
Type::VARCHAR | Type::TEXT | Type::BPCHAR | Type::NAME | Type::UNKNOWN => true,
ref ty
if (ty.name() == "citext"
|| ty.name() == "ltree"
|| ty.name() == "lquery"
|| ty.name() == "ltxtquery") =>
{
true
}
_ => false,
}
}
}
macro_rules! simple_from {
($t:ty, $f:ident, $($expected:ident),+) => {
impl<'a> FromSql<'a> for $t {
fn from_sql(_: &Type, raw: &'a [u8]) -> Result<$t, Box<dyn Error + Sync + Send>> {
types::$f(raw)
}
accepts!($($expected),+);
}
}
}
simple_from!(i8, char_from_sql, CHAR);
simple_from!(u32, oid_from_sql, OID);
/// An enum representing the nullability of a Postgres value.
pub enum IsNull {
/// The value is NULL.
Yes,
/// The value is not NULL.
No,
}
/// A trait for types that can be converted into Postgres values.
pub trait ToSql: fmt::Debug {
/// Converts the value of `self` into the binary format of the specified
/// Postgres `Type`, appending it to `out`.
///
/// The caller of this method is responsible for ensuring that this type
/// is compatible with the Postgres `Type`.
///
/// The return value indicates if this value should be represented as
/// `NULL`. If this is the case, implementations **must not** write
/// anything to `out`.
fn to_sql(&self, ty: &Type, out: &mut BytesMut) -> Result<IsNull, Box<dyn Error + Sync + Send>>
where
Self: Sized;
/// Determines if a value of this type can be converted to the specified
/// Postgres `Type`.
fn accepts(ty: &Type) -> bool
where
Self: Sized;
/// An adaptor method used internally by Rust-Postgres.
///
/// *All* implementations of this method should be generated by the
/// `to_sql_checked!()` macro.
fn to_sql_checked(
&self,
ty: &Type,
out: &mut BytesMut,
) -> Result<IsNull, Box<dyn Error + Sync + Send>>;
/// Specify the encode format
fn encode_format(&self, _ty: &Type) -> Format {
Format::Binary
}
}
/// Supported Postgres message format types
///
/// Using Text format in a message assumes a Postgres `SERVER_ENCODING` of `UTF8`
#[derive(Clone, Copy, Debug, PartialEq)]
pub enum Format {
/// Text format (UTF-8)
Text,
/// Compact, typed binary format
Binary,
}
impl ToSql for &str {
fn to_sql(&self, ty: &Type, w: &mut BytesMut) -> Result<IsNull, Box<dyn Error + Sync + Send>> {
match *ty {
ref ty if ty.name() == "ltree" => types::ltree_to_sql(self, w),
ref ty if ty.name() == "lquery" => types::lquery_to_sql(self, w),
ref ty if ty.name() == "ltxtquery" => types::ltxtquery_to_sql(self, w),
_ => types::text_to_sql(self, w),
}
Ok(IsNull::No)
}
fn accepts(ty: &Type) -> bool {
match *ty {
Type::VARCHAR | Type::TEXT | Type::BPCHAR | Type::NAME | Type::UNKNOWN => true,
ref ty
if (ty.name() == "citext"
|| ty.name() == "ltree"
|| ty.name() == "lquery"
|| ty.name() == "ltxtquery") =>
{
true
}
_ => false,
}
}
to_sql_checked!();
}
macro_rules! simple_to {
($t:ty, $f:ident, $($expected:ident),+) => {
impl ToSql for $t {
fn to_sql(&self,
_: &Type,
w: &mut BytesMut)
-> Result<IsNull, Box<dyn Error + Sync + Send>> {
types::$f(*self, w);
Ok(IsNull::No)
}
accepts!($($expected),+);
to_sql_checked!();
}
}
}
simple_to!(u32, oid_to_sql, OID);

View File

@@ -1,34 +0,0 @@
use crate::{FromSql, Type};
pub use bytes::BytesMut;
use std::error::Error;
pub fn read_be_i32(buf: &mut &[u8]) -> Result<i32, Box<dyn Error + Sync + Send>> {
if buf.len() < 4 {
return Err("invalid buffer size".into());
}
let mut bytes = [0; 4];
bytes.copy_from_slice(&buf[..4]);
*buf = &buf[4..];
Ok(i32::from_be_bytes(bytes))
}
pub fn read_value<'a, T>(
type_: &Type,
buf: &mut &'a [u8],
) -> Result<T, Box<dyn Error + Sync + Send>>
where
T: FromSql<'a>,
{
let len = read_be_i32(buf)?;
let value = if len < 0 {
None
} else {
if len as usize > buf.len() {
return Err("invalid buffer size".into());
}
let (head, tail) = buf.split_at(len as usize);
*buf = tail;
Some(head)
};
T::from_sql_nullable(type_, value)
}

File diff suppressed because it is too large Load Diff

View File

@@ -1,21 +0,0 @@
[package]
name = "tokio-postgres2"
version = "0.1.0"
edition = "2018"
license = "MIT/Apache-2.0"
[dependencies]
async-trait.workspace = true
bytes.workspace = true
byteorder.workspace = true
fallible-iterator.workspace = true
futures-util = { workspace = true, features = ["sink"] }
log = "0.4"
parking_lot.workspace = true
percent-encoding = "2.0"
pin-project-lite.workspace = true
phf = "0.11"
postgres-protocol2 = { path = "../postgres-protocol2" }
postgres-types2 = { path = "../postgres-types2" }
tokio = { workspace = true, features = ["io-util", "time", "net"] }
tokio-util = { workspace = true, features = ["codec"] }

View File

@@ -1,40 +0,0 @@
use tokio::net::TcpStream;
use crate::client::SocketConfig;
use crate::config::{Host, SslMode};
use crate::tls::MakeTlsConnect;
use crate::{cancel_query_raw, connect_socket, Error};
use std::io;
pub(crate) async fn cancel_query<T>(
config: Option<SocketConfig>,
ssl_mode: SslMode,
mut tls: T,
process_id: i32,
secret_key: i32,
) -> Result<(), Error>
where
T: MakeTlsConnect<TcpStream>,
{
let config = match config {
Some(config) => config,
None => {
return Err(Error::connect(io::Error::new(
io::ErrorKind::InvalidInput,
"unknown host",
)))
}
};
let hostname = match &config.host {
Host::Tcp(host) => &**host,
};
let tls = tls
.make_tls_connect(hostname)
.map_err(|e| Error::tls(e.into()))?;
let socket =
connect_socket::connect_socket(&config.host, config.port, config.connect_timeout).await?;
cancel_query_raw::cancel_query_raw(socket, ssl_mode, tls, process_id, secret_key).await
}

View File

@@ -1,29 +0,0 @@
use crate::config::SslMode;
use crate::tls::TlsConnect;
use crate::{connect_tls, Error};
use bytes::BytesMut;
use postgres_protocol2::message::frontend;
use tokio::io::{AsyncRead, AsyncWrite, AsyncWriteExt};
pub async fn cancel_query_raw<S, T>(
stream: S,
mode: SslMode,
tls: T,
process_id: i32,
secret_key: i32,
) -> Result<(), Error>
where
S: AsyncRead + AsyncWrite + Unpin,
T: TlsConnect<S>,
{
let mut stream = connect_tls::connect_tls(stream, mode, tls).await?;
let mut buf = BytesMut::new();
frontend::cancel_request(process_id, secret_key, &mut buf);
stream.write_all(&buf).await.map_err(Error::io)?;
stream.flush().await.map_err(Error::io)?;
stream.shutdown().await.map_err(Error::io)?;
Ok(())
}

View File

@@ -1,62 +0,0 @@
use crate::config::SslMode;
use crate::tls::TlsConnect;
use crate::{cancel_query, client::SocketConfig, tls::MakeTlsConnect};
use crate::{cancel_query_raw, Error};
use tokio::io::{AsyncRead, AsyncWrite};
use tokio::net::TcpStream;
/// The capability to request cancellation of in-progress queries on a
/// connection.
#[derive(Clone)]
pub struct CancelToken {
pub socket_config: Option<SocketConfig>,
pub ssl_mode: SslMode,
pub process_id: i32,
pub secret_key: i32,
}
impl CancelToken {
/// Attempts to cancel the in-progress query on the connection associated
/// with this `CancelToken`.
///
/// The server provides no information about whether a cancellation attempt was successful or not. An error will
/// only be returned if the client was unable to connect to the database.
///
/// Cancellation is inherently racy. There is no guarantee that the
/// cancellation request will reach the server before the query terminates
/// normally, or that the connection associated with this token is still
/// active.
///
/// Requires the `runtime` Cargo feature (enabled by default).
pub async fn cancel_query<T>(&self, tls: T) -> Result<(), Error>
where
T: MakeTlsConnect<TcpStream>,
{
cancel_query::cancel_query(
self.socket_config.clone(),
self.ssl_mode,
tls,
self.process_id,
self.secret_key,
)
.await
}
/// Like `cancel_query`, but uses a stream which is already connected to the server rather than opening a new
/// connection itself.
pub async fn cancel_query_raw<S, T>(&self, stream: S, tls: T) -> Result<(), Error>
where
S: AsyncRead + AsyncWrite + Unpin,
T: TlsConnect<S>,
{
cancel_query_raw::cancel_query_raw(
stream,
self.ssl_mode,
tls,
self.process_id,
self.secret_key,
)
.await
}
}

View File

@@ -1,271 +0,0 @@
use crate::codec::{BackendMessages, FrontendMessage};
use crate::config::Host;
use crate::config::SslMode;
use crate::connection::{Request, RequestMessages};
use crate::types::{Oid, Type};
use crate::{
simple_query, CancelToken, Error, ReadyForQueryStatus, Statement, Transaction,
TransactionBuilder,
};
use bytes::BytesMut;
use fallible_iterator::FallibleIterator;
use futures_util::{future, ready};
use postgres_protocol2::message::{backend::Message, frontend};
use std::collections::HashMap;
use std::fmt;
use std::task::{Context, Poll};
use tokio::sync::mpsc;
use std::time::Duration;
pub struct Responses {
receiver: mpsc::Receiver<BackendMessages>,
cur: BackendMessages,
}
impl Responses {
pub fn poll_next(&mut self, cx: &mut Context<'_>) -> Poll<Result<Message, Error>> {
loop {
match self.cur.next().map_err(Error::parse)? {
Some(Message::ErrorResponse(body)) => return Poll::Ready(Err(Error::db(body))),
Some(message) => return Poll::Ready(Ok(message)),
None => {}
}
match ready!(self.receiver.poll_recv(cx)) {
Some(messages) => self.cur = messages,
None => return Poll::Ready(Err(Error::closed())),
}
}
}
pub async fn next(&mut self) -> Result<Message, Error> {
future::poll_fn(|cx| self.poll_next(cx)).await
}
}
/// A cache of type info and prepared statements for fetching type info
/// (corresponding to the queries in the [prepare] module).
#[derive(Default)]
pub(crate) struct CachedTypeInfo {
/// A statement for basic information for a type from its
/// OID. Corresponds to [TYPEINFO_QUERY](prepare::TYPEINFO_QUERY) (or its
/// fallback).
typeinfo: Option<Statement>,
/// A statement for getting information for a composite type from its OID.
/// Corresponds to [TYPEINFO_QUERY](prepare::TYPEINFO_COMPOSITE_QUERY).
typeinfo_composite: Option<Statement>,
/// A statement for getting information for a composite type from its OID.
/// Corresponds to [TYPEINFO_QUERY](prepare::TYPEINFO_COMPOSITE_QUERY) (or
/// its fallback).
typeinfo_enum: Option<Statement>,
/// Cache of types already looked up.
types: HashMap<Oid, Type>,
}
impl CachedTypeInfo {
pub(crate) fn typeinfo(&mut self) -> Option<&Statement> {
self.typeinfo.as_ref()
}
pub(crate) fn set_typeinfo(&mut self, statement: Statement) -> &Statement {
self.typeinfo.insert(statement)
}
pub(crate) fn typeinfo_composite(&mut self) -> Option<&Statement> {
self.typeinfo_composite.as_ref()
}
pub(crate) fn set_typeinfo_composite(&mut self, statement: Statement) -> &Statement {
self.typeinfo_composite.insert(statement)
}
pub(crate) fn typeinfo_enum(&mut self) -> Option<&Statement> {
self.typeinfo_enum.as_ref()
}
pub(crate) fn set_typeinfo_enum(&mut self, statement: Statement) -> &Statement {
self.typeinfo_enum.insert(statement)
}
pub(crate) fn type_(&mut self, oid: Oid) -> Option<Type> {
self.types.get(&oid).cloned()
}
pub(crate) fn set_type(&mut self, oid: Oid, type_: &Type) {
self.types.insert(oid, type_.clone());
}
}
pub struct InnerClient {
sender: mpsc::UnboundedSender<Request>,
/// A buffer to use when writing out postgres commands.
buffer: BytesMut,
}
impl InnerClient {
pub fn send(&self, messages: RequestMessages) -> Result<Responses, Error> {
let (sender, receiver) = mpsc::channel(1);
let request = Request { messages, sender };
self.sender.send(request).map_err(|_| Error::closed())?;
Ok(Responses {
receiver,
cur: BackendMessages::empty(),
})
}
/// Call the given function with a buffer to be used when writing out
/// postgres commands.
pub fn with_buf<F, R>(&mut self, f: F) -> R
where
F: FnOnce(&mut BytesMut) -> R,
{
let r = f(&mut self.buffer);
self.buffer.clear();
r
}
}
#[derive(Clone)]
pub struct SocketConfig {
pub host: Host,
pub port: u16,
pub connect_timeout: Option<Duration>,
// pub keepalive: Option<KeepaliveConfig>,
}
/// An asynchronous PostgreSQL client.
///
/// The client is one half of what is returned when a connection is established. Users interact with the database
/// through this client object.
pub struct Client {
pub(crate) inner: InnerClient,
pub(crate) cached_typeinfo: CachedTypeInfo,
socket_config: SocketConfig,
ssl_mode: SslMode,
process_id: i32,
secret_key: i32,
}
impl Client {
pub(crate) fn new(
sender: mpsc::UnboundedSender<Request>,
socket_config: SocketConfig,
ssl_mode: SslMode,
process_id: i32,
secret_key: i32,
) -> Client {
Client {
inner: InnerClient {
sender,
buffer: Default::default(),
},
cached_typeinfo: Default::default(),
socket_config,
ssl_mode,
process_id,
secret_key,
}
}
/// Returns process_id.
pub fn get_process_id(&self) -> i32 {
self.process_id
}
/// Executes a sequence of SQL statements using the simple query protocol.
///
/// Statements should be separated by semicolons. If an error occurs, execution of the sequence will stop at that
/// point. This is intended for use when, for example, initializing a database schema.
///
/// # Warning
///
/// Prepared statements should be use for any query which contains user-specified data, as they provided the
/// functionality to safely embed that data in the request. Do not form statements via string concatenation and pass
/// them to this method!
pub async fn batch_execute(&mut self, query: &str) -> Result<ReadyForQueryStatus, Error> {
simple_query::batch_execute(&mut self.inner, query).await
}
/// Begins a new database transaction.
///
/// The transaction will roll back by default - use the `commit` method to commit it.
pub async fn transaction(&mut self) -> Result<Transaction<'_>, Error> {
struct RollbackIfNotDone<'me> {
client: &'me mut Client,
done: bool,
}
impl Drop for RollbackIfNotDone<'_> {
fn drop(&mut self) {
if self.done {
return;
}
let buf = self.client.inner.with_buf(|buf| {
frontend::query("ROLLBACK", buf).unwrap();
buf.split().freeze()
});
let _ = self
.client
.inner
.send(RequestMessages::Single(FrontendMessage::Raw(buf)));
}
}
// This is done, as `Future` created by this method can be dropped after
// `RequestMessages` is synchronously send to the `Connection` by
// `batch_execute()`, but before `Responses` is asynchronously polled to
// completion. In that case `Transaction` won't be created and thus
// won't be rolled back.
{
let mut cleaner = RollbackIfNotDone {
client: self,
done: false,
};
cleaner.client.batch_execute("BEGIN").await?;
cleaner.done = true;
}
Ok(Transaction::new(self))
}
/// Returns a builder for a transaction with custom settings.
///
/// Unlike the `transaction` method, the builder can be used to control the transaction's isolation level and other
/// attributes.
pub fn build_transaction(&mut self) -> TransactionBuilder<'_> {
TransactionBuilder::new(self)
}
/// Constructs a cancellation token that can later be used to request cancellation of a query running on the
/// connection associated with this client.
pub fn cancel_token(&self) -> CancelToken {
CancelToken {
socket_config: Some(self.socket_config.clone()),
ssl_mode: self.ssl_mode,
process_id: self.process_id,
secret_key: self.secret_key,
}
}
/// Determines if the connection to the server has already closed.
///
/// In that case, all future queries will fail.
pub fn is_closed(&self) -> bool {
self.inner.sender.is_closed()
}
}
impl fmt::Debug for Client {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
f.debug_struct("Client").finish()
}
}

View File

@@ -1,98 +0,0 @@
use bytes::{Buf, Bytes, BytesMut};
use fallible_iterator::FallibleIterator;
use postgres_protocol2::message::backend;
use postgres_protocol2::message::frontend::CopyData;
use std::io;
use tokio_util::codec::{Decoder, Encoder};
pub enum FrontendMessage {
Raw(Bytes),
CopyData(CopyData<Box<dyn Buf + Send>>),
}
pub enum BackendMessage {
Normal {
messages: BackendMessages,
request_complete: bool,
},
Async(backend::Message),
}
pub struct BackendMessages(BytesMut);
impl BackendMessages {
pub fn empty() -> BackendMessages {
BackendMessages(BytesMut::new())
}
}
impl FallibleIterator for BackendMessages {
type Item = backend::Message;
type Error = io::Error;
fn next(&mut self) -> io::Result<Option<backend::Message>> {
backend::Message::parse(&mut self.0)
}
}
pub struct PostgresCodec;
impl Encoder<FrontendMessage> for PostgresCodec {
type Error = io::Error;
fn encode(&mut self, item: FrontendMessage, dst: &mut BytesMut) -> io::Result<()> {
match item {
FrontendMessage::Raw(buf) => dst.extend_from_slice(&buf),
FrontendMessage::CopyData(data) => data.write(dst),
}
Ok(())
}
}
impl Decoder for PostgresCodec {
type Item = BackendMessage;
type Error = io::Error;
fn decode(&mut self, src: &mut BytesMut) -> Result<Option<BackendMessage>, io::Error> {
let mut idx = 0;
let mut request_complete = false;
while let Some(header) = backend::Header::parse(&src[idx..])? {
let len = header.len() as usize + 1;
if src[idx..].len() < len {
break;
}
match header.tag() {
backend::NOTICE_RESPONSE_TAG
| backend::NOTIFICATION_RESPONSE_TAG
| backend::PARAMETER_STATUS_TAG => {
if idx == 0 {
let message = backend::Message::parse(src)?.unwrap();
return Ok(Some(BackendMessage::Async(message)));
} else {
break;
}
}
_ => {}
}
idx += len;
if header.tag() == backend::READY_FOR_QUERY_TAG {
request_complete = true;
break;
}
}
if idx == 0 {
Ok(None)
} else {
Ok(Some(BackendMessage::Normal {
messages: BackendMessages(src.split_to(idx)),
request_complete,
}))
}
}
}

View File

@@ -1,264 +0,0 @@
//! Connection configuration.
use crate::connect::connect;
use crate::connect_raw::connect_raw;
use crate::connect_raw::RawConnection;
use crate::tls::MakeTlsConnect;
use crate::tls::TlsConnect;
use crate::{Client, Connection, Error};
use postgres_protocol2::message::frontend::StartupMessageParams;
use std::fmt;
use std::str;
use std::time::Duration;
use tokio::io::{AsyncRead, AsyncWrite};
pub use postgres_protocol2::authentication::sasl::ScramKeys;
use tokio::net::TcpStream;
/// TLS configuration.
#[derive(Debug, Copy, Clone, PartialEq, Eq)]
#[non_exhaustive]
pub enum SslMode {
/// Do not use TLS.
Disable,
/// Attempt to connect with TLS but allow sessions without.
Prefer,
/// Require the use of TLS.
Require,
}
/// Channel binding configuration.
#[derive(Debug, Copy, Clone, PartialEq, Eq)]
#[non_exhaustive]
pub enum ChannelBinding {
/// Do not use channel binding.
Disable,
/// Attempt to use channel binding but allow sessions without.
Prefer,
/// Require the use of channel binding.
Require,
}
/// Replication mode configuration.
#[derive(Debug, Copy, Clone, PartialEq, Eq)]
#[non_exhaustive]
pub enum ReplicationMode {
/// Physical replication.
Physical,
/// Logical replication.
Logical,
}
/// A host specification.
#[derive(Debug, Clone, PartialEq, Eq)]
pub enum Host {
/// A TCP hostname.
Tcp(String),
}
/// Precomputed keys which may override password during auth.
#[derive(Debug, Clone, Copy, PartialEq, Eq)]
pub enum AuthKeys {
/// A `ClientKey` & `ServerKey` pair for `SCRAM-SHA-256`.
ScramSha256(ScramKeys<32>),
}
/// Connection configuration.
#[derive(Clone, PartialEq, Eq)]
pub struct Config {
pub(crate) host: Host,
pub(crate) port: u16,
pub(crate) password: Option<Vec<u8>>,
pub(crate) auth_keys: Option<Box<AuthKeys>>,
pub(crate) ssl_mode: SslMode,
pub(crate) connect_timeout: Option<Duration>,
pub(crate) channel_binding: ChannelBinding,
pub(crate) server_params: StartupMessageParams,
database: bool,
username: bool,
}
impl Config {
/// Creates a new configuration.
pub fn new(host: String, port: u16) -> Config {
Config {
host: Host::Tcp(host),
port,
password: None,
auth_keys: None,
ssl_mode: SslMode::Prefer,
connect_timeout: None,
channel_binding: ChannelBinding::Prefer,
server_params: StartupMessageParams::default(),
database: false,
username: false,
}
}
/// Sets the user to authenticate with.
///
/// Required.
pub fn user(&mut self, user: &str) -> &mut Config {
self.set_param("user", user)
}
/// Gets the user to authenticate with, if one has been configured with
/// the `user` method.
pub fn user_is_set(&self) -> bool {
self.username
}
/// Sets the password to authenticate with.
pub fn password<T>(&mut self, password: T) -> &mut Config
where
T: AsRef<[u8]>,
{
self.password = Some(password.as_ref().to_vec());
self
}
/// Gets the password to authenticate with, if one has been configured with
/// the `password` method.
pub fn get_password(&self) -> Option<&[u8]> {
self.password.as_deref()
}
/// Sets precomputed protocol-specific keys to authenticate with.
/// When set, this option will override `password`.
/// See [`AuthKeys`] for more information.
pub fn auth_keys(&mut self, keys: AuthKeys) -> &mut Config {
self.auth_keys = Some(Box::new(keys));
self
}
/// Gets precomputed protocol-specific keys to authenticate with.
/// if one has been configured with the `auth_keys` method.
pub fn get_auth_keys(&self) -> Option<AuthKeys> {
self.auth_keys.as_deref().copied()
}
/// Sets the name of the database to connect to.
///
/// Defaults to the user.
pub fn dbname(&mut self, dbname: &str) -> &mut Config {
self.set_param("database", dbname)
}
/// Gets the name of the database to connect to, if one has been configured
/// with the `dbname` method.
pub fn db_is_set(&self) -> bool {
self.database
}
pub fn set_param(&mut self, name: &str, value: &str) -> &mut Config {
if name == "database" {
self.database = true;
} else if name == "user" {
self.username = true;
}
self.server_params.insert(name, value);
self
}
/// Sets the SSL configuration.
///
/// Defaults to `prefer`.
pub fn ssl_mode(&mut self, ssl_mode: SslMode) -> &mut Config {
self.ssl_mode = ssl_mode;
self
}
/// Gets the SSL configuration.
pub fn get_ssl_mode(&self) -> SslMode {
self.ssl_mode
}
/// Gets the hosts that have been added to the configuration with `host`.
pub fn get_host(&self) -> &Host {
&self.host
}
/// Gets the ports that have been added to the configuration with `port`.
pub fn get_port(&self) -> u16 {
self.port
}
/// Sets the timeout applied to socket-level connection attempts.
///
/// Note that hostnames can resolve to multiple IP addresses, and this timeout will apply to each address of each
/// host separately. Defaults to no limit.
pub fn connect_timeout(&mut self, connect_timeout: Duration) -> &mut Config {
self.connect_timeout = Some(connect_timeout);
self
}
/// Gets the connection timeout, if one has been set with the
/// `connect_timeout` method.
pub fn get_connect_timeout(&self) -> Option<&Duration> {
self.connect_timeout.as_ref()
}
/// Sets the channel binding behavior.
///
/// Defaults to `prefer`.
pub fn channel_binding(&mut self, channel_binding: ChannelBinding) -> &mut Config {
self.channel_binding = channel_binding;
self
}
/// Gets the channel binding behavior.
pub fn get_channel_binding(&self) -> ChannelBinding {
self.channel_binding
}
/// Opens a connection to a PostgreSQL database.
///
/// Requires the `runtime` Cargo feature (enabled by default).
pub async fn connect<T>(
&self,
tls: T,
) -> Result<(Client, Connection<TcpStream, T::Stream>), Error>
where
T: MakeTlsConnect<TcpStream>,
{
connect(tls, self).await
}
pub async fn connect_raw<S, T>(
&self,
stream: S,
tls: T,
) -> Result<RawConnection<S, T::Stream>, Error>
where
S: AsyncRead + AsyncWrite + Unpin,
T: TlsConnect<S>,
{
connect_raw(stream, tls, self).await
}
}
// Omit password from debug output
impl fmt::Debug for Config {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
struct Redaction {}
impl fmt::Debug for Redaction {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
write!(f, "_")
}
}
f.debug_struct("Config")
.field("password", &self.password.as_ref().map(|_| Redaction {}))
.field("ssl_mode", &self.ssl_mode)
.field("host", &self.host)
.field("port", &self.port)
.field("connect_timeout", &self.connect_timeout)
.field("channel_binding", &self.channel_binding)
.field("server_params", &self.server_params)
.finish()
}
}

View File

@@ -1,75 +0,0 @@
use crate::client::SocketConfig;
use crate::codec::BackendMessage;
use crate::config::Host;
use crate::connect_raw::connect_raw;
use crate::connect_socket::connect_socket;
use crate::tls::{MakeTlsConnect, TlsConnect};
use crate::{Client, Config, Connection, Error, RawConnection};
use postgres_protocol2::message::backend::Message;
use tokio::net::TcpStream;
use tokio::sync::mpsc;
pub async fn connect<T>(
mut tls: T,
config: &Config,
) -> Result<(Client, Connection<TcpStream, T::Stream>), Error>
where
T: MakeTlsConnect<TcpStream>,
{
let hostname = match &config.host {
Host::Tcp(host) => host.as_str(),
};
let tls = tls
.make_tls_connect(hostname)
.map_err(|e| Error::tls(e.into()))?;
match connect_once(&config.host, config.port, tls, config).await {
Ok((client, connection)) => Ok((client, connection)),
Err(e) => Err(e),
}
}
async fn connect_once<T>(
host: &Host,
port: u16,
tls: T,
config: &Config,
) -> Result<(Client, Connection<TcpStream, T::Stream>), Error>
where
T: TlsConnect<TcpStream>,
{
let socket = connect_socket(host, port, config.connect_timeout).await?;
let RawConnection {
stream,
parameters,
delayed_notice,
process_id,
secret_key,
} = connect_raw(socket, tls, config).await?;
let socket_config = SocketConfig {
host: host.clone(),
port,
connect_timeout: config.connect_timeout,
};
let (sender, receiver) = mpsc::unbounded_channel();
let client = Client::new(
sender,
socket_config,
config.ssl_mode,
process_id,
secret_key,
);
// delayed notices are always sent as "Async" messages.
let delayed = delayed_notice
.into_iter()
.map(|m| BackendMessage::Async(Message::NoticeResponse(m)))
.collect();
let connection = Connection::new(stream, delayed, parameters, receiver);
Ok((client, connection))
}

View File

@@ -1,326 +0,0 @@
use crate::codec::{BackendMessage, BackendMessages, FrontendMessage, PostgresCodec};
use crate::config::{self, AuthKeys, Config};
use crate::connect_tls::connect_tls;
use crate::maybe_tls_stream::MaybeTlsStream;
use crate::tls::{TlsConnect, TlsStream};
use crate::Error;
use bytes::BytesMut;
use fallible_iterator::FallibleIterator;
use futures_util::{ready, Sink, SinkExt, Stream, TryStreamExt};
use postgres_protocol2::authentication::sasl;
use postgres_protocol2::authentication::sasl::ScramSha256;
use postgres_protocol2::message::backend::{AuthenticationSaslBody, Message, NoticeResponseBody};
use postgres_protocol2::message::frontend;
use std::collections::HashMap;
use std::io;
use std::pin::Pin;
use std::task::{Context, Poll};
use tokio::io::{AsyncRead, AsyncWrite};
use tokio_util::codec::Framed;
pub struct StartupStream<S, T> {
inner: Framed<MaybeTlsStream<S, T>, PostgresCodec>,
buf: BackendMessages,
delayed_notice: Vec<NoticeResponseBody>,
}
impl<S, T> Sink<FrontendMessage> for StartupStream<S, T>
where
S: AsyncRead + AsyncWrite + Unpin,
T: AsyncRead + AsyncWrite + Unpin,
{
type Error = io::Error;
fn poll_ready(mut self: Pin<&mut Self>, cx: &mut Context<'_>) -> Poll<io::Result<()>> {
Pin::new(&mut self.inner).poll_ready(cx)
}
fn start_send(mut self: Pin<&mut Self>, item: FrontendMessage) -> io::Result<()> {
Pin::new(&mut self.inner).start_send(item)
}
fn poll_flush(mut self: Pin<&mut Self>, cx: &mut Context<'_>) -> Poll<io::Result<()>> {
Pin::new(&mut self.inner).poll_flush(cx)
}
fn poll_close(mut self: Pin<&mut Self>, cx: &mut Context<'_>) -> Poll<io::Result<()>> {
Pin::new(&mut self.inner).poll_close(cx)
}
}
impl<S, T> Stream for StartupStream<S, T>
where
S: AsyncRead + AsyncWrite + Unpin,
T: AsyncRead + AsyncWrite + Unpin,
{
type Item = io::Result<Message>;
fn poll_next(
mut self: Pin<&mut Self>,
cx: &mut Context<'_>,
) -> Poll<Option<io::Result<Message>>> {
loop {
match self.buf.next() {
Ok(Some(message)) => return Poll::Ready(Some(Ok(message))),
Ok(None) => {}
Err(e) => return Poll::Ready(Some(Err(e))),
}
match ready!(Pin::new(&mut self.inner).poll_next(cx)) {
Some(Ok(BackendMessage::Normal { messages, .. })) => self.buf = messages,
Some(Ok(BackendMessage::Async(message))) => return Poll::Ready(Some(Ok(message))),
Some(Err(e)) => return Poll::Ready(Some(Err(e))),
None => return Poll::Ready(None),
}
}
}
}
pub struct RawConnection<S, T> {
pub stream: Framed<MaybeTlsStream<S, T>, PostgresCodec>,
pub parameters: HashMap<String, String>,
pub delayed_notice: Vec<NoticeResponseBody>,
pub process_id: i32,
pub secret_key: i32,
}
pub async fn connect_raw<S, T>(
stream: S,
tls: T,
config: &Config,
) -> Result<RawConnection<S, T::Stream>, Error>
where
S: AsyncRead + AsyncWrite + Unpin,
T: TlsConnect<S>,
{
let stream = connect_tls(stream, config.ssl_mode, tls).await?;
let mut stream = StartupStream {
inner: Framed::new(stream, PostgresCodec),
buf: BackendMessages::empty(),
delayed_notice: Vec::new(),
};
startup(&mut stream, config).await?;
authenticate(&mut stream, config).await?;
let (process_id, secret_key, parameters) = read_info(&mut stream).await?;
Ok(RawConnection {
stream: stream.inner,
parameters,
delayed_notice: stream.delayed_notice,
process_id,
secret_key,
})
}
async fn startup<S, T>(stream: &mut StartupStream<S, T>, config: &Config) -> Result<(), Error>
where
S: AsyncRead + AsyncWrite + Unpin,
T: AsyncRead + AsyncWrite + Unpin,
{
let mut buf = BytesMut::new();
frontend::startup_message(&config.server_params, &mut buf).map_err(Error::encode)?;
stream
.send(FrontendMessage::Raw(buf.freeze()))
.await
.map_err(Error::io)
}
async fn authenticate<S, T>(stream: &mut StartupStream<S, T>, config: &Config) -> Result<(), Error>
where
S: AsyncRead + AsyncWrite + Unpin,
T: TlsStream + Unpin,
{
match stream.try_next().await.map_err(Error::io)? {
Some(Message::AuthenticationOk) => {
can_skip_channel_binding(config)?;
return Ok(());
}
Some(Message::AuthenticationCleartextPassword) => {
can_skip_channel_binding(config)?;
let pass = config
.password
.as_ref()
.ok_or_else(|| Error::config("password missing".into()))?;
authenticate_password(stream, pass).await?;
}
Some(Message::AuthenticationSasl(body)) => {
authenticate_sasl(stream, body, config).await?;
}
Some(Message::AuthenticationMd5Password)
| Some(Message::AuthenticationKerberosV5)
| Some(Message::AuthenticationScmCredential)
| Some(Message::AuthenticationGss)
| Some(Message::AuthenticationSspi) => {
return Err(Error::authentication(
"unsupported authentication method".into(),
))
}
Some(Message::ErrorResponse(body)) => return Err(Error::db(body)),
Some(_) => return Err(Error::unexpected_message()),
None => return Err(Error::closed()),
}
match stream.try_next().await.map_err(Error::io)? {
Some(Message::AuthenticationOk) => Ok(()),
Some(Message::ErrorResponse(body)) => Err(Error::db(body)),
Some(_) => Err(Error::unexpected_message()),
None => Err(Error::closed()),
}
}
fn can_skip_channel_binding(config: &Config) -> Result<(), Error> {
match config.channel_binding {
config::ChannelBinding::Disable | config::ChannelBinding::Prefer => Ok(()),
config::ChannelBinding::Require => Err(Error::authentication(
"server did not use channel binding".into(),
)),
}
}
async fn authenticate_password<S, T>(
stream: &mut StartupStream<S, T>,
password: &[u8],
) -> Result<(), Error>
where
S: AsyncRead + AsyncWrite + Unpin,
T: AsyncRead + AsyncWrite + Unpin,
{
let mut buf = BytesMut::new();
frontend::password_message(password, &mut buf).map_err(Error::encode)?;
stream
.send(FrontendMessage::Raw(buf.freeze()))
.await
.map_err(Error::io)
}
async fn authenticate_sasl<S, T>(
stream: &mut StartupStream<S, T>,
body: AuthenticationSaslBody,
config: &Config,
) -> Result<(), Error>
where
S: AsyncRead + AsyncWrite + Unpin,
T: TlsStream + Unpin,
{
let mut has_scram = false;
let mut has_scram_plus = false;
let mut mechanisms = body.mechanisms();
while let Some(mechanism) = mechanisms.next().map_err(Error::parse)? {
match mechanism {
sasl::SCRAM_SHA_256 => has_scram = true,
sasl::SCRAM_SHA_256_PLUS => has_scram_plus = true,
_ => {}
}
}
let channel_binding = stream
.inner
.get_ref()
.channel_binding()
.tls_server_end_point
.filter(|_| config.channel_binding != config::ChannelBinding::Disable)
.map(sasl::ChannelBinding::tls_server_end_point);
let (channel_binding, mechanism) = if has_scram_plus {
match channel_binding {
Some(channel_binding) => (channel_binding, sasl::SCRAM_SHA_256_PLUS),
None => (sasl::ChannelBinding::unsupported(), sasl::SCRAM_SHA_256),
}
} else if has_scram {
match channel_binding {
Some(_) => (sasl::ChannelBinding::unrequested(), sasl::SCRAM_SHA_256),
None => (sasl::ChannelBinding::unsupported(), sasl::SCRAM_SHA_256),
}
} else {
return Err(Error::authentication("unsupported SASL mechanism".into()));
};
if mechanism != sasl::SCRAM_SHA_256_PLUS {
can_skip_channel_binding(config)?;
}
let mut scram = if let Some(AuthKeys::ScramSha256(keys)) = config.get_auth_keys() {
ScramSha256::new_with_keys(keys, channel_binding)
} else if let Some(password) = config.get_password() {
ScramSha256::new(password, channel_binding)
} else {
return Err(Error::config("password or auth keys missing".into()));
};
let mut buf = BytesMut::new();
frontend::sasl_initial_response(mechanism, scram.message(), &mut buf).map_err(Error::encode)?;
stream
.send(FrontendMessage::Raw(buf.freeze()))
.await
.map_err(Error::io)?;
let body = match stream.try_next().await.map_err(Error::io)? {
Some(Message::AuthenticationSaslContinue(body)) => body,
Some(Message::ErrorResponse(body)) => return Err(Error::db(body)),
Some(_) => return Err(Error::unexpected_message()),
None => return Err(Error::closed()),
};
scram
.update(body.data())
.await
.map_err(|e| Error::authentication(e.into()))?;
let mut buf = BytesMut::new();
frontend::sasl_response(scram.message(), &mut buf).map_err(Error::encode)?;
stream
.send(FrontendMessage::Raw(buf.freeze()))
.await
.map_err(Error::io)?;
let body = match stream.try_next().await.map_err(Error::io)? {
Some(Message::AuthenticationSaslFinal(body)) => body,
Some(Message::ErrorResponse(body)) => return Err(Error::db(body)),
Some(_) => return Err(Error::unexpected_message()),
None => return Err(Error::closed()),
};
scram
.finish(body.data())
.map_err(|e| Error::authentication(e.into()))?;
Ok(())
}
async fn read_info<S, T>(
stream: &mut StartupStream<S, T>,
) -> Result<(i32, i32, HashMap<String, String>), Error>
where
S: AsyncRead + AsyncWrite + Unpin,
T: AsyncRead + AsyncWrite + Unpin,
{
let mut process_id = 0;
let mut secret_key = 0;
let mut parameters = HashMap::new();
loop {
match stream.try_next().await.map_err(Error::io)? {
Some(Message::BackendKeyData(body)) => {
process_id = body.process_id();
secret_key = body.secret_key();
}
Some(Message::ParameterStatus(body)) => {
parameters.insert(
body.name().map_err(Error::parse)?.to_string(),
body.value().map_err(Error::parse)?.to_string(),
);
}
Some(Message::NoticeResponse(body)) => stream.delayed_notice.push(body),
Some(Message::ReadyForQuery(_)) => return Ok((process_id, secret_key, parameters)),
Some(Message::ErrorResponse(body)) => return Err(Error::db(body)),
Some(_) => return Err(Error::unexpected_message()),
None => return Err(Error::closed()),
}
}
}

View File

@@ -1,65 +0,0 @@
use crate::config::Host;
use crate::Error;
use std::future::Future;
use std::io;
use std::time::Duration;
use tokio::net::{self, TcpStream};
use tokio::time;
pub(crate) async fn connect_socket(
host: &Host,
port: u16,
connect_timeout: Option<Duration>,
) -> Result<TcpStream, Error> {
match host {
Host::Tcp(host) => {
let addrs = net::lookup_host((&**host, port))
.await
.map_err(Error::connect)?;
let mut last_err = None;
for addr in addrs {
let stream =
match connect_with_timeout(TcpStream::connect(addr), connect_timeout).await {
Ok(stream) => stream,
Err(e) => {
last_err = Some(e);
continue;
}
};
stream.set_nodelay(true).map_err(Error::connect)?;
return Ok(stream);
}
Err(last_err.unwrap_or_else(|| {
Error::connect(io::Error::new(
io::ErrorKind::InvalidInput,
"could not resolve any addresses",
))
}))
}
}
}
async fn connect_with_timeout<F, T>(connect: F, timeout: Option<Duration>) -> Result<T, Error>
where
F: Future<Output = io::Result<T>>,
{
match timeout {
Some(timeout) => match time::timeout(timeout, connect).await {
Ok(Ok(socket)) => Ok(socket),
Ok(Err(e)) => Err(Error::connect(e)),
Err(_) => Err(Error::connect(io::Error::new(
io::ErrorKind::TimedOut,
"connection timed out",
))),
},
None => match connect.await {
Ok(socket) => Ok(socket),
Err(e) => Err(Error::connect(e)),
},
}
}

View File

@@ -1,48 +0,0 @@
use crate::config::SslMode;
use crate::maybe_tls_stream::MaybeTlsStream;
use crate::tls::private::ForcePrivateApi;
use crate::tls::TlsConnect;
use crate::Error;
use bytes::BytesMut;
use postgres_protocol2::message::frontend;
use tokio::io::{AsyncRead, AsyncReadExt, AsyncWrite, AsyncWriteExt};
pub async fn connect_tls<S, T>(
mut stream: S,
mode: SslMode,
tls: T,
) -> Result<MaybeTlsStream<S, T::Stream>, Error>
where
S: AsyncRead + AsyncWrite + Unpin,
T: TlsConnect<S>,
{
match mode {
SslMode::Disable => return Ok(MaybeTlsStream::Raw(stream)),
SslMode::Prefer if !tls.can_connect(ForcePrivateApi) => {
return Ok(MaybeTlsStream::Raw(stream))
}
SslMode::Prefer | SslMode::Require => {}
}
let mut buf = BytesMut::new();
frontend::ssl_request(&mut buf);
stream.write_all(&buf).await.map_err(Error::io)?;
let mut buf = [0];
stream.read_exact(&mut buf).await.map_err(Error::io)?;
if buf[0] != b'S' {
if SslMode::Require == mode {
return Err(Error::tls("server does not support TLS".into()));
} else {
return Ok(MaybeTlsStream::Raw(stream));
}
}
let stream = tls
.connect(stream)
.await
.map_err(|e| Error::tls(e.into()))?;
Ok(MaybeTlsStream::Tls(stream))
}

View File

@@ -1,323 +0,0 @@
use crate::codec::{BackendMessage, BackendMessages, FrontendMessage, PostgresCodec};
use crate::error::DbError;
use crate::maybe_tls_stream::MaybeTlsStream;
use crate::{AsyncMessage, Error, Notification};
use bytes::BytesMut;
use fallible_iterator::FallibleIterator;
use futures_util::{ready, Sink, Stream};
use log::{info, trace};
use postgres_protocol2::message::backend::Message;
use postgres_protocol2::message::frontend;
use std::collections::{HashMap, VecDeque};
use std::future::Future;
use std::pin::Pin;
use std::task::{Context, Poll};
use tokio::io::{AsyncRead, AsyncWrite};
use tokio::sync::mpsc;
use tokio_util::codec::Framed;
use tokio_util::sync::PollSender;
pub enum RequestMessages {
Single(FrontendMessage),
}
pub struct Request {
pub messages: RequestMessages,
pub sender: mpsc::Sender<BackendMessages>,
}
pub struct Response {
sender: PollSender<BackendMessages>,
}
#[derive(PartialEq, Debug)]
enum State {
Active,
Terminating,
Closing,
}
/// A connection to a PostgreSQL database.
///
/// This is one half of what is returned when a new connection is established. It performs the actual IO with the
/// server, and should generally be spawned off onto an executor to run in the background.
///
/// `Connection` implements `Future`, and only resolves when the connection is closed, either because a fatal error has
/// occurred, or because its associated `Client` has dropped and all outstanding work has completed.
#[must_use = "futures do nothing unless polled"]
pub struct Connection<S, T> {
/// HACK: we need this in the Neon Proxy.
pub stream: Framed<MaybeTlsStream<S, T>, PostgresCodec>,
/// HACK: we need this in the Neon Proxy to forward params.
pub parameters: HashMap<String, String>,
receiver: mpsc::UnboundedReceiver<Request>,
pending_request: Option<RequestMessages>,
pending_responses: VecDeque<BackendMessage>,
responses: VecDeque<Response>,
state: State,
}
impl<S, T> Connection<S, T>
where
S: AsyncRead + AsyncWrite + Unpin,
T: AsyncRead + AsyncWrite + Unpin,
{
pub(crate) fn new(
stream: Framed<MaybeTlsStream<S, T>, PostgresCodec>,
pending_responses: VecDeque<BackendMessage>,
parameters: HashMap<String, String>,
receiver: mpsc::UnboundedReceiver<Request>,
) -> Connection<S, T> {
Connection {
stream,
parameters,
receiver,
pending_request: None,
pending_responses,
responses: VecDeque::new(),
state: State::Active,
}
}
fn poll_response(
&mut self,
cx: &mut Context<'_>,
) -> Poll<Option<Result<BackendMessage, Error>>> {
if let Some(message) = self.pending_responses.pop_front() {
trace!("retrying pending response");
return Poll::Ready(Some(Ok(message)));
}
Pin::new(&mut self.stream)
.poll_next(cx)
.map(|o| o.map(|r| r.map_err(Error::io)))
}
fn poll_read(&mut self, cx: &mut Context<'_>) -> Result<Option<AsyncMessage>, Error> {
if self.state != State::Active {
trace!("poll_read: done");
return Ok(None);
}
loop {
let message = match self.poll_response(cx)? {
Poll::Ready(Some(message)) => message,
Poll::Ready(None) => return Err(Error::closed()),
Poll::Pending => {
trace!("poll_read: waiting on response");
return Ok(None);
}
};
let (mut messages, request_complete) = match message {
BackendMessage::Async(Message::NoticeResponse(body)) => {
let error = DbError::parse(&mut body.fields()).map_err(Error::parse)?;
return Ok(Some(AsyncMessage::Notice(error)));
}
BackendMessage::Async(Message::NotificationResponse(body)) => {
let notification = Notification {
process_id: body.process_id(),
channel: body.channel().map_err(Error::parse)?.to_string(),
payload: body.message().map_err(Error::parse)?.to_string(),
};
return Ok(Some(AsyncMessage::Notification(notification)));
}
BackendMessage::Async(Message::ParameterStatus(body)) => {
self.parameters.insert(
body.name().map_err(Error::parse)?.to_string(),
body.value().map_err(Error::parse)?.to_string(),
);
continue;
}
BackendMessage::Async(_) => unreachable!(),
BackendMessage::Normal {
messages,
request_complete,
} => (messages, request_complete),
};
let mut response = match self.responses.pop_front() {
Some(response) => response,
None => match messages.next().map_err(Error::parse)? {
Some(Message::ErrorResponse(error)) => return Err(Error::db(error)),
_ => return Err(Error::unexpected_message()),
},
};
match response.sender.poll_reserve(cx) {
Poll::Ready(Ok(())) => {
let _ = response.sender.send_item(messages);
if !request_complete {
self.responses.push_front(response);
}
}
Poll::Ready(Err(_)) => {
// we need to keep paging through the rest of the messages even if the receiver's hung up
if !request_complete {
self.responses.push_front(response);
}
}
Poll::Pending => {
self.responses.push_front(response);
self.pending_responses.push_back(BackendMessage::Normal {
messages,
request_complete,
});
trace!("poll_read: waiting on sender");
return Ok(None);
}
}
}
}
fn poll_request(&mut self, cx: &mut Context<'_>) -> Poll<Option<RequestMessages>> {
if let Some(messages) = self.pending_request.take() {
trace!("retrying pending request");
return Poll::Ready(Some(messages));
}
if self.receiver.is_closed() {
return Poll::Ready(None);
}
match self.receiver.poll_recv(cx) {
Poll::Ready(Some(request)) => {
trace!("polled new request");
self.responses.push_back(Response {
sender: PollSender::new(request.sender),
});
Poll::Ready(Some(request.messages))
}
Poll::Ready(None) => Poll::Ready(None),
Poll::Pending => Poll::Pending,
}
}
fn poll_write(&mut self, cx: &mut Context<'_>) -> Result<bool, Error> {
loop {
if self.state == State::Closing {
trace!("poll_write: done");
return Ok(false);
}
if Pin::new(&mut self.stream)
.poll_ready(cx)
.map_err(Error::io)?
.is_pending()
{
trace!("poll_write: waiting on socket");
return Ok(false);
}
let request = match self.poll_request(cx) {
Poll::Ready(Some(request)) => request,
Poll::Ready(None) if self.responses.is_empty() && self.state == State::Active => {
trace!("poll_write: at eof, terminating");
self.state = State::Terminating;
let mut request = BytesMut::new();
frontend::terminate(&mut request);
RequestMessages::Single(FrontendMessage::Raw(request.freeze()))
}
Poll::Ready(None) => {
trace!(
"poll_write: at eof, pending responses {}",
self.responses.len()
);
return Ok(true);
}
Poll::Pending => {
trace!("poll_write: waiting on request");
return Ok(true);
}
};
match request {
RequestMessages::Single(request) => {
Pin::new(&mut self.stream)
.start_send(request)
.map_err(Error::io)?;
if self.state == State::Terminating {
trace!("poll_write: sent eof, closing");
self.state = State::Closing;
}
}
}
}
}
fn poll_flush(&mut self, cx: &mut Context<'_>) -> Result<(), Error> {
match Pin::new(&mut self.stream)
.poll_flush(cx)
.map_err(Error::io)?
{
Poll::Ready(()) => trace!("poll_flush: flushed"),
Poll::Pending => trace!("poll_flush: waiting on socket"),
}
Ok(())
}
fn poll_shutdown(&mut self, cx: &mut Context<'_>) -> Poll<Result<(), Error>> {
if self.state != State::Closing {
return Poll::Pending;
}
match Pin::new(&mut self.stream)
.poll_close(cx)
.map_err(Error::io)?
{
Poll::Ready(()) => {
trace!("poll_shutdown: complete");
Poll::Ready(Ok(()))
}
Poll::Pending => {
trace!("poll_shutdown: waiting on socket");
Poll::Pending
}
}
}
/// Returns the value of a runtime parameter for this connection.
pub fn parameter(&self, name: &str) -> Option<&str> {
self.parameters.get(name).map(|s| &**s)
}
/// Polls for asynchronous messages from the server.
///
/// The server can send notices as well as notifications asynchronously to the client. Applications that wish to
/// examine those messages should use this method to drive the connection rather than its `Future` implementation.
pub fn poll_message(
&mut self,
cx: &mut Context<'_>,
) -> Poll<Option<Result<AsyncMessage, Error>>> {
let message = self.poll_read(cx)?;
let want_flush = self.poll_write(cx)?;
if want_flush {
self.poll_flush(cx)?;
}
match message {
Some(message) => Poll::Ready(Some(Ok(message))),
None => match self.poll_shutdown(cx) {
Poll::Ready(Ok(())) => Poll::Ready(None),
Poll::Ready(Err(e)) => Poll::Ready(Some(Err(e))),
Poll::Pending => Poll::Pending,
},
}
}
}
impl<S, T> Future for Connection<S, T>
where
S: AsyncRead + AsyncWrite + Unpin,
T: AsyncRead + AsyncWrite + Unpin,
{
type Output = Result<(), Error>;
fn poll(mut self: Pin<&mut Self>, cx: &mut Context<'_>) -> Poll<Result<(), Error>> {
while let Some(message) = ready!(self.poll_message(cx)?) {
if let AsyncMessage::Notice(notice) = message {
info!("{}: {}", notice.severity(), notice.message());
}
}
Poll::Ready(Ok(()))
}
}

View File

@@ -1,495 +0,0 @@
//! Errors.
use fallible_iterator::FallibleIterator;
use postgres_protocol2::message::backend::{ErrorFields, ErrorResponseBody};
use std::error::{self, Error as _Error};
use std::fmt;
use std::io;
pub use self::sqlstate::*;
#[allow(clippy::unreadable_literal)]
mod sqlstate;
/// The severity of a Postgres error or notice.
#[derive(Debug, Copy, Clone, PartialEq, Eq)]
pub enum Severity {
/// PANIC
Panic,
/// FATAL
Fatal,
/// ERROR
Error,
/// WARNING
Warning,
/// NOTICE
Notice,
/// DEBUG
Debug,
/// INFO
Info,
/// LOG
Log,
}
impl fmt::Display for Severity {
fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result {
let s = match *self {
Severity::Panic => "PANIC",
Severity::Fatal => "FATAL",
Severity::Error => "ERROR",
Severity::Warning => "WARNING",
Severity::Notice => "NOTICE",
Severity::Debug => "DEBUG",
Severity::Info => "INFO",
Severity::Log => "LOG",
};
fmt.write_str(s)
}
}
impl Severity {
fn from_str(s: &str) -> Option<Severity> {
match s {
"PANIC" => Some(Severity::Panic),
"FATAL" => Some(Severity::Fatal),
"ERROR" => Some(Severity::Error),
"WARNING" => Some(Severity::Warning),
"NOTICE" => Some(Severity::Notice),
"DEBUG" => Some(Severity::Debug),
"INFO" => Some(Severity::Info),
"LOG" => Some(Severity::Log),
_ => None,
}
}
}
/// A Postgres error or notice.
#[derive(Debug, Clone, PartialEq, Eq)]
pub struct DbError {
severity: String,
parsed_severity: Option<Severity>,
code: SqlState,
message: String,
detail: Option<String>,
hint: Option<String>,
position: Option<ErrorPosition>,
where_: Option<String>,
schema: Option<String>,
table: Option<String>,
column: Option<String>,
datatype: Option<String>,
constraint: Option<String>,
file: Option<String>,
line: Option<u32>,
routine: Option<String>,
}
impl DbError {
pub(crate) fn parse(fields: &mut ErrorFields<'_>) -> io::Result<DbError> {
let mut severity = None;
let mut parsed_severity = None;
let mut code = None;
let mut message = None;
let mut detail = None;
let mut hint = None;
let mut normal_position = None;
let mut internal_position = None;
let mut internal_query = None;
let mut where_ = None;
let mut schema = None;
let mut table = None;
let mut column = None;
let mut datatype = None;
let mut constraint = None;
let mut file = None;
let mut line = None;
let mut routine = None;
while let Some(field) = fields.next()? {
match field.type_() {
b'S' => severity = Some(field.value().to_owned()),
b'C' => code = Some(SqlState::from_code(field.value())),
b'M' => message = Some(field.value().to_owned()),
b'D' => detail = Some(field.value().to_owned()),
b'H' => hint = Some(field.value().to_owned()),
b'P' => {
normal_position = Some(field.value().parse::<u32>().map_err(|_| {
io::Error::new(
io::ErrorKind::InvalidInput,
"`P` field did not contain an integer",
)
})?);
}
b'p' => {
internal_position = Some(field.value().parse::<u32>().map_err(|_| {
io::Error::new(
io::ErrorKind::InvalidInput,
"`p` field did not contain an integer",
)
})?);
}
b'q' => internal_query = Some(field.value().to_owned()),
b'W' => where_ = Some(field.value().to_owned()),
b's' => schema = Some(field.value().to_owned()),
b't' => table = Some(field.value().to_owned()),
b'c' => column = Some(field.value().to_owned()),
b'd' => datatype = Some(field.value().to_owned()),
b'n' => constraint = Some(field.value().to_owned()),
b'F' => file = Some(field.value().to_owned()),
b'L' => {
line = Some(field.value().parse::<u32>().map_err(|_| {
io::Error::new(
io::ErrorKind::InvalidInput,
"`L` field did not contain an integer",
)
})?);
}
b'R' => routine = Some(field.value().to_owned()),
b'V' => {
parsed_severity = Some(Severity::from_str(field.value()).ok_or_else(|| {
io::Error::new(
io::ErrorKind::InvalidInput,
"`V` field contained an invalid value",
)
})?);
}
_ => {}
}
}
Ok(DbError {
severity: severity
.ok_or_else(|| io::Error::new(io::ErrorKind::InvalidInput, "`S` field missing"))?,
parsed_severity,
code: code
.ok_or_else(|| io::Error::new(io::ErrorKind::InvalidInput, "`C` field missing"))?,
message: message
.ok_or_else(|| io::Error::new(io::ErrorKind::InvalidInput, "`M` field missing"))?,
detail,
hint,
position: match normal_position {
Some(position) => Some(ErrorPosition::Original(position)),
None => match internal_position {
Some(position) => Some(ErrorPosition::Internal {
position,
query: internal_query.ok_or_else(|| {
io::Error::new(
io::ErrorKind::InvalidInput,
"`q` field missing but `p` field present",
)
})?,
}),
None => None,
},
},
where_,
schema,
table,
column,
datatype,
constraint,
file,
line,
routine,
})
}
/// The field contents are ERROR, FATAL, or PANIC (in an error message),
/// or WARNING, NOTICE, DEBUG, INFO, or LOG (in a notice message), or a
/// localized translation of one of these.
pub fn severity(&self) -> &str {
&self.severity
}
/// A parsed, nonlocalized version of `severity`. (PostgreSQL 9.6+)
pub fn parsed_severity(&self) -> Option<Severity> {
self.parsed_severity
}
/// The SQLSTATE code for the error.
pub fn code(&self) -> &SqlState {
&self.code
}
/// The primary human-readable error message.
///
/// This should be accurate but terse (typically one line).
pub fn message(&self) -> &str {
&self.message
}
/// An optional secondary error message carrying more detail about the
/// problem.
///
/// Might run to multiple lines.
pub fn detail(&self) -> Option<&str> {
self.detail.as_deref()
}
/// An optional suggestion what to do about the problem.
///
/// This is intended to differ from `detail` in that it offers advice
/// (potentially inappropriate) rather than hard facts. Might run to
/// multiple lines.
pub fn hint(&self) -> Option<&str> {
self.hint.as_deref()
}
/// An optional error cursor position into either the original query string
/// or an internally generated query.
pub fn position(&self) -> Option<&ErrorPosition> {
self.position.as_ref()
}
/// An indication of the context in which the error occurred.
///
/// Presently this includes a call stack traceback of active procedural
/// language functions and internally-generated queries. The trace is one
/// entry per line, most recent first.
pub fn where_(&self) -> Option<&str> {
self.where_.as_deref()
}
/// If the error was associated with a specific database object, the name
/// of the schema containing that object, if any. (PostgreSQL 9.3+)
pub fn schema(&self) -> Option<&str> {
self.schema.as_deref()
}
/// If the error was associated with a specific table, the name of the
/// table. (Refer to the schema name field for the name of the table's
/// schema.) (PostgreSQL 9.3+)
pub fn table(&self) -> Option<&str> {
self.table.as_deref()
}
/// If the error was associated with a specific table column, the name of
/// the column.
///
/// (Refer to the schema and table name fields to identify the table.)
/// (PostgreSQL 9.3+)
pub fn column(&self) -> Option<&str> {
self.column.as_deref()
}
/// If the error was associated with a specific data type, the name of the
/// data type. (Refer to the schema name field for the name of the data
/// type's schema.) (PostgreSQL 9.3+)
pub fn datatype(&self) -> Option<&str> {
self.datatype.as_deref()
}
/// If the error was associated with a specific constraint, the name of the
/// constraint.
///
/// Refer to fields listed above for the associated table or domain.
/// (For this purpose, indexes are treated as constraints, even if they
/// weren't created with constraint syntax.) (PostgreSQL 9.3+)
pub fn constraint(&self) -> Option<&str> {
self.constraint.as_deref()
}
/// The file name of the source-code location where the error was reported.
pub fn file(&self) -> Option<&str> {
self.file.as_deref()
}
/// The line number of the source-code location where the error was
/// reported.
pub fn line(&self) -> Option<u32> {
self.line
}
/// The name of the source-code routine reporting the error.
pub fn routine(&self) -> Option<&str> {
self.routine.as_deref()
}
}
impl fmt::Display for DbError {
fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result {
write!(fmt, "{}: {}", self.severity, self.message)?;
if let Some(detail) = &self.detail {
write!(fmt, "\nDETAIL: {}", detail)?;
}
if let Some(hint) = &self.hint {
write!(fmt, "\nHINT: {}", hint)?;
}
Ok(())
}
}
impl error::Error for DbError {}
/// Represents the position of an error in a query.
#[derive(Clone, PartialEq, Eq, Debug)]
pub enum ErrorPosition {
/// A position in the original query.
Original(u32),
/// A position in an internally generated query.
Internal {
/// The byte position.
position: u32,
/// A query generated by the Postgres server.
query: String,
},
}
#[derive(Debug, PartialEq)]
enum Kind {
Io,
UnexpectedMessage,
Tls,
ToSql(usize),
FromSql(usize),
Column(String),
Closed,
Db,
Parse,
Encode,
Authentication,
Config,
Connect,
Timeout,
}
struct ErrorInner {
kind: Kind,
cause: Option<Box<dyn error::Error + Sync + Send>>,
}
/// An error communicating with the Postgres server.
pub struct Error(Box<ErrorInner>);
impl fmt::Debug for Error {
fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result {
fmt.debug_struct("Error")
.field("kind", &self.0.kind)
.field("cause", &self.0.cause)
.finish()
}
}
impl fmt::Display for Error {
fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result {
match &self.0.kind {
Kind::Io => fmt.write_str("error communicating with the server")?,
Kind::UnexpectedMessage => fmt.write_str("unexpected message from server")?,
Kind::Tls => fmt.write_str("error performing TLS handshake")?,
Kind::ToSql(idx) => write!(fmt, "error serializing parameter {}", idx)?,
Kind::FromSql(idx) => write!(fmt, "error deserializing column {}", idx)?,
Kind::Column(column) => write!(fmt, "invalid column `{}`", column)?,
Kind::Closed => fmt.write_str("connection closed")?,
Kind::Db => fmt.write_str("db error")?,
Kind::Parse => fmt.write_str("error parsing response from server")?,
Kind::Encode => fmt.write_str("error encoding message to server")?,
Kind::Authentication => fmt.write_str("authentication error")?,
Kind::Config => fmt.write_str("invalid configuration")?,
Kind::Connect => fmt.write_str("error connecting to server")?,
Kind::Timeout => fmt.write_str("timeout waiting for server")?,
};
if let Some(ref cause) = self.0.cause {
write!(fmt, ": {}", cause)?;
}
Ok(())
}
}
impl error::Error for Error {
fn source(&self) -> Option<&(dyn error::Error + 'static)> {
self.0.cause.as_ref().map(|e| &**e as _)
}
}
impl Error {
/// Consumes the error, returning its cause.
pub fn into_source(self) -> Option<Box<dyn error::Error + Sync + Send>> {
self.0.cause
}
/// Returns the source of this error if it was a `DbError`.
///
/// This is a simple convenience method.
pub fn as_db_error(&self) -> Option<&DbError> {
self.source().and_then(|e| e.downcast_ref::<DbError>())
}
/// Determines if the error was associated with closed connection.
pub fn is_closed(&self) -> bool {
self.0.kind == Kind::Closed
}
/// Returns the SQLSTATE error code associated with the error.
///
/// This is a convenience method that downcasts the cause to a `DbError` and returns its code.
pub fn code(&self) -> Option<&SqlState> {
self.as_db_error().map(DbError::code)
}
fn new(kind: Kind, cause: Option<Box<dyn error::Error + Sync + Send>>) -> Error {
Error(Box::new(ErrorInner { kind, cause }))
}
pub(crate) fn closed() -> Error {
Error::new(Kind::Closed, None)
}
pub(crate) fn unexpected_message() -> Error {
Error::new(Kind::UnexpectedMessage, None)
}
#[allow(clippy::needless_pass_by_value)]
pub(crate) fn db(error: ErrorResponseBody) -> Error {
match DbError::parse(&mut error.fields()) {
Ok(e) => Error::new(Kind::Db, Some(Box::new(e))),
Err(e) => Error::new(Kind::Parse, Some(Box::new(e))),
}
}
pub(crate) fn parse(e: io::Error) -> Error {
Error::new(Kind::Parse, Some(Box::new(e)))
}
pub(crate) fn encode(e: io::Error) -> Error {
Error::new(Kind::Encode, Some(Box::new(e)))
}
#[allow(clippy::wrong_self_convention)]
pub(crate) fn to_sql(e: Box<dyn error::Error + Sync + Send>, idx: usize) -> Error {
Error::new(Kind::ToSql(idx), Some(e))
}
pub(crate) fn from_sql(e: Box<dyn error::Error + Sync + Send>, idx: usize) -> Error {
Error::new(Kind::FromSql(idx), Some(e))
}
pub(crate) fn column(column: String) -> Error {
Error::new(Kind::Column(column), None)
}
pub(crate) fn tls(e: Box<dyn error::Error + Sync + Send>) -> Error {
Error::new(Kind::Tls, Some(e))
}
pub(crate) fn io(e: io::Error) -> Error {
Error::new(Kind::Io, Some(Box::new(e)))
}
pub(crate) fn authentication(e: Box<dyn error::Error + Sync + Send>) -> Error {
Error::new(Kind::Authentication, Some(e))
}
pub(crate) fn config(e: Box<dyn error::Error + Sync + Send>) -> Error {
Error::new(Kind::Config, Some(e))
}
pub(crate) fn connect(e: io::Error) -> Error {
Error::new(Kind::Connect, Some(Box::new(e)))
}
#[doc(hidden)]
pub fn __private_api_timeout() -> Error {
Error::new(Kind::Timeout, None)
}
}

File diff suppressed because it is too large Load Diff

View File

@@ -1,64 +0,0 @@
use crate::query::{self, RowStream};
use crate::types::Type;
use crate::{Client, Error, Transaction};
use async_trait::async_trait;
use postgres_protocol2::Oid;
mod private {
pub trait Sealed {}
}
/// A trait allowing abstraction over connections and transactions.
///
/// This trait is "sealed", and cannot be implemented outside of this crate.
#[async_trait]
pub trait GenericClient: private::Sealed {
async fn query_raw_txt<S, I>(&mut self, statement: &str, params: I) -> Result<RowStream, Error>
where
S: AsRef<str> + Sync + Send,
I: IntoIterator<Item = Option<S>> + Sync + Send,
I::IntoIter: ExactSizeIterator + Sync + Send;
/// Query for type information
async fn get_type(&mut self, oid: Oid) -> Result<Type, Error>;
}
impl private::Sealed for Client {}
#[async_trait]
impl GenericClient for Client {
async fn query_raw_txt<S, I>(&mut self, statement: &str, params: I) -> Result<RowStream, Error>
where
S: AsRef<str> + Sync + Send,
I: IntoIterator<Item = Option<S>> + Sync + Send,
I::IntoIter: ExactSizeIterator + Sync + Send,
{
query::query_txt(&mut self.inner, statement, params).await
}
/// Query for type information
async fn get_type(&mut self, oid: Oid) -> Result<Type, Error> {
crate::prepare::get_type(&mut self.inner, &mut self.cached_typeinfo, oid).await
}
}
impl private::Sealed for Transaction<'_> {}
#[async_trait]
#[allow(clippy::needless_lifetimes)]
impl GenericClient for Transaction<'_> {
async fn query_raw_txt<S, I>(&mut self, statement: &str, params: I) -> Result<RowStream, Error>
where
S: AsRef<str> + Sync + Send,
I: IntoIterator<Item = Option<S>> + Sync + Send,
I::IntoIter: ExactSizeIterator + Sync + Send,
{
query::query_txt(&mut self.client().inner, statement, params).await
}
/// Query for type information
async fn get_type(&mut self, oid: Oid) -> Result<Type, Error> {
let client = self.client();
crate::prepare::get_type(&mut client.inner, &mut client.cached_typeinfo, oid).await
}
}

View File

@@ -1,115 +0,0 @@
//! An asynchronous, pipelined, PostgreSQL client.
#![warn(rust_2018_idioms, clippy::all)]
pub use crate::cancel_token::CancelToken;
pub use crate::client::{Client, SocketConfig};
pub use crate::config::Config;
pub use crate::connect_raw::RawConnection;
pub use crate::connection::Connection;
use crate::error::DbError;
pub use crate::error::Error;
pub use crate::generic_client::GenericClient;
pub use crate::query::RowStream;
pub use crate::row::Row;
pub use crate::statement::{Column, Statement};
pub use crate::tls::NoTls;
// pub use crate::to_statement::ToStatement;
pub use crate::transaction::Transaction;
pub use crate::transaction_builder::{IsolationLevel, TransactionBuilder};
use crate::types::ToSql;
use postgres_protocol2::message::backend::ReadyForQueryBody;
/// After executing a query, the connection will be in one of these states
#[derive(Clone, Copy, Debug, PartialEq)]
#[repr(u8)]
pub enum ReadyForQueryStatus {
/// Connection state is unknown
Unknown,
/// Connection is idle (no transactions)
Idle = b'I',
/// Connection is in a transaction block
Transaction = b'T',
/// Connection is in a failed transaction block
FailedTransaction = b'E',
}
impl From<ReadyForQueryBody> for ReadyForQueryStatus {
fn from(value: ReadyForQueryBody) -> Self {
match value.status() {
b'I' => Self::Idle,
b'T' => Self::Transaction,
b'E' => Self::FailedTransaction,
_ => Self::Unknown,
}
}
}
mod cancel_query;
mod cancel_query_raw;
mod cancel_token;
mod client;
mod codec;
pub mod config;
mod connect;
mod connect_raw;
mod connect_socket;
mod connect_tls;
mod connection;
pub mod error;
mod generic_client;
pub mod maybe_tls_stream;
mod prepare;
mod query;
pub mod row;
mod simple_query;
mod statement;
pub mod tls;
// mod to_statement;
mod transaction;
mod transaction_builder;
pub mod types;
/// An asynchronous notification.
#[derive(Clone, Debug)]
pub struct Notification {
process_id: i32,
channel: String,
payload: String,
}
impl Notification {
/// The process ID of the notifying backend process.
pub fn process_id(&self) -> i32 {
self.process_id
}
/// The name of the channel that the notify has been raised on.
pub fn channel(&self) -> &str {
&self.channel
}
/// The "payload" string passed from the notifying process.
pub fn payload(&self) -> &str {
&self.payload
}
}
/// An asynchronous message from the server.
#[allow(clippy::large_enum_variant)]
#[derive(Debug, Clone)]
pub enum AsyncMessage {
/// A notice.
///
/// Notices use the same format as errors, but aren't "errors" per-se.
Notice(DbError),
/// A notification.
///
/// Connections can subscribe to notifications with the `LISTEN` command.
Notification(Notification),
}
fn slice_iter<'a>(
s: &'a [&'a (dyn ToSql + Sync)],
) -> impl ExactSizeIterator<Item = &'a (dyn ToSql + Sync)> + 'a {
s.iter().map(|s| *s as _)
}

View File

@@ -1,77 +0,0 @@
//! MaybeTlsStream.
//!
//! Represents a stream that may or may not be encrypted with TLS.
use crate::tls::{ChannelBinding, TlsStream};
use std::io;
use std::pin::Pin;
use std::task::{Context, Poll};
use tokio::io::{AsyncRead, AsyncWrite, ReadBuf};
/// A stream that may or may not be encrypted with TLS.
pub enum MaybeTlsStream<S, T> {
/// An unencrypted stream.
Raw(S),
/// An encrypted stream.
Tls(T),
}
impl<S, T> AsyncRead for MaybeTlsStream<S, T>
where
S: AsyncRead + Unpin,
T: AsyncRead + Unpin,
{
fn poll_read(
mut self: Pin<&mut Self>,
cx: &mut Context<'_>,
buf: &mut ReadBuf<'_>,
) -> Poll<io::Result<()>> {
match &mut *self {
MaybeTlsStream::Raw(s) => Pin::new(s).poll_read(cx, buf),
MaybeTlsStream::Tls(s) => Pin::new(s).poll_read(cx, buf),
}
}
}
impl<S, T> AsyncWrite for MaybeTlsStream<S, T>
where
S: AsyncWrite + Unpin,
T: AsyncWrite + Unpin,
{
fn poll_write(
mut self: Pin<&mut Self>,
cx: &mut Context<'_>,
buf: &[u8],
) -> Poll<io::Result<usize>> {
match &mut *self {
MaybeTlsStream::Raw(s) => Pin::new(s).poll_write(cx, buf),
MaybeTlsStream::Tls(s) => Pin::new(s).poll_write(cx, buf),
}
}
fn poll_flush(mut self: Pin<&mut Self>, cx: &mut Context<'_>) -> Poll<io::Result<()>> {
match &mut *self {
MaybeTlsStream::Raw(s) => Pin::new(s).poll_flush(cx),
MaybeTlsStream::Tls(s) => Pin::new(s).poll_flush(cx),
}
}
fn poll_shutdown(mut self: Pin<&mut Self>, cx: &mut Context<'_>) -> Poll<io::Result<()>> {
match &mut *self {
MaybeTlsStream::Raw(s) => Pin::new(s).poll_shutdown(cx),
MaybeTlsStream::Tls(s) => Pin::new(s).poll_shutdown(cx),
}
}
}
impl<S, T> TlsStream for MaybeTlsStream<S, T>
where
S: AsyncRead + AsyncWrite + Unpin,
T: TlsStream + Unpin,
{
fn channel_binding(&self) -> ChannelBinding {
match self {
MaybeTlsStream::Raw(_) => ChannelBinding::none(),
MaybeTlsStream::Tls(s) => s.channel_binding(),
}
}
}

View File

@@ -1,295 +0,0 @@
use crate::client::{CachedTypeInfo, InnerClient};
use crate::codec::FrontendMessage;
use crate::connection::RequestMessages;
use crate::error::SqlState;
use crate::types::{Field, Kind, Oid, Type};
use crate::{query, slice_iter};
use crate::{Column, Error, Statement};
use bytes::Bytes;
use fallible_iterator::FallibleIterator;
use futures_util::{pin_mut, StreamExt, TryStreamExt};
use log::debug;
use postgres_protocol2::message::backend::Message;
use postgres_protocol2::message::frontend;
use std::future::Future;
use std::pin::{pin, Pin};
use std::sync::atomic::{AtomicUsize, Ordering};
pub(crate) const TYPEINFO_QUERY: &str = "\
SELECT t.typname, t.typtype, t.typelem, r.rngsubtype, t.typbasetype, n.nspname, t.typrelid
FROM pg_catalog.pg_type t
LEFT OUTER JOIN pg_catalog.pg_range r ON r.rngtypid = t.oid
INNER JOIN pg_catalog.pg_namespace n ON t.typnamespace = n.oid
WHERE t.oid = $1
";
// Range types weren't added until Postgres 9.2, so pg_range may not exist
const TYPEINFO_FALLBACK_QUERY: &str = "\
SELECT t.typname, t.typtype, t.typelem, NULL::OID, t.typbasetype, n.nspname, t.typrelid
FROM pg_catalog.pg_type t
INNER JOIN pg_catalog.pg_namespace n ON t.typnamespace = n.oid
WHERE t.oid = $1
";
const TYPEINFO_ENUM_QUERY: &str = "\
SELECT enumlabel
FROM pg_catalog.pg_enum
WHERE enumtypid = $1
ORDER BY enumsortorder
";
// Postgres 9.0 didn't have enumsortorder
const TYPEINFO_ENUM_FALLBACK_QUERY: &str = "\
SELECT enumlabel
FROM pg_catalog.pg_enum
WHERE enumtypid = $1
ORDER BY oid
";
pub(crate) const TYPEINFO_COMPOSITE_QUERY: &str = "\
SELECT attname, atttypid
FROM pg_catalog.pg_attribute
WHERE attrelid = $1
AND NOT attisdropped
AND attnum > 0
ORDER BY attnum
";
static NEXT_ID: AtomicUsize = AtomicUsize::new(0);
pub async fn prepare(
client: &mut InnerClient,
cache: &mut CachedTypeInfo,
query: &str,
types: &[Type],
) -> Result<Statement, Error> {
let name = format!("s{}", NEXT_ID.fetch_add(1, Ordering::SeqCst));
let buf = encode(client, &name, query, types)?;
let mut responses = client.send(RequestMessages::Single(FrontendMessage::Raw(buf)))?;
match responses.next().await? {
Message::ParseComplete => {}
_ => return Err(Error::unexpected_message()),
}
let parameter_description = match responses.next().await? {
Message::ParameterDescription(body) => body,
_ => return Err(Error::unexpected_message()),
};
let row_description = match responses.next().await? {
Message::RowDescription(body) => Some(body),
Message::NoData => None,
_ => return Err(Error::unexpected_message()),
};
let mut parameters = vec![];
let mut it = parameter_description.parameters();
while let Some(oid) = it.next().map_err(Error::parse)? {
let type_ = get_type(client, cache, oid).await?;
parameters.push(type_);
}
let mut columns = vec![];
if let Some(row_description) = row_description {
let mut it = row_description.fields();
while let Some(field) = it.next().map_err(Error::parse)? {
let type_ = get_type(client, cache, field.type_oid()).await?;
let column = Column::new(field.name().to_string(), type_, field);
columns.push(column);
}
}
Ok(Statement::new(name, parameters, columns))
}
fn prepare_rec<'a>(
client: &'a mut InnerClient,
cache: &'a mut CachedTypeInfo,
query: &'a str,
types: &'a [Type],
) -> Pin<Box<dyn Future<Output = Result<Statement, Error>> + 'a + Send>> {
Box::pin(prepare(client, cache, query, types))
}
fn encode(
client: &mut InnerClient,
name: &str,
query: &str,
types: &[Type],
) -> Result<Bytes, Error> {
if types.is_empty() {
debug!("preparing query {}: {}", name, query);
} else {
debug!("preparing query {} with types {:?}: {}", name, types, query);
}
client.with_buf(|buf| {
frontend::parse(name, query, types.iter().map(Type::oid), buf).map_err(Error::encode)?;
frontend::describe(b'S', name, buf).map_err(Error::encode)?;
frontend::sync(buf);
Ok(buf.split().freeze())
})
}
pub async fn get_type(
client: &mut InnerClient,
cache: &mut CachedTypeInfo,
oid: Oid,
) -> Result<Type, Error> {
if let Some(type_) = Type::from_oid(oid) {
return Ok(type_);
}
if let Some(type_) = cache.type_(oid) {
return Ok(type_);
}
let stmt = typeinfo_statement(client, cache).await?;
let rows = query::query(client, stmt, slice_iter(&[&oid])).await?;
pin_mut!(rows);
let row = match rows.try_next().await? {
Some(row) => row,
None => return Err(Error::unexpected_message()),
};
let name: String = row.try_get(stmt.columns(), 0)?;
let type_: i8 = row.try_get(stmt.columns(), 1)?;
let elem_oid: Oid = row.try_get(stmt.columns(), 2)?;
let rngsubtype: Option<Oid> = row.try_get(stmt.columns(), 3)?;
let basetype: Oid = row.try_get(stmt.columns(), 4)?;
let schema: String = row.try_get(stmt.columns(), 5)?;
let relid: Oid = row.try_get(stmt.columns(), 6)?;
let kind = if type_ == b'e' as i8 {
let variants = get_enum_variants(client, cache, oid).await?;
Kind::Enum(variants)
} else if type_ == b'p' as i8 {
Kind::Pseudo
} else if basetype != 0 {
let type_ = get_type_rec(client, cache, basetype).await?;
Kind::Domain(type_)
} else if elem_oid != 0 {
let type_ = get_type_rec(client, cache, elem_oid).await?;
Kind::Array(type_)
} else if relid != 0 {
let fields = get_composite_fields(client, cache, relid).await?;
Kind::Composite(fields)
} else if let Some(rngsubtype) = rngsubtype {
let type_ = get_type_rec(client, cache, rngsubtype).await?;
Kind::Range(type_)
} else {
Kind::Simple
};
let type_ = Type::new(name, oid, kind, schema);
cache.set_type(oid, &type_);
Ok(type_)
}
fn get_type_rec<'a>(
client: &'a mut InnerClient,
cache: &'a mut CachedTypeInfo,
oid: Oid,
) -> Pin<Box<dyn Future<Output = Result<Type, Error>> + Send + 'a>> {
Box::pin(get_type(client, cache, oid))
}
async fn typeinfo_statement<'c>(
client: &mut InnerClient,
cache: &'c mut CachedTypeInfo,
) -> Result<&'c Statement, Error> {
if cache.typeinfo().is_some() {
// needed to get around a borrow checker limitation
return Ok(cache.typeinfo().unwrap());
}
let stmt = match prepare_rec(client, cache, TYPEINFO_QUERY, &[]).await {
Ok(stmt) => stmt,
Err(ref e) if e.code() == Some(&SqlState::UNDEFINED_TABLE) => {
prepare_rec(client, cache, TYPEINFO_FALLBACK_QUERY, &[]).await?
}
Err(e) => return Err(e),
};
Ok(cache.set_typeinfo(stmt))
}
async fn get_enum_variants(
client: &mut InnerClient,
cache: &mut CachedTypeInfo,
oid: Oid,
) -> Result<Vec<String>, Error> {
let stmt = typeinfo_enum_statement(client, cache).await?;
let mut out = vec![];
let mut rows = pin!(query::query(client, stmt, slice_iter(&[&oid])).await?);
while let Some(row) = rows.next().await {
out.push(row?.try_get(stmt.columns(), 0)?)
}
Ok(out)
}
async fn typeinfo_enum_statement<'c>(
client: &mut InnerClient,
cache: &'c mut CachedTypeInfo,
) -> Result<&'c Statement, Error> {
if cache.typeinfo_enum().is_some() {
// needed to get around a borrow checker limitation
return Ok(cache.typeinfo_enum().unwrap());
}
let stmt = match prepare_rec(client, cache, TYPEINFO_ENUM_QUERY, &[]).await {
Ok(stmt) => stmt,
Err(ref e) if e.code() == Some(&SqlState::UNDEFINED_COLUMN) => {
prepare_rec(client, cache, TYPEINFO_ENUM_FALLBACK_QUERY, &[]).await?
}
Err(e) => return Err(e),
};
Ok(cache.set_typeinfo_enum(stmt))
}
async fn get_composite_fields(
client: &mut InnerClient,
cache: &mut CachedTypeInfo,
oid: Oid,
) -> Result<Vec<Field>, Error> {
let stmt = typeinfo_composite_statement(client, cache).await?;
let mut rows = pin!(query::query(client, stmt, slice_iter(&[&oid])).await?);
let mut oids = vec![];
while let Some(row) = rows.next().await {
let row = row?;
let name = row.try_get(stmt.columns(), 0)?;
let oid = row.try_get(stmt.columns(), 1)?;
oids.push((name, oid));
}
let mut fields = vec![];
for (name, oid) in oids {
let type_ = get_type_rec(client, cache, oid).await?;
fields.push(Field::new(name, type_));
}
Ok(fields)
}
async fn typeinfo_composite_statement<'c>(
client: &mut InnerClient,
cache: &'c mut CachedTypeInfo,
) -> Result<&'c Statement, Error> {
if cache.typeinfo_composite().is_some() {
// needed to get around a borrow checker limitation
return Ok(cache.typeinfo_composite().unwrap());
}
let stmt = prepare_rec(client, cache, TYPEINFO_COMPOSITE_QUERY, &[]).await?;
Ok(cache.set_typeinfo_composite(stmt))
}

View File

@@ -1,333 +0,0 @@
use crate::client::{InnerClient, Responses};
use crate::codec::FrontendMessage;
use crate::connection::RequestMessages;
use crate::types::IsNull;
use crate::{Column, Error, ReadyForQueryStatus, Row, Statement};
use bytes::{BufMut, Bytes, BytesMut};
use fallible_iterator::FallibleIterator;
use futures_util::{ready, Stream};
use log::{debug, log_enabled, Level};
use pin_project_lite::pin_project;
use postgres_protocol2::message::backend::Message;
use postgres_protocol2::message::frontend;
use postgres_types2::{Format, ToSql, Type};
use std::fmt;
use std::marker::PhantomPinned;
use std::pin::Pin;
use std::task::{Context, Poll};
struct BorrowToSqlParamsDebug<'a>(&'a [&'a (dyn ToSql + Sync)]);
impl fmt::Debug for BorrowToSqlParamsDebug<'_> {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
f.debug_list().entries(self.0.iter()).finish()
}
}
pub async fn query<'a, I>(
client: &mut InnerClient,
statement: &Statement,
params: I,
) -> Result<RawRowStream, Error>
where
I: IntoIterator<Item = &'a (dyn ToSql + Sync)>,
I::IntoIter: ExactSizeIterator,
{
let buf = if log_enabled!(Level::Debug) {
let params = params.into_iter().collect::<Vec<_>>();
debug!(
"executing statement {} with parameters: {:?}",
statement.name(),
BorrowToSqlParamsDebug(params.as_slice()),
);
encode(client, statement, params)?
} else {
encode(client, statement, params)?
};
let responses = start(client, buf).await?;
Ok(RawRowStream {
responses,
command_tag: None,
status: ReadyForQueryStatus::Unknown,
output_format: Format::Binary,
_p: PhantomPinned,
})
}
pub async fn query_txt<S, I>(
client: &mut InnerClient,
query: &str,
params: I,
) -> Result<RowStream, Error>
where
S: AsRef<str>,
I: IntoIterator<Item = Option<S>>,
I::IntoIter: ExactSizeIterator,
{
let params = params.into_iter();
let buf = client.with_buf(|buf| {
frontend::parse(
"", // unnamed prepared statement
query, // query to parse
std::iter::empty(), // give no type info
buf,
)
.map_err(Error::encode)?;
frontend::describe(b'S', "", buf).map_err(Error::encode)?;
// Bind, pass params as text, retrieve as binary
match frontend::bind(
"", // empty string selects the unnamed portal
"", // unnamed prepared statement
std::iter::empty(), // all parameters use the default format (text)
params,
|param, buf| match param {
Some(param) => {
buf.put_slice(param.as_ref().as_bytes());
Ok(postgres_protocol2::IsNull::No)
}
None => Ok(postgres_protocol2::IsNull::Yes),
},
Some(0), // all text
buf,
) {
Ok(()) => Ok(()),
Err(frontend::BindError::Conversion(e)) => Err(Error::to_sql(e, 0)),
Err(frontend::BindError::Serialization(e)) => Err(Error::encode(e)),
}?;
// Execute
frontend::execute("", 0, buf).map_err(Error::encode)?;
// Sync
frontend::sync(buf);
Ok(buf.split().freeze())
})?;
// now read the responses
let mut responses = client.send(RequestMessages::Single(FrontendMessage::Raw(buf)))?;
match responses.next().await? {
Message::ParseComplete => {}
_ => return Err(Error::unexpected_message()),
}
let parameter_description = match responses.next().await? {
Message::ParameterDescription(body) => body,
_ => return Err(Error::unexpected_message()),
};
let row_description = match responses.next().await? {
Message::RowDescription(body) => Some(body),
Message::NoData => None,
_ => return Err(Error::unexpected_message()),
};
match responses.next().await? {
Message::BindComplete => {}
_ => return Err(Error::unexpected_message()),
}
let mut parameters = vec![];
let mut it = parameter_description.parameters();
while let Some(oid) = it.next().map_err(Error::parse)? {
let type_ = Type::from_oid(oid).unwrap_or(Type::UNKNOWN);
parameters.push(type_);
}
let mut columns = vec![];
if let Some(row_description) = row_description {
let mut it = row_description.fields();
while let Some(field) = it.next().map_err(Error::parse)? {
let type_ = Type::from_oid(field.type_oid()).unwrap_or(Type::UNKNOWN);
let column = Column::new(field.name().to_string(), type_, field);
columns.push(column);
}
}
Ok(RowStream {
statement: Statement::new_anonymous(parameters, columns),
responses,
command_tag: None,
status: ReadyForQueryStatus::Unknown,
output_format: Format::Text,
_p: PhantomPinned,
})
}
async fn start(client: &InnerClient, buf: Bytes) -> Result<Responses, Error> {
let mut responses = client.send(RequestMessages::Single(FrontendMessage::Raw(buf)))?;
match responses.next().await? {
Message::BindComplete => {}
_ => return Err(Error::unexpected_message()),
}
Ok(responses)
}
pub fn encode<'a, I>(
client: &mut InnerClient,
statement: &Statement,
params: I,
) -> Result<Bytes, Error>
where
I: IntoIterator<Item = &'a (dyn ToSql + Sync)>,
I::IntoIter: ExactSizeIterator,
{
client.with_buf(|buf| {
encode_bind(statement, params, "", buf)?;
frontend::execute("", 0, buf).map_err(Error::encode)?;
frontend::sync(buf);
Ok(buf.split().freeze())
})
}
pub fn encode_bind<'a, I>(
statement: &Statement,
params: I,
portal: &str,
buf: &mut BytesMut,
) -> Result<(), Error>
where
I: IntoIterator<Item = &'a (dyn ToSql + Sync)>,
I::IntoIter: ExactSizeIterator,
{
let param_types = statement.params();
let params = params.into_iter();
assert!(
param_types.len() == params.len(),
"expected {} parameters but got {}",
param_types.len(),
params.len()
);
let (param_formats, params): (Vec<_>, Vec<_>) = params
.zip(param_types.iter())
.map(|(p, ty)| (p.encode_format(ty) as i16, p))
.unzip();
let params = params.into_iter();
let mut error_idx = 0;
let r = frontend::bind(
portal,
statement.name(),
param_formats,
params.zip(param_types).enumerate(),
|(idx, (param, ty)), buf| match param.to_sql_checked(ty, buf) {
Ok(IsNull::No) => Ok(postgres_protocol2::IsNull::No),
Ok(IsNull::Yes) => Ok(postgres_protocol2::IsNull::Yes),
Err(e) => {
error_idx = idx;
Err(e)
}
},
Some(1),
buf,
);
match r {
Ok(()) => Ok(()),
Err(frontend::BindError::Conversion(e)) => Err(Error::to_sql(e, error_idx)),
Err(frontend::BindError::Serialization(e)) => Err(Error::encode(e)),
}
}
pin_project! {
/// A stream of table rows.
pub struct RowStream {
statement: Statement,
responses: Responses,
command_tag: Option<String>,
output_format: Format,
status: ReadyForQueryStatus,
#[pin]
_p: PhantomPinned,
}
}
impl Stream for RowStream {
type Item = Result<Row, Error>;
fn poll_next(self: Pin<&mut Self>, cx: &mut Context<'_>) -> Poll<Option<Self::Item>> {
let this = self.project();
loop {
match ready!(this.responses.poll_next(cx)?) {
Message::DataRow(body) => {
return Poll::Ready(Some(Ok(Row::new(body, *this.output_format)?)))
}
Message::EmptyQueryResponse | Message::PortalSuspended => {}
Message::CommandComplete(body) => {
if let Ok(tag) = body.tag() {
*this.command_tag = Some(tag.to_string());
}
}
Message::ReadyForQuery(status) => {
*this.status = status.into();
return Poll::Ready(None);
}
_ => return Poll::Ready(Some(Err(Error::unexpected_message()))),
}
}
}
}
impl RowStream {
/// Returns information about the columns of data in the row.
pub fn columns(&self) -> &[Column] {
self.statement.columns()
}
/// Returns the command tag of this query.
///
/// This is only available after the stream has been exhausted.
pub fn command_tag(&self) -> Option<String> {
self.command_tag.clone()
}
/// Returns if the connection is ready for querying, with the status of the connection.
///
/// This might be available only after the stream has been exhausted.
pub fn ready_status(&self) -> ReadyForQueryStatus {
self.status
}
}
pin_project! {
/// A stream of table rows.
pub struct RawRowStream {
responses: Responses,
command_tag: Option<String>,
output_format: Format,
status: ReadyForQueryStatus,
#[pin]
_p: PhantomPinned,
}
}
impl Stream for RawRowStream {
type Item = Result<Row, Error>;
fn poll_next(self: Pin<&mut Self>, cx: &mut Context<'_>) -> Poll<Option<Self::Item>> {
let this = self.project();
loop {
match ready!(this.responses.poll_next(cx)?) {
Message::DataRow(body) => {
return Poll::Ready(Some(Ok(Row::new(body, *this.output_format)?)))
}
Message::EmptyQueryResponse | Message::PortalSuspended => {}
Message::CommandComplete(body) => {
if let Ok(tag) = body.tag() {
*this.command_tag = Some(tag.to_string());
}
}
Message::ReadyForQuery(status) => {
*this.status = status.into();
return Poll::Ready(None);
}
_ => return Poll::Ready(Some(Err(Error::unexpected_message()))),
}
}
}
}

View File

@@ -1,84 +0,0 @@
//! Rows.
use crate::statement::Column;
use crate::types::{FromSql, Type, WrongType};
use crate::Error;
use fallible_iterator::FallibleIterator;
use postgres_protocol2::message::backend::DataRowBody;
use postgres_types2::{Format, WrongFormat};
use std::fmt;
use std::ops::Range;
use std::str;
/// A row of data returned from the database by a query.
pub struct Row {
output_format: Format,
body: DataRowBody,
ranges: Vec<Option<Range<usize>>>,
}
impl fmt::Debug for Row {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
f.debug_struct("Row").finish()
}
}
impl Row {
pub(crate) fn new(
// statement: Statement,
body: DataRowBody,
output_format: Format,
) -> Result<Row, Error> {
let ranges = body.ranges().collect().map_err(Error::parse)?;
Ok(Row {
body,
ranges,
output_format,
})
}
pub(crate) fn try_get<'a, T>(&'a self, columns: &[Column], idx: usize) -> Result<T, Error>
where
T: FromSql<'a>,
{
let Some(column) = columns.get(idx) else {
return Err(Error::column(idx.to_string()));
};
let ty = column.type_();
if !T::accepts(ty) {
return Err(Error::from_sql(
Box::new(WrongType::new::<T>(ty.clone())),
idx,
));
}
FromSql::from_sql_nullable(ty, self.col_buffer(idx)).map_err(|e| Error::from_sql(e, idx))
}
/// Get the raw bytes for the column at the given index.
fn col_buffer(&self, idx: usize) -> Option<&[u8]> {
let range = self.ranges.get(idx)?.to_owned()?;
Some(&self.body.buffer()[range])
}
/// Interpret the column at the given index as text
///
/// Useful when using query_raw_txt() which sets text transfer mode
pub fn as_text(&self, idx: usize) -> Result<Option<&str>, Error> {
if self.output_format == Format::Text {
match self.col_buffer(idx) {
Some(raw) => {
FromSql::from_sql(&Type::TEXT, raw).map_err(|e| Error::from_sql(e, idx))
}
None => Ok(None),
}
} else {
Err(Error::from_sql(Box::new(WrongFormat {}), idx))
}
}
/// Row byte size
pub fn body_len(&self) -> usize {
self.body.buffer().len()
}
}

View File

@@ -1,36 +0,0 @@
use crate::client::InnerClient;
use crate::codec::FrontendMessage;
use crate::connection::RequestMessages;
use crate::{Error, ReadyForQueryStatus};
use bytes::Bytes;
use log::debug;
use postgres_protocol2::message::backend::Message;
use postgres_protocol2::message::frontend;
pub async fn batch_execute(
client: &mut InnerClient,
query: &str,
) -> Result<ReadyForQueryStatus, Error> {
debug!("executing statement batch: {}", query);
let buf = encode(client, query)?;
let mut responses = client.send(RequestMessages::Single(FrontendMessage::Raw(buf)))?;
loop {
match responses.next().await? {
Message::ReadyForQuery(status) => return Ok(status.into()),
Message::CommandComplete(_)
| Message::EmptyQueryResponse
| Message::RowDescription(_)
| Message::DataRow(_) => {}
_ => return Err(Error::unexpected_message()),
}
}
}
pub(crate) fn encode(client: &mut InnerClient, query: &str) -> Result<Bytes, Error> {
client.with_buf(|buf| {
frontend::query(query, buf).map_err(Error::encode)?;
Ok(buf.split().freeze())
})
}

View File

@@ -1,126 +0,0 @@
use crate::types::Type;
use postgres_protocol2::{message::backend::Field, Oid};
use std::fmt;
struct StatementInner {
name: String,
params: Vec<Type>,
columns: Vec<Column>,
}
/// A prepared statement.
///
/// Prepared statements can only be used with the connection that created them.
pub struct Statement(StatementInner);
impl Statement {
pub(crate) fn new(name: String, params: Vec<Type>, columns: Vec<Column>) -> Statement {
Statement(StatementInner {
name,
params,
columns,
})
}
pub(crate) fn new_anonymous(params: Vec<Type>, columns: Vec<Column>) -> Statement {
Statement(StatementInner {
name: String::new(),
params,
columns,
})
}
pub(crate) fn name(&self) -> &str {
&self.0.name
}
/// Returns the expected types of the statement's parameters.
pub fn params(&self) -> &[Type] {
&self.0.params
}
/// Returns information about the columns returned when the statement is queried.
pub fn columns(&self) -> &[Column] {
&self.0.columns
}
}
/// Information about a column of a query.
pub struct Column {
name: String,
type_: Type,
// raw fields from RowDescription
table_oid: Oid,
column_id: i16,
format: i16,
// that better be stored in self.type_, but that is more radical refactoring
type_oid: Oid,
type_size: i16,
type_modifier: i32,
}
impl Column {
pub(crate) fn new(name: String, type_: Type, raw_field: Field<'_>) -> Column {
Column {
name,
type_,
table_oid: raw_field.table_oid(),
column_id: raw_field.column_id(),
format: raw_field.format(),
type_oid: raw_field.type_oid(),
type_size: raw_field.type_size(),
type_modifier: raw_field.type_modifier(),
}
}
/// Returns the name of the column.
pub fn name(&self) -> &str {
&self.name
}
/// Returns the type of the column.
pub fn type_(&self) -> &Type {
&self.type_
}
/// Returns the table OID of the column.
pub fn table_oid(&self) -> Oid {
self.table_oid
}
/// Returns the column ID of the column.
pub fn column_id(&self) -> i16 {
self.column_id
}
/// Returns the format of the column.
pub fn format(&self) -> i16 {
self.format
}
/// Returns the type OID of the column.
pub fn type_oid(&self) -> Oid {
self.type_oid
}
/// Returns the type size of the column.
pub fn type_size(&self) -> i16 {
self.type_size
}
/// Returns the type modifier of the column.
pub fn type_modifier(&self) -> i32 {
self.type_modifier
}
}
impl fmt::Debug for Column {
fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result {
fmt.debug_struct("Column")
.field("name", &self.name)
.field("type", &self.type_)
.finish()
}
}

View File

@@ -1,162 +0,0 @@
//! TLS support.
use std::error::Error;
use std::future::Future;
use std::pin::Pin;
use std::task::{Context, Poll};
use std::{fmt, io};
use tokio::io::{AsyncRead, AsyncWrite, ReadBuf};
pub(crate) mod private {
pub struct ForcePrivateApi;
}
/// Channel binding information returned from a TLS handshake.
pub struct ChannelBinding {
pub(crate) tls_server_end_point: Option<Vec<u8>>,
}
impl ChannelBinding {
/// Creates a `ChannelBinding` containing no information.
pub fn none() -> ChannelBinding {
ChannelBinding {
tls_server_end_point: None,
}
}
/// Creates a `ChannelBinding` containing `tls-server-end-point` channel binding information.
pub fn tls_server_end_point(tls_server_end_point: Vec<u8>) -> ChannelBinding {
ChannelBinding {
tls_server_end_point: Some(tls_server_end_point),
}
}
}
/// A constructor of `TlsConnect`ors.
///
/// Requires the `runtime` Cargo feature (enabled by default).
pub trait MakeTlsConnect<S> {
/// The stream type created by the `TlsConnect` implementation.
type Stream: TlsStream + Unpin;
/// The `TlsConnect` implementation created by this type.
type TlsConnect: TlsConnect<S, Stream = Self::Stream>;
/// The error type returned by the `TlsConnect` implementation.
type Error: Into<Box<dyn Error + Sync + Send>>;
/// Creates a new `TlsConnect`or.
///
/// The domain name is provided for certificate verification and SNI.
fn make_tls_connect(&mut self, domain: &str) -> Result<Self::TlsConnect, Self::Error>;
}
/// An asynchronous function wrapping a stream in a TLS session.
pub trait TlsConnect<S> {
/// The stream returned by the future.
type Stream: TlsStream + Unpin;
/// The error returned by the future.
type Error: Into<Box<dyn Error + Sync + Send>>;
/// The future returned by the connector.
type Future: Future<Output = Result<Self::Stream, Self::Error>>;
/// Returns a future performing a TLS handshake over the stream.
fn connect(self, stream: S) -> Self::Future;
#[doc(hidden)]
fn can_connect(&self, _: private::ForcePrivateApi) -> bool {
true
}
}
/// A TLS-wrapped connection to a PostgreSQL database.
pub trait TlsStream: AsyncRead + AsyncWrite {
/// Returns channel binding information for the session.
fn channel_binding(&self) -> ChannelBinding;
}
/// A `MakeTlsConnect` and `TlsConnect` implementation which simply returns an error.
///
/// This can be used when `sslmode` is `none` or `prefer`.
#[derive(Debug, Copy, Clone)]
pub struct NoTls;
impl<S> MakeTlsConnect<S> for NoTls {
type Stream = NoTlsStream;
type TlsConnect = NoTls;
type Error = NoTlsError;
fn make_tls_connect(&mut self, _: &str) -> Result<NoTls, NoTlsError> {
Ok(NoTls)
}
}
impl<S> TlsConnect<S> for NoTls {
type Stream = NoTlsStream;
type Error = NoTlsError;
type Future = NoTlsFuture;
fn connect(self, _: S) -> NoTlsFuture {
NoTlsFuture(())
}
fn can_connect(&self, _: private::ForcePrivateApi) -> bool {
false
}
}
/// The future returned by `NoTls`.
pub struct NoTlsFuture(());
impl Future for NoTlsFuture {
type Output = Result<NoTlsStream, NoTlsError>;
fn poll(self: Pin<&mut Self>, _: &mut Context<'_>) -> Poll<Self::Output> {
Poll::Ready(Err(NoTlsError(())))
}
}
/// The TLS "stream" type produced by the `NoTls` connector.
///
/// Since `NoTls` doesn't support TLS, this type is uninhabited.
pub enum NoTlsStream {}
impl AsyncRead for NoTlsStream {
fn poll_read(
self: Pin<&mut Self>,
_: &mut Context<'_>,
_: &mut ReadBuf<'_>,
) -> Poll<io::Result<()>> {
match *self {}
}
}
impl AsyncWrite for NoTlsStream {
fn poll_write(self: Pin<&mut Self>, _: &mut Context<'_>, _: &[u8]) -> Poll<io::Result<usize>> {
match *self {}
}
fn poll_flush(self: Pin<&mut Self>, _: &mut Context<'_>) -> Poll<io::Result<()>> {
match *self {}
}
fn poll_shutdown(self: Pin<&mut Self>, _: &mut Context<'_>) -> Poll<io::Result<()>> {
match *self {}
}
}
impl TlsStream for NoTlsStream {
fn channel_binding(&self) -> ChannelBinding {
match *self {}
}
}
/// The error returned by `NoTls`.
#[derive(Debug)]
pub struct NoTlsError(());
impl fmt::Display for NoTlsError {
fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result {
fmt.write_str("no TLS implementation configured")
}
}
impl Error for NoTlsError {}

View File

@@ -1,63 +0,0 @@
use crate::codec::FrontendMessage;
use crate::connection::RequestMessages;
use crate::{CancelToken, Client, Error, ReadyForQueryStatus};
use postgres_protocol2::message::frontend;
/// A representation of a PostgreSQL database transaction.
///
/// Transactions will implicitly roll back when dropped. Use the `commit` method to commit the changes made in the
/// transaction. Transactions can be nested, with inner transactions implemented via safepoints.
pub struct Transaction<'a> {
client: &'a mut Client,
done: bool,
}
impl Drop for Transaction<'_> {
fn drop(&mut self) {
if self.done {
return;
}
let buf = self.client.inner.with_buf(|buf| {
frontend::query("ROLLBACK", buf).unwrap();
buf.split().freeze()
});
let _ = self
.client
.inner
.send(RequestMessages::Single(FrontendMessage::Raw(buf)));
}
}
impl<'a> Transaction<'a> {
pub(crate) fn new(client: &'a mut Client) -> Transaction<'a> {
Transaction {
client,
done: false,
}
}
/// Consumes the transaction, committing all changes made within it.
pub async fn commit(mut self) -> Result<ReadyForQueryStatus, Error> {
self.done = true;
self.client.batch_execute("COMMIT").await
}
/// Rolls the transaction back, discarding all changes made within it.
///
/// This is equivalent to `Transaction`'s `Drop` implementation, but provides any error encountered to the caller.
pub async fn rollback(mut self) -> Result<ReadyForQueryStatus, Error> {
self.done = true;
self.client.batch_execute("ROLLBACK").await
}
/// Like `Client::cancel_token`.
pub fn cancel_token(&self) -> CancelToken {
self.client.cancel_token()
}
/// Returns a reference to the underlying `Client`.
pub fn client(&mut self) -> &mut Client {
self.client
}
}

View File

@@ -1,113 +0,0 @@
use crate::{Client, Error, Transaction};
/// The isolation level of a database transaction.
#[derive(Debug, Copy, Clone)]
#[non_exhaustive]
pub enum IsolationLevel {
/// Equivalent to `ReadCommitted`.
ReadUncommitted,
/// An individual statement in the transaction will see rows committed before it began.
ReadCommitted,
/// All statements in the transaction will see the same view of rows committed before the first query in the
/// transaction.
RepeatableRead,
/// The reads and writes in this transaction must be able to be committed as an atomic "unit" with respect to reads
/// and writes of all other concurrent serializable transactions without interleaving.
Serializable,
}
/// A builder for database transactions.
pub struct TransactionBuilder<'a> {
client: &'a mut Client,
isolation_level: Option<IsolationLevel>,
read_only: Option<bool>,
deferrable: Option<bool>,
}
impl<'a> TransactionBuilder<'a> {
pub(crate) fn new(client: &'a mut Client) -> TransactionBuilder<'a> {
TransactionBuilder {
client,
isolation_level: None,
read_only: None,
deferrable: None,
}
}
/// Sets the isolation level of the transaction.
pub fn isolation_level(mut self, isolation_level: IsolationLevel) -> Self {
self.isolation_level = Some(isolation_level);
self
}
/// Sets the access mode of the transaction.
pub fn read_only(mut self, read_only: bool) -> Self {
self.read_only = Some(read_only);
self
}
/// Sets the deferrability of the transaction.
///
/// If the transaction is also serializable and read only, creation of the transaction may block, but when it
/// completes the transaction is able to run with less overhead and a guarantee that it will not be aborted due to
/// serialization failure.
pub fn deferrable(mut self, deferrable: bool) -> Self {
self.deferrable = Some(deferrable);
self
}
/// Begins the transaction.
///
/// The transaction will roll back by default - use the `commit` method to commit it.
pub async fn start(self) -> Result<Transaction<'a>, Error> {
let mut query = "START TRANSACTION".to_string();
let mut first = true;
if let Some(level) = self.isolation_level {
first = false;
query.push_str(" ISOLATION LEVEL ");
let level = match level {
IsolationLevel::ReadUncommitted => "READ UNCOMMITTED",
IsolationLevel::ReadCommitted => "READ COMMITTED",
IsolationLevel::RepeatableRead => "REPEATABLE READ",
IsolationLevel::Serializable => "SERIALIZABLE",
};
query.push_str(level);
}
if let Some(read_only) = self.read_only {
if !first {
query.push(',');
}
first = false;
let s = if read_only {
" READ ONLY"
} else {
" READ WRITE"
};
query.push_str(s);
}
if let Some(deferrable) = self.deferrable {
if !first {
query.push(',');
}
let s = if deferrable {
" DEFERRABLE"
} else {
" NOT DEFERRABLE"
};
query.push_str(s);
}
self.client.batch_execute(&query).await?;
Ok(Transaction::new(self.client))
}
}

View File

@@ -1,6 +0,0 @@
//! Types.
//!
//! This module is a reexport of the `postgres_types` crate.
#[doc(inline)]
pub use postgres_types2::*;

View File

@@ -35,7 +35,6 @@ use utils::backoff;
use utils::backoff::exponential_backoff_duration_seconds;
use crate::metrics::{start_measuring_requests, AttemptOutcome, RequestKind};
use crate::DownloadKind;
use crate::{
config::AzureConfig, error::Cancelled, ConcurrencyLimiter, Download, DownloadError,
DownloadOpts, Listing, ListingMode, ListingObject, RemotePath, RemoteStorage, StorageMetadata,
@@ -50,17 +49,10 @@ pub struct AzureBlobStorage {
concurrency_limiter: ConcurrencyLimiter,
// Per-request timeout. Accessible for tests.
pub timeout: Duration,
// Alternative timeout used for metadata objects which are expected to be small
pub small_timeout: Duration,
}
impl AzureBlobStorage {
pub fn new(
azure_config: &AzureConfig,
timeout: Duration,
small_timeout: Duration,
) -> Result<Self> {
pub fn new(azure_config: &AzureConfig, timeout: Duration) -> Result<Self> {
debug!(
"Creating azure remote storage for azure container {}",
azure_config.container_name
@@ -102,7 +94,6 @@ impl AzureBlobStorage {
max_keys_per_list_response,
concurrency_limiter: ConcurrencyLimiter::new(azure_config.concurrency_limit.get()),
timeout,
small_timeout,
})
}
@@ -142,7 +133,6 @@ impl AzureBlobStorage {
async fn download_for_builder(
&self,
builder: GetBlobBuilder,
timeout: Duration,
cancel: &CancellationToken,
) -> Result<Download, DownloadError> {
let kind = RequestKind::Get;
@@ -166,7 +156,7 @@ impl AzureBlobStorage {
.map_err(to_download_error);
// apply per request timeout
let response = tokio_stream::StreamExt::timeout(response, timeout);
let response = tokio_stream::StreamExt::timeout(response, self.timeout);
// flatten
let response = response.map(|res| match res {
@@ -230,11 +220,6 @@ impl AzureBlobStorage {
let started_at = ScopeGuard::into_inner(started_at);
let outcome = match &download {
Ok(_) => AttemptOutcome::Ok,
// At this level in the stack 404 and 304 responses do not indicate an error.
// There's expected cases when a blob may not exist or hasn't been modified since
// the last get (e.g. probing for timeline indices and heatmap downloads).
// Callers should handle errors if they are unexpected.
Err(DownloadError::NotFound | DownloadError::Unmodified) => AttemptOutcome::Ok,
Err(_) => AttemptOutcome::Err,
};
crate::metrics::BUCKET_METRICS
@@ -425,7 +410,7 @@ impl RemoteStorage for AzureBlobStorage {
let blob_client = self.client.blob_client(self.relative_path_to_name(key));
let properties_future = blob_client.get_properties().into_future();
let properties_future = tokio::time::timeout(self.small_timeout, properties_future);
let properties_future = tokio::time::timeout(self.timeout, properties_future);
let res = tokio::select! {
res = properties_future => res,
@@ -531,12 +516,7 @@ impl RemoteStorage for AzureBlobStorage {
});
}
let timeout = match opts.kind {
DownloadKind::Small => self.small_timeout,
DownloadKind::Large => self.timeout,
};
self.download_for_builder(builder, timeout, cancel).await
self.download_for_builder(builder, cancel).await
}
async fn delete(&self, path: &RemotePath, cancel: &CancellationToken) -> anyhow::Result<()> {

View File

@@ -24,13 +24,6 @@ pub struct RemoteStorageConfig {
skip_serializing_if = "is_default_timeout"
)]
pub timeout: Duration,
/// Alternative timeout used for metadata objects which are expected to be small
#[serde(
with = "humantime_serde",
default = "default_small_timeout",
skip_serializing_if = "is_default_small_timeout"
)]
pub small_timeout: Duration,
}
impl RemoteStorageKind {
@@ -47,18 +40,10 @@ fn default_timeout() -> Duration {
RemoteStorageConfig::DEFAULT_TIMEOUT
}
fn default_small_timeout() -> Duration {
RemoteStorageConfig::DEFAULT_SMALL_TIMEOUT
}
fn is_default_timeout(d: &Duration) -> bool {
*d == RemoteStorageConfig::DEFAULT_TIMEOUT
}
fn is_default_small_timeout(d: &Duration) -> bool {
*d == RemoteStorageConfig::DEFAULT_SMALL_TIMEOUT
}
/// A kind of a remote storage to connect to, with its connection configuration.
#[derive(Debug, Clone, PartialEq, Eq, Deserialize, Serialize)]
#[serde(untagged)]
@@ -199,7 +184,6 @@ fn serialize_storage_class<S: serde::Serializer>(
impl RemoteStorageConfig {
pub const DEFAULT_TIMEOUT: Duration = std::time::Duration::from_secs(120);
pub const DEFAULT_SMALL_TIMEOUT: Duration = std::time::Duration::from_secs(30);
pub fn from_toml(toml: &toml_edit::Item) -> anyhow::Result<RemoteStorageConfig> {
Ok(utils::toml_edit_ext::deserialize_item(toml)?)
@@ -235,8 +219,7 @@ timeout = '5s'";
storage: RemoteStorageKind::LocalFs {
local_path: Utf8PathBuf::from(".")
},
timeout: Duration::from_secs(5),
small_timeout: RemoteStorageConfig::DEFAULT_SMALL_TIMEOUT
timeout: Duration::from_secs(5)
}
);
}
@@ -264,8 +247,7 @@ timeout = '5s'";
max_keys_per_list_response: DEFAULT_MAX_KEYS_PER_LIST_RESPONSE,
upload_storage_class: Some(StorageClass::IntelligentTiering),
}),
timeout: Duration::from_secs(7),
small_timeout: RemoteStorageConfig::DEFAULT_SMALL_TIMEOUT
timeout: Duration::from_secs(7)
}
);
}
@@ -317,8 +299,7 @@ timeout = '5s'";
concurrency_limit: default_remote_storage_azure_concurrency_limit(),
max_keys_per_list_response: DEFAULT_MAX_KEYS_PER_LIST_RESPONSE,
}),
timeout: Duration::from_secs(7),
small_timeout: RemoteStorageConfig::DEFAULT_SMALL_TIMEOUT
timeout: Duration::from_secs(7)
}
);
}

View File

@@ -178,15 +178,6 @@ pub struct DownloadOpts {
/// The end of the byte range to download, or unbounded. Must be after the
/// start bound.
pub byte_end: Bound<u64>,
/// Indicate whether we're downloading something small or large: this indirectly controls
/// timeouts: for something like an index/manifest/heatmap, we should time out faster than
/// for layer files
pub kind: DownloadKind,
}
pub enum DownloadKind {
Large,
Small,
}
impl Default for DownloadOpts {
@@ -195,7 +186,6 @@ impl Default for DownloadOpts {
etag: Default::default(),
byte_start: Bound::Unbounded,
byte_end: Bound::Unbounded,
kind: DownloadKind::Large,
}
}
}
@@ -594,10 +584,6 @@ impl<Other: RemoteStorage> GenericRemoteStorage<Arc<Other>> {
impl GenericRemoteStorage {
pub async fn from_config(storage_config: &RemoteStorageConfig) -> anyhow::Result<Self> {
let timeout = storage_config.timeout;
// If somkeone overrides timeout to be small without adjusting small_timeout, then adjust it automatically
let small_timeout = std::cmp::min(storage_config.small_timeout, timeout);
Ok(match &storage_config.storage {
RemoteStorageKind::LocalFs { local_path: path } => {
info!("Using fs root '{path}' as a remote storage");
@@ -620,11 +606,7 @@ impl GenericRemoteStorage {
.unwrap_or("<AZURE_STORAGE_ACCOUNT>");
info!("Using azure container '{}' in account '{storage_account}' in region '{}' as a remote storage, prefix in container: '{:?}'",
azure_config.container_name, azure_config.container_region, azure_config.prefix_in_container);
Self::AzureBlob(Arc::new(AzureBlobStorage::new(
azure_config,
timeout,
small_timeout,
)?))
Self::AzureBlob(Arc::new(AzureBlobStorage::new(azure_config, timeout)?))
}
})
}

View File

@@ -219,8 +219,7 @@ async fn create_azure_client(
concurrency_limit: NonZeroUsize::new(100).unwrap(),
max_keys_per_list_response,
}),
timeout: RemoteStorageConfig::DEFAULT_TIMEOUT,
small_timeout: RemoteStorageConfig::DEFAULT_SMALL_TIMEOUT,
timeout: Duration::from_secs(120),
};
Ok(Arc::new(
GenericRemoteStorage::from_config(&remote_storage_config)

View File

@@ -396,7 +396,6 @@ async fn create_s3_client(
upload_storage_class: None,
}),
timeout: RemoteStorageConfig::DEFAULT_TIMEOUT,
small_timeout: RemoteStorageConfig::DEFAULT_SMALL_TIMEOUT,
};
Ok(Arc::new(
GenericRemoteStorage::from_config(&remote_storage_config)

View File

@@ -19,14 +19,12 @@ bincode.workspace = true
bytes.workspace = true
camino.workspace = true
chrono.workspace = true
diatomic-waker.workspace = true
git-version.workspace = true
hex = { workspace = true, features = ["serde"] }
humantime.workspace = true
hyper0 = { workspace = true, features = ["full"] }
fail.workspace = true
futures = { workspace = true}
jemalloc_pprof.workspace = true
jsonwebtoken.workspace = true
nix.workspace = true
once_cell.workspace = true
@@ -47,7 +45,6 @@ tracing.workspace = true
tracing-error.workspace = true
tracing-subscriber = { workspace = true, features = ["json", "registry"] }
rand.workspace = true
scopeguard.workspace = true
strum.workspace = true
strum_macros.workspace = true
url.workspace = true

View File

@@ -10,7 +10,6 @@ use metrics::{register_int_counter, Encoder, IntCounter, TextEncoder};
use once_cell::sync::Lazy;
use routerify::ext::RequestExt;
use routerify::{Middleware, RequestInfo, Router, RouterBuilder};
use tokio_util::io::ReaderStream;
use tracing::{debug, info, info_span, warn, Instrument};
use std::future::Future;
@@ -408,69 +407,6 @@ pub async fn profile_cpu_handler(req: Request<Body>) -> Result<Response<Body>, A
}
}
/// Generates heap profiles.
///
/// This only works with jemalloc on Linux.
pub async fn profile_heap_handler(req: Request<Body>) -> Result<Response<Body>, ApiError> {
enum Format {
Jemalloc,
Pprof,
}
// Parameters.
let format = match get_query_param(&req, "format")?.as_deref() {
None => Format::Pprof,
Some("jemalloc") => Format::Jemalloc,
Some("pprof") => Format::Pprof,
Some(format) => return Err(ApiError::BadRequest(anyhow!("invalid format {format}"))),
};
// Obtain profiler handle.
let mut prof_ctl = jemalloc_pprof::PROF_CTL
.as_ref()
.ok_or(ApiError::InternalServerError(anyhow!(
"heap profiling not enabled"
)))?
.lock()
.await;
if !prof_ctl.activated() {
return Err(ApiError::InternalServerError(anyhow!(
"heap profiling not enabled"
)));
}
// Take and return the profile.
match format {
Format::Jemalloc => {
// NB: file is an open handle to a tempfile that's already deleted.
let file = tokio::task::spawn_blocking(move || prof_ctl.dump())
.await
.map_err(|join_err| ApiError::InternalServerError(join_err.into()))?
.map_err(ApiError::InternalServerError)?;
let stream = ReaderStream::new(tokio::fs::File::from_std(file));
Response::builder()
.status(200)
.header(CONTENT_TYPE, "application/octet-stream")
.header(CONTENT_DISPOSITION, "attachment; filename=\"heap.dump\"")
.body(Body::wrap_stream(stream))
.map_err(|err| ApiError::InternalServerError(err.into()))
}
Format::Pprof => {
let data = tokio::task::spawn_blocking(move || prof_ctl.dump_pprof())
.await
.map_err(|join_err| ApiError::InternalServerError(join_err.into()))?
.map_err(ApiError::InternalServerError)?;
Response::builder()
.status(200)
.header(CONTENT_TYPE, "application/octet-stream")
.header(CONTENT_DISPOSITION, "attachment; filename=\"heap.pb\"")
.body(Body::from(data))
.map_err(|err| ApiError::InternalServerError(err.into()))
}
}
}
pub fn add_request_id_middleware<B: hyper::body::HttpBody + Send + Sync + 'static>(
) -> Middleware<B, ApiError> {
Middleware::pre(move |req| async move {

View File

@@ -164,12 +164,6 @@ impl TenantShardId {
}
}
impl std::fmt::Display for ShardNumber {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
self.0.fmt(f)
}
}
impl std::fmt::Display for ShardSlug<'_> {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
write!(

Some files were not shown because too many files have changed in this diff Show More