Commit Graph
109 Commits
Author SHA1 Message Date
ricoromaandClaude Opus 5.5 edad1fda2b fix: exit at once on SIGTERM before the server starts
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-13 16:53:15 +03:00
ricoromaandClaude Opus 5.5 f6052f6c30 fix: stop starting replicas with the compose grace period in the driver
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-13 16:51:25 +03:00
ricoromaandClaude Opus 5.5 42e6edf420 feat: add a driver command that stops a starting replica
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-13 16:50:18 +03:00
ricoromaandClaude Opus 5.5 f8c81878cb fix: run the fault stand postgres under an init process
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-13 16:45:25 +03:00
ricoromaandClaude Opus 5.5 249b67e8c0 test: fail fault checks when the stand postgres crashes
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-13 16:43:54 +03:00
ricoromaandClaude Opus 5.5 b37c45cbb8 revert: drop the failover events cross-check
This reverts commit a24832c.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-13 16:39:44 +03:00
ricoromaandClaude Opus 5.5 a24832c72b test: cross-check failover events against the daemon buffer
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-13 16:16:31 +03:00
ricoromaandClaude Opus 5.5 fe75fcf8a9 fix: stream docker events during failover phases
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-13 16:16:11 +03:00
ricoromaandClaude Opus 5.5 18238588b4 fix: record a zero swap-out delta in bench passports
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-13 16:15:25 +03:00
ricoromaandClaude Opus 5.5 628bbfa639 docs: propose fix-known-issues change
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-13 16:13:04 +03:00
ricoromaandClaude Opus 5.5 f74a3f384e docs: quantify the docker events buffer margin in known issues
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-13 15:28:34 +03:00
ricoromaandClaude Opus 5.5 669a0d9df7 docs: repeat the clean clone check on the final code
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-13 15:19:51 +03:00
ricoromaandClaude Opus 5.5 6476b1e673 docs: track label logs referenced by the report
The *.log rule kept LB, rollout script and 5xx logs of committed
labels out of git while REPORT.md cites them. The 59 logs (556 KB)
hold no tokens, passwords or host paths; quick-* labels stay ignored.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-13 15:18:00 +03:00
ricoromaandClaude Opus 5.5 dbb91507ed docs: note local-only label logs in the report
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-12 19:29:17 +03:00
ricoromaandClaude Opus 5.5 3416ca16b9 docs: archive profile-resources and sync its specs
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-12 19:28:03 +03:00
ricoromaandClaude Opus 5.5 2514ff177d docs: address profile-resources verify findings
Report the failover ranges strictly over phases 1-5 with phase 06
apart, add per-window memory peaks to the failover table, describe
which intervals the window maximum covers, and note the per-second
rows and per-request zeros of the taken labels fixed in f34cb9a.
Record the design findings, the raw logs ignored by git as a known
issue, and the checks after the fixes.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-12 19:26:18 +03:00
ricoromaandClaude Opus 5.5 f34cb9af5c fix: label per-second rows around pause by their unfrozen part
The samples.csv row of an interval that touches the start of a freeze
now gets the window of its unfrozen part, as the window aggregates do
since the thaw correction. CPU per successful request is no longer
computed for a container absent from the window. The definitions in
the summary describe which intervals the window maximum covers.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-12 19:23:29 +03:00
ricoromaandClaude Opus 5.5 c5752962f6 docs: report CPU and memory of failover and rollout profiles
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-12 19:04:33 +03:00
ricoromaandClaude Opus 5.5 9c4249d8d1 docs: add rollout resources label
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-12 18:57:42 +03:00
ricoromaandClaude Opus 5.5 309a00da2a docs: add failover resources label
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-12 18:35:27 +03:00
ricoromaandClaude Opus 5.5 2d267bfbe6 fix: attribute replica CPU around pause to the unfrozen part
A sample interval that crosses docker pause or unpause carried the CPU
spent before the pause and the burst after unpause into the failover
window by proportional split. Use the pause and unpause moments from
Docker events so that interval CPU lands outside the frozen span.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-12 18:09:05 +03:00
ricoromaandClaude Opus 5.5 6505b5e929 feat: sample CPU and memory in failover and rollout profiles
Run bench/sampler.sh in a new scan mode (cgroup walk, no network, no
psql or LB polling) for the whole load of the failover and rollout
profiles. bench/resources.py turns the samples into per-window CPU,
throttling and memory of every component and the VM, with
samples.csv, resources.csv and summary sections; CPU per successful
request only in steady windows. Check the peak CPU budget before the
stand starts.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-12 17:42:04 +03:00
ricoromaandClaude Opus 5.5 417fdd2f22 docs: propose profile-resources change
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-12 17:22:14 +03:00
ricoromaandClaude Opus 5.5 8f007f7bab docs: add network and ports, p50 latency and README compose commands to the report
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-12 16:49:08 +03:00
ricoromaandClaude Opus 5.5 7f00128fbe docs: address final report review and remove PLAN.md
Fix the findings of the final REPORT.md review: prerequisites, the
defence scenario, swap sources, the single explanation of the R* drop
at PostgreSQL 0.5 CPU, known defects in the limitations, a dead link,
stale counts and references. Remove the device name from evidence.

Remove the lab plan now that every change is archived: its decisions
live in REPORT.md and the archived designs, live references point
there, and AGENTS.md drops the transitional section.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 22:20:22 +03:00
ricoromaandClaude Opus 5.5 b0637addea fix: run the make instances driver on system python3 3.9
datetime.UTC exists only since Python 3.11, so make instances failed on
the macOS system python3 listed in the report prerequisites. Use
timezone.utc and postponed annotations instead.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 22:20:22 +03:00
ricoromaandClaude Opus 5.5 063eedbc71 docs: complete lab 3 report and add final clean clone check
Fill the remaining report sections: clean clone run with versions and
prerequisites, lab 2 carry-over, transaction boundaries, teacher
agreements, skeleton variable mapping, rubric cross-check, experiment
overview, limitations summary, sources and AI usage. Record the clean
clone run of every documented command on default ports.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 21:53:13 +03:00
ricoromaandClaude Opus 5.5 67b4303a0a docs: archive scaling-improvement and sync its specs
Sync load-benchmark (series comparison table, track 1 before/after
rules, db4 series) and service-configuration (default pool 10 + 0)
into the main specs, move the change to the archive and drop it from
the transitional state in AGENTS.md.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 21:38:16 +03:00
ricoromaandClaude Opus 5.5 25f1962021 docs: address scaling-improvement verify findings
Fix the two-step threshold (x1.118), CPU-per-request spread and
PostgreSQL ranges in REPORT.md and design, record the LB side check
(lb_chkdown, error_codes) in evidence, add the db4 series to the
load-benchmark delta, refresh AGENTS.md, bench.md, PLAN.md and
proposal, and log the swapouts_delta passport defect.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 21:36:21 +03:00
ricoromaandClaude Opus 5.5 cb984c1d8d docs: count the pool improvement for track 1 with a recorded caveat
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 21:11:19 +03:00
ricoromaandClaude Opus 5.5 dee571c9cd docs: compare scaling-improvement result against the same-session A/A
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 21:11:19 +03:00
ricoromaandClaude Opus 5.5 adbb06d8f9 docs: report second scaling-improvement attempt for track 1
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 17:51:55 +03:00
ricoromaandClaude Opus 5.5 4648cd0a8c docs: add second scaling-improvement attempt comparisons and checks
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 17:51:55 +03:00
ricoromaandClaude Opus 5.5 17998c92ce docs: add scaling-db-nooverflow2 label
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 17:48:32 +03:00
ricoroma 7a38c5dbcc Revert "revert: measure scaling-improvement baseline again without the pool change"
This reverts commit 5c07923d8c.
2026-10-11 17:23:34 +03:00
ricoroma de6001c526 docs: add scaling-db-before2 bench evidence 2026-10-11 17:23:34 +03:00
ricoromaandClaude Opus 5.5 45a0ad723b docs: add second scaling db A/A label for scaling-improvement
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 16:58:20 +03:00
ricoromaandClaude Opus 5.5 5c07923d8c revert: measure scaling-improvement baseline again without the pool change
This reverts commit 53eb38e (the pool change 9c3959d, 623a973, 80e64cd).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 16:30:27 +03:00
ricoromaandClaude Opus 5.5 a676a520b0 docs: plan the second scaling-improvement measurement attempt
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 16:30:27 +03:00
ricoromaandClaude Opus 5.5 056ca19a53 docs: format probe script in connection cost evidence
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 15:46:34 +03:00
ricoromaandClaude Opus 5.5 8b69913995 docs: report scaling-improvement pool result for track 1
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 15:46:24 +03:00
ricoromaandClaude Opus 5.5 1694b1d0ae docs: add scaling-improvement comparisons, connection cost and checks
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 15:46:13 +03:00
ricoromaandClaude Opus 5.5 cd88b48b41 docs: add scaling-db-nooverflow label
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 15:30:47 +03:00
ricoroma 53eb38e19b Revert "revert: measure scaling-improvement baseline without the pool change"
This reverts commit 62fa57535e.
2026-10-11 15:04:46 +03:00
ricoroma dfd9ecce29 docs: add scaling-db-before bench evidence 2026-10-11 15:04:46 +03:00
ricoromaandClaude Opus 5.5 3df111cd80 docs: add scaling db A/A label for scaling-improvement
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 14:37:29 +03:00
ricoromaandClaude Opus 5.5 62fa57535e revert: measure scaling-improvement baseline without the pool change
This reverts commits 9c3959d, 623a973 and 80e64cd.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 14:10:43 +03:00
ricoromaandClaude Opus 5.5 1bb8ea67ad feat: add two-cell db4 series for the scaling-improvement measurements
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 14:06:57 +03:00
ricoromaandClaude Opus 5.5 80e64cda4a test: count PostgreSQL sessions without healthcheck and flush races
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 13:46:39 +03:00
ricoromaandClaude Opus 5.5 623a973dac test: wait for PostgreSQL session stats before counting pool reconnections
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 13:30:38 +03:00