scripts/detection.sh runs four failures one after another and brings the stand back between them: SIGTERM through docker stop -t 6 (the balancer must exclude the replica within 1.5 s and before the process exits after its drain window), docker kill (1.5 s), docker pause for a hung process (3.5 s, the check timeout counts twice) and stopping PostgreSQL (all replicas out within 1.5 s, the balancer answers 503 itself). A 0.25 s slack covers the docker CLI and the 50 ms polling. Each run writes evidence/detection/<case>.json and the script fails if any case exceeds its declared maximum. scripts/failover-each.sh stops every replica in turn with stop-one.sh, keeps it down for 10 s and restores it, all inside one k6 load window, and collects a summary with RPS, errors, latency, detection, rejoin, probe recovery and balancer retry counters per replica. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
21 lines
872 B
Bash
Executable File
21 lines
872 B
Bash
Executable File
#!/bin/sh
|
|
set -eu
|
|
cd "$(dirname "$0")/.."
|
|
out=${1:-evidence/detection}
|
|
mkdir -p "$out"
|
|
scripts/restore.sh
|
|
replica() { docker compose ps -q "${API_SERVICE:-api}" | head -1; }
|
|
detect() { python3 scripts/probe.py detect "$@"; }
|
|
failed=0
|
|
detect --max 1.5 --within-action --restore scripts/restore.sh --out "$out/sigterm.json" \
|
|
-- docker stop -t "${STOP_TIMEOUT:-6}" "$(replica)" || failed=1
|
|
detect --max 1.5 --restore scripts/restore.sh --out "$out/kill.json" \
|
|
-- docker kill "$(replica)" || failed=1
|
|
hung=$(replica)
|
|
detect --max 3.5 --restore "docker unpause $hung && scripts/lb-wait.sh" --out "$out/pause.json" \
|
|
-- docker pause "$hung" || failed=1
|
|
db=$(docker compose ps -q postgres)
|
|
detect --max 1.5 --until 0 --lb-status 503 --restore "docker start $db && scripts/lb-wait.sh" \
|
|
--out "$out/postgres.json" -- docker stop "$db" || failed=1
|
|
exit "$failed"
|