v0.34.0_8
This commit is contained in:
Mateusz Gruszczyński committed
1 parent
b6a3e12225
commit
480c3ec769
92 files changed
+298
-10035
No files matched your search
@@ -1,53 +1,5 @@
|
||||
# RustProxy 0.34.0
|
||||
|
||||
## DEV: OPT5 multi-hot-path tuning (source candidate; uncompiled)
|
||||
|
||||
OPT5 builds on the **compiled OPT4** (151/151 executed live HTTP checks passed;
|
||||
1 optional Zstandard check skipped). The shared-host 1,040-byte HTTP/1.1
|
||||
comparison shows OPT4 at about 39,054 RPS with 16 connections and 36,425 RPS
|
||||
with 64; mixed results versus OPT3 mean there is no proven broad RPS gain.
|
||||
OPT5 removes unnecessary ID hashing and unauthenticated original-URL
|
||||
construction, avoids per-success health-detail allocation, batches bounded
|
||||
traffic analytics, avoids routine status/User-Agent allocations, and fast-paths
|
||||
unoffered response compression. All policies, observability, caches and
|
||||
protocol support remain enabled; no version bump. **OPT5 has not been built
|
||||
or measured.** See [`docs/PROXY_1KB_OPT5_DEV_0.34.0.md`](docs/PROXY_1KB_OPT5_DEV_0.34.0.md).
|
||||
|
||||
## DEV: OPT4 small-response proxy hot path (subsequently compiled)
|
||||
|
||||
The compiled OPT4 binary passed 151 real HTTP acceptance checks (1 optional
|
||||
Zstandard case skipped). OPT4 removes avoidable allocations and string work
|
||||
for ordinary uncached HTTP/1.1 requests, with regressions for empty request
|
||||
bodies, URI/Host/Connection semantics and Unicode log truncation. Its brief
|
||||
shared-host RPS comparison is mixed, so no stable improvement is claimed. No settings or functionality disabled;
|
||||
version remains 0.34.0. Details:
|
||||
[`docs/PROXY_1KB_OPT4_DEV_0.34.0.md`](docs/PROXY_1KB_OPT4_DEV_0.34.0.md).
|
||||
|
||||
|
||||
## DEV: small-response proxy hot path, OPT3 source candidate
|
||||
|
||||
Seven compatible optimizations reduce path copies and work constructing
|
||||
forwarded headers, log IPs, hop-by-hop headers, request IDs and cache-off
|
||||
reason totals. HTTP/1.1 proxy with a 1,040-byte response remains the first
|
||||
benchmark priority; local Web Server is secondary. No version or policy changes.
|
||||
A subsequently supplied compiled OPT3 executable passed 151 real HTTP
|
||||
assertions (one optional codec skipped). Two short same-host benchmark runs
|
||||
did **not** establish a stable RPS gain over OPT2. See the historical
|
||||
[`OPT3 source report`](docs/PROXY_1KB_OPT3_DEV_0.34.0.md)
|
||||
and the newer [`OPT4 analysis`](docs/PROXY_1KB_OPT4_DEV_0.34.0.md).
|
||||
|
||||
## DEV: cache counter contention and saturated analytics (source candidate)
|
||||
|
||||
Per-domain cache outcome/reason accounting now uses bounded immutable indexes
|
||||
and atomic counters rather than the HOT cache's global store mutex on every
|
||||
request. The analytics producer reserves a queue slot before cloning fields;
|
||||
full-queue/drop behavior and its counter remain unchanged. Per-vhost
|
||||
latency histograms now update one exclusive atomic bin per completion and
|
||||
reconstruct unchanged cumulative Prometheus buckets during scraping. The supplied
|
||||
previous DEV binary passed 151 live HTTP assertions, but this newest code
|
||||
**has not been built**. No version bump, no measured RPS claim. See
|
||||
[`docs/PROXY_1KB_COUNTERS_DEV_0.34.0.md`](docs/PROXY_1KB_COUNTERS_DEV_0.34.0.md).
|
||||
|
||||
## 0.34.0: HOT HIT critical-section follow-up (source candidate)
|
||||
|
||||
Immutable response headers are shared until a HIT; the mutable per-response
|
||||
@@ -57,12 +9,6 @@ shorter paths. The existing stale-for-revalidation policy makes the previous
|
||||
full HOT sweep unnecessary. Exact LRU, generation barriers, BAN, COLD,
|
||||
transformation/ETag, TTL and cache bypass settings remain unchanged.
|
||||
|
||||
This is **uncompiled source**; do not infer network RPS or production acceptance
|
||||
from source checks. See
|
||||
[`docs/CACHE_HOTPATH_0.34.0.md`](docs/CACHE_HOTPATH_0.34.0.md).
|
||||
|
||||
## 0.34.0: reverse-proxy hot-path optimization revision (source candidate)
|
||||
|
||||
This revision retains the **0.34.0** version and all prior scoped-cache,
|
||||
`sendfile`, security, logging and vhost/administration controls. In the
|
||||
uncached reverse-proxy hot path, request observation now passes the immutable
|
||||
@@ -87,9 +33,7 @@ subscribers only while subscribers exist; history and external sinks remain.
|
||||
|
||||
These changes have **not** been compiled or timed in this environment, and
|
||||
there is no claimed 0.34.0 RPS improvement. Build and run the regression and
|
||||
nginx benchmark matrix before deployment. See
|
||||
[`docs/PROXY_HOTPATH_0.34.0.md`](docs/PROXY_HOTPATH_0.34.0.md) for invariants,
|
||||
checks and exact measurement commands.
|
||||
nginx benchmark matrix before deployment.
|
||||
|
||||
## 0.34.0: scoped cache, proxy accounting and opt-in Linux sendfile
|
||||
|
||||
@@ -116,12 +60,6 @@ existing streaming path. The default is `off`; toggling this transport option
|
||||
does not purge HOT/COLD cache. **Native compilation and live protocol tests
|
||||
remain pending; do not deploy this candidate without them.**
|
||||
|
||||
An offline cache-retention simulation for a 20-vhost workload shows 92.123%
|
||||
fewer invalidated objects and 92.113% fewer cache misses than the old global
|
||||
purge policy **under its stated synthetic assumptions**. These are not RPS
|
||||
figures; real 0.34.0 performance remains unmeasured.
|
||||
|
||||
See `docs/TEST_REPORT_0.34.0.md` for verification scope and remaining gates.
|
||||
|
||||
## 0.33.8: request-path performance and compatibility verification
|
||||
|
||||
@@ -150,12 +88,9 @@ to exercise the existing documented behavior, including scheme-isolated cache
|
||||
keys, purge generations, private administration responses and bundled error pages.
|
||||
|
||||
The LB panel, database schema, default TLS/caching policies and bounded pool
|
||||
recycling retain their existing behavior. Live process regression, verified TLS,
|
||||
H2/H3/QUIC and WebSocket runners, repeated nginx comparisons, resource sampling,
|
||||
syscall profiling and a mixed-traffic soak are included in
|
||||
[`scripts/performance/`](scripts/performance/README.md). See the
|
||||
[verification contract](docs/TEST_REPORT.md) and the separately generated
|
||||
[performance results](docs/PERFORMANCE_0.33.8.md) for measured outcomes and limits.
|
||||
recycling retain their existing behavior. Historical verification methods and
|
||||
results are documented in [docs/TEST_REPORT.md](docs/TEST_REPORT.md) and
|
||||
[docs/PERFORMANCE_0.33.8.md](docs/PERFORMANCE_0.33.8.md).
|
||||
|
||||
## 0.33.7: named dashboard presets, panel components and chart typography
|
||||
|
||||
@@ -780,7 +715,6 @@ Historical release information remains in [CHANGELOG.md](CHANGELOG.md) and [docs
|
||||
- [docs/FILE_LOGS.md](docs/FILE_LOGS.md) - stored-log explorer, protocol-error records, bounds and filesystem security.
|
||||
- [docs/PHP_FPM.md](docs/PHP_FPM.md) - implemented FastCGI/PHP-FPM integration, safety model and runtime requirements.
|
||||
- [docs/TEST_REPORT.md](docs/TEST_REPORT.md) - verification results and environment limitations.
|
||||
- [docs/TODO.md](docs/TODO.md) - unresolved follow-up only.
|
||||
|
||||
## Release history
|
||||
|
||||
|
||||
Reference in new issue
Block a user