[murex971]

Summary

LISTEN and NOTIFY with pub/sub disabled (pub_sub_channel_size = 0, the default) returned FATAL: pub/sub disabled and closed the client connection. Fixes #1808.

The error is now reported with SQLSTATE 0A000, the connection stays open, and an open transaction is marked failed. The hint tells the client to set pub_sub_channel_size.

NOTIFY inside a transaction is rejected immediately. Those notifications are buffered and sent after COMMIT, so failing then would close the connection after the client was told the commit succeeded.

[codecov]

Codecov Report

❌ Patch coverage is 98.87640% with 1 line in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
pgdog/src/frontend/client/query_engine/pub_sub.rs	96.42%	1 Missing ⚠️

📢 Thoughts on this report? Let us know!

[github-actions]

Bencher Report

Project	PgDog
Branch	fix/pub-sub-disabled
Testbed	Intel v1

⚠️ WARNING: Truncated view!

The full continuous benchmarking report exceeds the maximum length allowed on this platform.

🐰 View full continuous benchmarking report in Bencher

[github-actions]

Bencher Report

Project	PgDog
Branch	fix/pub-sub-disabled
Testbed	Intel v1

Click to view all benchmark results

Benchmark	Throughput	Benchmark Result operations / second (ops/s) x 1e3 (Result Δ%)	Lower Boundary operations / second (ops/s) x 1e3 (Limit %)	cpu-per-txn	Benchmark Result Measure (units) (Result Δ%)	Upper Boundary Measure (units) (Limit %)
pooler-v3	📈 view plot 🚷 view threshold	16.99 ops/s x 1e3 (+0.00%) Baseline: 16.99 ops/s x 1e3	16.82 ops/s x 1e3 (99.00%)	📈 view plot 🚷 view threshold	40.43 units (-4.25%) Baseline: 42.23 units	42.86 units (94.33%)

🐰 View full continuous benchmarking report in Bencher

[github-actions]

Bencher Report

Project	PgDog
Branch	fix/pub-sub-disabled
Testbed	Intel v1

Click to view all benchmark results

Benchmark	Throughput	Benchmark Result operations / second (ops/s) x 1e3 (Result Δ%)	Lower Boundary operations / second (ops/s) x 1e3 (Limit %)	cpu-per-txn	Benchmark Result Measure (units) (Result Δ%)	Upper Boundary Measure (units) (Limit %)
lb-v3	📈 view plot 🚷 view threshold	15.99 ops/s x 1e3 (+0.00%) Baseline: 15.99 ops/s x 1e3	15.83 ops/s x 1e3 (99.00%)	📈 view plot 🚷 view threshold	41.43 units (-4.60%) Baseline: 43.43 units	44.08 units (93.99%)

🐰 View full continuous benchmarking report in Bencher

[github-actions]

Metal bench passed. All 150 measures within bounds.

suite	throughput	cpu/txn	user cycles/txn	user instr/txn	kernel cycles/txn	ctx/txn
pooler		28.73 (+0.6% ± 1.3%)	29,940 (+0.4% ± 1.6%)	35,018 (-0.0% ± 0.03%)	38,862 (+1.0% ± 1.9%)
lb		29.66 (+1.3% ± 0.8%)	31,234 (+0.3% ± 1.1%)	35,894 (-0.1% ± 0.03%)	39,783 (+2.5% ± 1.5%)
sharding-w2		52.94 (-0.5%)	68,126 (-3.1%)	72,262 (-0.0%)	61,389 (+1.9%)	1.404 (+0.5%)
sharding-w8	154,222 (-2.0%)	47.6 (+1.9%)	68,566 (+1.4%)	56,095 (+0.0%)	54,126 (+2.4%)	0.1444 (+0.9%)

Counters per txn

suite	frontend stall	l1i refill	backend stall
pooler	16,492 (+1.0% ± 3.2%)	3,451 (-0.4% ± 1.3%)	2,202 (-1.6% ± 2.7%)
lb	17,249 (+1.1% ± 2.2%)	3,508 (-0.6% ± 1.2%)	2,417 (-3.5% ± 3.0%)
sharding-w2	39,786 (-6.1%)	5,486 (-0.6%)	6,372 (+5.5%)
sharding-w8	39,973 (+2.8%)	5,168 (+1.5%)	8,709 (-2.4%)

Microbench: all 135 measures within bounds.

• extended_cached: 15 of 15 benches faster, median -3.8% ± 0.5%
• extended_cold: 14 of 15 benches faster, median -3.3% ± 0.6%
• simple_cold: 14 of 15 benches faster, median -1.7% ± 0.3%

lb flamegraph vs b66fc9e

Open interactive for hover, zoom and search.

Open interactive.

microbench flamegraph vs b66fc9e

Open interactive for hover, zoom and search.

Open interactive.

pooler flamegraph vs b66fc9e

Open interactive for hover, zoom and search.

Open interactive.

sharding-w2 flamegraph vs b66fc9e

Open interactive for hover, zoom and search.

Open interactive.

sharding-w8 flamegraph vs b66fc9e

Open interactive for hover, zoom and search.

Open interactive.

Logs

lb, microbench, pooler, sharding-w2, sharding-w8, verdict.json

Compared 335259b with b66fc9e.