Files
fleetd/scripts/test-redeploy-fleetd.sh
Dai Ha 599419f9e6
CI / contract (pull_request) Successful in 47s
CI / build (pull_request) Successful in 1m53s
fleetd #492 follow-up: carry the unclear detail across detect_supervisor's subshell boundary
b17f37a set SUPERVISOR_UNCLEAR_DETAIL as a global inside detect_supervisor, but the real call
site invokes it as $(detect_supervisor) — a subshell — so that global died with the subshell and
the die() message's ${VAR:-fallback} silently masked the loss with generic text.

- detect_supervisor now packs kind and detail onto its one stdout line (joined by the ASCII unit
  separator byte, $SUPERVISOR_DETAIL_SEP), the only channel that survives $( ). The real call site
  unpacks both with in-shell parameter expansion — no extra subshell.
- Dropped the ${SUPERVISOR_UNCLEAR_DETAIL:-...} fallback at the die() message: under set -u, a
  missing detail now fails loudly instead of silently defaulting (same defect class as #497).
- Added a constraints comment block above detect_supervisor for future callers: stdout-only,
  no ${VAR:-default} papering over a lost value, and every case on the return value needs an
  explicit *) arm.
- Added *) arms to the three `case "$SUPERVISOR_KIND"` switches (report/stop/start): report warns
  and continues (display-only), stop/start die naming the value (they act on it).
- Rewrote the "unclear" test to go through the real call-site shape ($(detect_supervisor) then
  the same split), not a hand-constructed value, and tightened its final assertion to check for
  the actual detail text rather than $SYSTEMD_UNIT alone (the die() boilerplate names the unit
  either way, so that check could pass on a lost value).
2026-09-12 09:58:45 +07:00

432 lines
23 KiB
Bash
Executable File

#!/usr/bin/env bash
# Self-contained checks for the pure log classifier in redeploy-fleetd.sh.
set -euo pipefail
ROOT="$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)"
TMP="$(mktemp -d "$ROOT/.redeploy-log-test.XXXXXX")"
trap 'rm -rf "$TMP"' EXIT
# Sourcing stops before redeploy-fleetd.sh can build, stop, or start the daemon.
source "$ROOT/scripts/redeploy-fleetd.sh"
fail() {
printf 'FAIL: %s\n' "$*" >&2
return 1
}
assert_equals() {
local expected="$1" actual="$2" description="$3"
[ "$expected" = "$actual" ] || fail "$description: expected $expected, got $actual"
}
classify_fixture() {
local name="$1"
classify_amqp_connection_errors "$TMP/$name"
}
# fleetd #492 follow-up: detect_supervisor's stdout is now "kind<SEP>detail" (see the constraints
# comment above detect_supervisor in redeploy-fleetd.sh) — every test below that only cares about
# the kind must split it out with the SAME in-shell parameter expansion the real call site (:438)
# uses, never a bare string comparison against the raw output.
supervisor_kind_of() {
printf '%s' "${1%%"$SUPERVISOR_DETAIL_SEP"*}"
}
supervisor_detail_of() {
printf '%s' "${1#*"$SUPERVISOR_DETAIL_SEP"}"
}
# fleetd #492 — supervisor detection. Detect_supervisor() reads launchd_loaded/systemd_loaded, so
# each test overrides BOTH pairs (installed + loaded) explicitly, rather than relying on either
# being naturally absent: this machine may itself be running a real fleetd under launchd right now
# (see CLAUDE.md/MEMORY.md — launchd supervision has been live here since 2026-08-26), so leaving
# launchd_loaded unmocked in a "systemd only" test would silently read this host's own live state
# instead of the fixture.
test_detect_supervisor_launchd_only() {
launchd_installed() { return 0; }
launchd_loaded() { return 0; }
systemd_installed() { return 1; }
systemd_loaded() { return 1; }
assert_equals "launchd" "$(supervisor_kind_of "$(detect_supervisor)")" "launchd-only detection"
}
test_detect_supervisor_systemd_only() {
launchd_installed() { return 1; }
launchd_loaded() { return 1; }
systemd_installed() { return 0; }
systemd_loaded() { return 0; }
assert_equals "systemd" "$(supervisor_kind_of "$(detect_supervisor)")" "systemd-only detection"
}
test_detect_supervisor_none() {
launchd_installed() { return 1; }
launchd_loaded() { return 1; }
systemd_installed() { return 1; }
systemd_loaded() { return 1; }
assert_equals "none" "$(supervisor_kind_of "$(detect_supervisor)")" "unsupervised detection"
}
# fleetd #492 follow-up — detect_supervisor must never answer "none" when the truth is "could not
# tell". `systemd_installed`/`systemd_loaded` already know a unit file exists; this proves that
# fact is now actually consulted, not just printed as a warning: an installed-but-not-loaded unit
# reads as unclear, because is-active answers "no" for activating/deactivating/failed/pending
# auto-restart too, and every one of those is a host that IS under systemd.
test_detect_supervisor_systemd_installed_not_loaded_is_unclear() {
launchd_installed() { return 1; }
launchd_loaded() { return 1; }
systemd_installed() { return 0; } # the unit file IS there
systemd_loaded() { return 1; } # is-active says no — could be activating/failed/pending restart
local raw
raw="$(detect_supervisor)"
assert_equals "unclear" "$(supervisor_kind_of "$raw")" "systemd installed-but-not-loaded must read as unclear, not none"
printf '%s' "$(supervisor_detail_of "$raw")" | grep -qF "$SYSTEMD_UNIT" \
|| fail "detail does not name the systemd unit it found installed-but-not-loaded"
}
# Same fact, the launchd side: a plist on disk that is not currently loaded (unloaded without being
# removed, or about to be reloaded) must not read as "no supervisor" either.
test_detect_supervisor_launchd_installed_not_loaded_is_unclear() {
launchd_installed() { return 0; } # the plist IS there
launchd_loaded() { return 1; } # launchctl list says not loaded
systemd_installed() { return 1; }
systemd_loaded() { return 1; }
local raw
raw="$(detect_supervisor)"
assert_equals "unclear" "$(supervisor_kind_of "$raw")" "launchd installed-but-not-loaded must read as unclear, not none"
printf '%s' "$(supervisor_detail_of "$raw")" | grep -qF "$LAUNCHD_LABEL" \
|| fail "detail does not name the launchd label it found installed-but-not-loaded"
}
# Drives the REAL systemd_loaded/systemd_installed bodies (never stubbed) through a `systemctl`
# stub placed first on PATH that exits non-zero AND writes to stderr — the shape of a systemctl
# that runs but cannot reach the user bus (measured elsewhere as a headless ssh session with no
# lingering). This must read as unclear, never none: a probe that could not answer at all is not
# the same fact as "no supervisor is loaded".
test_detect_supervisor_systemd_probe_error_is_unclear() {
# Re-source first to restore the REAL launchd_*/systemd_* probe bodies. Earlier tests in this
# file permanently override them with stub `return 0`/`return 1` bodies (that is the whole point
# of those tests), and a bash function definition is global for the rest of the process — without
# this, systemd_loaded here would still be whatever the previous test left it as, never touching
# a real `systemctl` call at all.
source "$ROOT/scripts/redeploy-fleetd.sh"
local bin_dir result rc=0
bin_dir="$TMP/stub-bin-systemctl-errors"
mkdir -p "$bin_dir"
cat > "$bin_dir/systemctl" <<'STUB'
#!/usr/bin/env bash
echo "Failed to connect to bus: No such file or directory" >&2
exit 1
STUB
chmod +x "$bin_dir/systemctl"
PATH="$bin_dir:$PATH" systemd_loaded && rc=0 || rc=$?
[ "$rc" -ne 0 ] || fail "systemd_loaded must not report loaded=true when systemctl only errored"
assert_equals "1" "$SYSTEMD_LOADED_ERRORED" "systemd_loaded must flag a probe error, not a clean negative"
launchd_installed() { return 1; }
launchd_loaded() { return 1; }
result="$(PATH="$bin_dir:$PATH" detect_supervisor)"
assert_equals "unclear" "$(supervisor_kind_of "$result")" "a systemd probe error must read as unclear, not none"
printf '%s' "$(supervisor_detail_of "$result")" | grep -qF "$SYSTEMD_UNIT" \
|| fail "detail does not name the systemd unit whose probe errored"
}
# fleetd #492 follow-up (Item 1): this must go through the REAL call-site shape at :437-440, not a
# hand-constructed "unclear" value — a test that builds "unclear" directly proves the switch, not
# the handoff, and that is exactly the gap that let SUPERVISOR_UNCLEAR_DETAIL never reach the real
# caller in b17f37a. detect_supervisor runs as $(detect_supervisor): a subshell. Only stdout
# survives that boundary, so kind AND detail must both cross on it — this test proves they do.
test_require_drivable_supervisor_refuses_unclear() {
launchd_installed() { return 1; }
launchd_loaded() { return 1; }
systemd_installed() { return 0; }
systemd_loaded() { return 1; }
local SUPERVISOR_RAW SUPERVISOR_KIND SUPERVISOR_UNCLEAR_DETAIL output rc=0
# Exactly what :437-439 does — do not shortcut this by constructing "unclear" by hand.
SUPERVISOR_RAW="$(detect_supervisor)"
SUPERVISOR_KIND="${SUPERVISOR_RAW%%"$SUPERVISOR_DETAIL_SEP"*}"
SUPERVISOR_UNCLEAR_DETAIL="${SUPERVISOR_RAW#*"$SUPERVISOR_DETAIL_SEP"}"
assert_equals "unclear" "$SUPERVISOR_KIND" "setup: expected unclear before testing the refusal"
[ -n "$SUPERVISOR_UNCLEAR_DETAIL" ] \
|| fail "detail did not survive the \$(...) call-site boundary — SUPERVISOR_UNCLEAR_DETAIL is empty in the parent shell"
printf '%s' "$SUPERVISOR_UNCLEAR_DETAIL" | grep -qF "$SYSTEMD_UNIT" \
|| fail "detail that crossed the subshell boundary does not name the systemd unit it found installed-but-not-loaded"
output="$(require_drivable_supervisor "$SUPERVISOR_KIND" 2>&1)" || rc=$?
[ "$rc" -ne 0 ] || fail "require_drivable_supervisor accepted an unclear (undrivable) supervisor"
# Check for the ACTUAL DETAIL TEXT, not just "$SYSTEMD_UNIT" — the die() message's boilerplate
# recovery instructions name the unit unconditionally either way ("systemctl --user status
# $SYSTEMD_UNIT"), so a bare unit-name grep here would pass even on a lost/fallback detail. Only
# the specific detail string proves the crossed value, not the boilerplate, reached the message.
printf '%s' "$output" | grep -qF "$SUPERVISOR_UNCLEAR_DETAIL" \
|| fail "refusal message does not contain the specific detail that crossed the subshell boundary"
}
# The heart of the ticket: a supervisor this script cannot drive must refuse, never fall through to
# `kill`. require_drivable_supervisor die()s, so it is invoked inside a command substitution — that
# forks a subshell, so its exit() only ends the subshell and this test script keeps running under
# `set -e`.
test_require_drivable_supervisor_refuses_ambiguous() {
local output rc=0
output="$(require_drivable_supervisor "ambiguous" 2>&1)" || rc=$?
[ "$rc" -ne 0 ] || fail "require_drivable_supervisor accepted an ambiguous (undrivable) supervisor"
printf '%s' "$output" | grep -qF "$LAUNCHD_LABEL" \
|| fail "refusal message does not name the launchd label it found"
printf '%s' "$output" | grep -qF "$SYSTEMD_UNIT" \
|| fail "refusal message does not name the systemd unit it found"
}
test_require_drivable_supervisor_accepts_known_kinds() {
require_drivable_supervisor "launchd" || fail "refused a drivable launchd supervisor"
require_drivable_supervisor "systemd" || fail "refused a drivable systemd supervisor"
require_drivable_supervisor "none" || fail "refused the unsupervised case"
}
# fleetd #492 — the one-daemon check. Two live pids is the exact symptom a racing supervisor
# produces, and none of the other post-restart checks (healthz, jar id, the fresh log line) can see
# it because either daemon alone satisfies them.
test_count_daemon_pids() {
assert_equals 0 "$(count_daemon_pids "")" "count of an empty pid list"
assert_equals 1 "$(count_daemon_pids "4242")" "count of a single pid"
assert_equals 2 "$(count_daemon_pids "$(printf '4242\n4343\n')")" "count of two pids"
}
test_assert_single_daemon_accepts_one_pid() {
assert_single_daemon "4242" || fail "assert_single_daemon rejected a single running pid"
}
test_assert_single_daemon_rejects_two_pids() {
local output rc=0
output="$(assert_single_daemon "$(printf '4242\n4343\n')" 2>&1)" || rc=$?
[ "$rc" -ne 0 ] || fail "assert_single_daemon accepted two simultaneously running pids"
printf '%s' "$output" | grep -qF '4242' || fail "refusal message does not list the pids it found"
printf '%s' "$output" | grep -qF '4343' || fail "refusal message does not list the pids it found"
}
test_no_errors() {
cat > "$TMP/no-errors.log" <<'LOG'
2026-09-05 12:00:00 INFO fleetd listening
LOG
classify_fixture no-errors.log
assert_equals 0 "$REDEPLOY_ERROR_COUNT" "no-errors total"
assert_equals 0 "$REDEPLOY_UNEXPLAINED_ERRORS" "no-errors unexplained"
}
test_recovery_patterns_match_source() {
grep -F 'AMQP connection {}: {}' "$ROOT/fleetd/src/main/java/dev/ltms/fleet/msg/AmqpReplyInbox.java" > /dev/null \
|| fail "AMQP failure pattern no longer matches source"
grep -F 'AMQP connection recovered; cleared held replies for fresh redelivery' \
"$ROOT/fleetd/src/main/java/dev/ltms/fleet/msg/AmqpReplyInbox.java" > /dev/null \
|| fail "reply-inbox recovery pattern no longer matches source"
grep -F 'AMQP lead mailbox connection recovered; cleared held messages for fresh redelivery' \
"$ROOT/fleetd/src/main/java/dev/ltms/fleet/msg/LeadMailbox.java" > /dev/null \
|| fail "lead-mailbox recovery pattern no longer matches source"
}
test_attributed_recovered_connection_error() {
cat > "$TMP/attributed-recovered.log" <<'LOG'
2026-09-05 12:00:00 ERROR [AMQP Connection broker:5672] d.l.fleet.msg.AmqpReplyInbox - AMQP connection fleetd-reply-inbox: An unexpected connection driver error occurred
2026-09-05 12:00:01 INFO [AMQP Connection broker:5672] d.l.fleet.msg.AmqpReplyInbox - AMQP connection recovered; cleared held replies for fresh redelivery
LOG
classify_fixture attributed-recovered.log
assert_equals 1 "$REDEPLOY_ERROR_COUNT" "attributed-recovered total"
assert_equals 1 "$REDEPLOY_RECOVERED_AMQP_ERRORS" "attributed-recovered errors"
assert_equals 0 "$REDEPLOY_UNEXPLAINED_ERRORS" "attributed-recovered unexplained"
}
test_source_derived_error_shapes_recover_by_connection() {
# These ERROR shapes come from AmqpConnectionFailureLogger on main. They need a live-log check
# after redeploy because the new code has not yet written a production line.
cat > "$TMP/source-derived.log" <<'LOG'
17:37:53.537 ERROR [AMQP Connection broker:5672] d.l.fleet.msg.AmqpReplyInbox - AMQP connection fleetd-reply-inbox: An unexpected connection driver error occurred
17:37:54.537 ERROR [AMQP Connection broker:5672] d.l.fleet.msg.AmqpReplyInbox - AMQP connection fleetd-reply-inbox: Caught an exception during connection recovery!
17:37:55.537 ERROR [AMQP Connection broker:5672] d.l.fleet.msg.AmqpReplyInbox - AMQP connection fleetd-reply-inbox: An unexpected connection driver error occurred
17:37:56.537 ERROR [AMQP Connection broker:5672] d.ltms.fleet.msg.LeadMailbox - AMQP connection fleetd-lead-mailbox: An unexpected connection driver error occurred
17:37:57.537 ERROR [AMQP Connection broker:5672] d.ltms.fleet.msg.LeadMailbox - AMQP connection fleetd-lead-mailbox: Caught an exception during connection recovery!
17:37:58.537 ERROR [AMQP Connection broker:5672] d.ltms.fleet.msg.LeadMailbox - AMQP connection fleetd-lead-mailbox: An unexpected connection driver error occurred
17:38:00.000 INFO [AMQP Connection broker:5672] d.l.fleet.msg.AmqpReplyInbox - AMQP connection recovered; cleared held replies for fresh redelivery
17:38:01.000 INFO [AMQP Connection broker:5672] d.l.fleet.msg.AmqpReplyInbox - AMQP connection recovered; cleared held replies for fresh redelivery
17:38:02.000 INFO [AMQP Connection broker:5672] d.l.fleet.msg.AmqpReplyInbox - AMQP connection recovered; cleared held replies for fresh redelivery
17:38:03.000 INFO [AMQP Connection broker:5672] d.ltms.fleet.msg.LeadMailbox - AMQP lead mailbox connection recovered; cleared held messages for fresh redelivery
17:38:04.000 INFO [AMQP Connection broker:5672] d.ltms.fleet.msg.LeadMailbox - AMQP lead mailbox connection recovered; cleared held messages for fresh redelivery
17:38:05.000 INFO [AMQP Connection broker:5672] d.ltms.fleet.msg.LeadMailbox - AMQP lead mailbox connection recovered; cleared held messages for fresh redelivery
LOG
classify_fixture source-derived.log
assert_equals 6 "$REDEPLOY_ERROR_COUNT" "source-derived total"
assert_equals 6 "$REDEPLOY_RECOVERED_AMQP_ERRORS" "source-derived recovered"
assert_equals 0 "$REDEPLOY_UNEXPLAINED_ERRORS" "source-derived unexplained"
}
test_cross_connection_unattributable_errors_stay_loud() {
# This candidate has neither stable connection name, so LeadMailbox recovery must not consume it.
cat > "$TMP/cross-unattributable.log" <<'LOG'
2026-09-05 12:00:00 ERROR [AMQP Connection broker:5672] unknown - AMQP connection: An unexpected connection driver error occurred
2026-09-05 12:00:01 ERROR [AMQP Connection broker:5672] unknown - AMQP connection: An unexpected connection driver error occurred
2026-09-05 12:00:02 INFO [AMQP Connection 10.10.20.13:5672] d.ltms.fleet.msg.LeadMailbox - AMQP lead mailbox connection recovered; cleared held messages for fresh redelivery
2026-09-05 12:00:03 INFO [AMQP Connection 10.10.20.13:5672] d.ltms.fleet.msg.LeadMailbox - AMQP lead mailbox connection recovered; cleared held messages for fresh redelivery
LOG
classify_fixture cross-unattributable.log
assert_equals 2 "$REDEPLOY_ERROR_COUNT" "cross-unattributable total"
assert_equals 0 "$REDEPLOY_RECOVERED_AMQP_ERRORS" "cross-unattributable recovered"
assert_equals 2 "$REDEPLOY_UNEXPLAINED_ERRORS" "cross-unattributable unexplained"
}
test_attributed_cross_connection_errors_stay_loud() {
# LeadMailbox recovery cannot heal AmqpReplyInbox errors.
cat > "$TMP/cross-attributed.log" <<'LOG'
2026-09-05 12:00:00 ERROR AmqpReplyInbox - AMQP connection fleetd-reply-inbox: An unexpected connection driver error occurred
2026-09-05 12:00:01 ERROR AmqpReplyInbox - AMQP connection fleetd-reply-inbox: An unexpected connection driver error occurred
2026-09-05 12:00:02 INFO LeadMailbox - AMQP lead mailbox connection recovered; cleared held messages for fresh redelivery
2026-09-05 12:00:03 INFO LeadMailbox - AMQP lead mailbox connection recovered; cleared held messages for fresh redelivery
LOG
classify_fixture cross-attributed.log
assert_equals 2 "$REDEPLOY_ERROR_COUNT" "cross-attributed total"
assert_equals 0 "$REDEPLOY_RECOVERED_AMQP_ERRORS" "cross-attributed recovered"
assert_equals 2 "$REDEPLOY_UNEXPLAINED_ERRORS" "cross-attributed unexplained"
}
test_attributed_unrecovered_connection_error() {
cat > "$TMP/unrecovered.log" <<'LOG'
2026-09-05 12:00:00 ERROR AmqpReplyInbox - AMQP connection fleetd-reply-inbox: An unexpected connection driver error occurred
LOG
classify_fixture unrecovered.log
assert_equals 1 "$REDEPLOY_ERROR_COUNT" "unrecovered total"
assert_equals 0 "$REDEPLOY_RECOVERED_AMQP_ERRORS" "unrecovered AMQP errors"
assert_equals 1 "$REDEPLOY_UNEXPLAINED_ERRORS" "unrecovered unexplained"
}
test_other_error_is_unexplained() {
cat > "$TMP/other-error.log" <<'LOG'
2026-09-05 12:00:00 ERROR dev.ltms.fleet.Fleetd - startup failed
2026-09-05 12:00:01 INFO dev.ltms.fleet.msg.AmqpReplyInbox - AMQP connection recovered; cleared held replies for fresh redelivery
LOG
classify_fixture other-error.log
assert_equals 1 "$REDEPLOY_ERROR_COUNT" "other-error total"
assert_equals 1 "$REDEPLOY_UNEXPLAINED_ERRORS" "other-error unexplained"
}
test_recovery_requirement_mutation_is_caught() {
classify_amqp_connection_errors() {
local log_file="$1" line
REDEPLOY_ERROR_COUNT=0
REDEPLOY_RECOVERED_AMQP_ERRORS=0
REDEPLOY_UNEXPLAINED_ERRORS=0
while IFS= read -r line || [ -n "$line" ]; do
case "$line" in
*' ERROR '*|*' SEVERE '*)
REDEPLOY_ERROR_COUNT=$((REDEPLOY_ERROR_COUNT + 1))
case "$line" in
*'AMQP connection fleetd-reply-inbox: An unexpected connection driver error occurred'*)
REDEPLOY_RECOVERED_AMQP_ERRORS=$((REDEPLOY_RECOVERED_AMQP_ERRORS + 1))
;;
*) REDEPLOY_UNEXPLAINED_ERRORS=$((REDEPLOY_UNEXPLAINED_ERRORS + 1)) ;;
esac
;;
esac
done < "$log_file"
}
if test_attributed_unrecovered_connection_error > "$TMP/mutation-output" 2>&1; then
fail "mutation accepted an unrecovered connection error"
fi
grep -F 'FAIL: unrecovered AMQP errors: expected 0, got 1' "$TMP/mutation-output" > /dev/null \
|| fail "mutation failed without the expected assertion"
printf 'Recovery mutation: FAIL: unrecovered AMQP errors: expected 0, got 1\n'
}
test_shared_counter_mutation_is_caught() {
classify_amqp_connection_errors() {
local log_file="$1" line pending=0
REDEPLOY_ERROR_COUNT=0
REDEPLOY_RECOVERED_AMQP_ERRORS=0
REDEPLOY_UNEXPLAINED_ERRORS=0
while IFS= read -r line || [ -n "$line" ]; do
case "$line" in
*' ERROR '*|*' SEVERE '*)
REDEPLOY_ERROR_COUNT=$((REDEPLOY_ERROR_COUNT + 1))
case "$line" in
*'AMQP connection'*'An unexpected connection driver error occurred'*|*'AMQP connection'*'Caught an exception during connection recovery!'*)
case "$line" in
*'fleetd-reply-inbox'*|*'fleetd-lead-mailbox'*) pending=$((pending + 1)) ;;
*) REDEPLOY_UNEXPLAINED_ERRORS=$((REDEPLOY_UNEXPLAINED_ERRORS + 1)) ;;
esac
;;
*) REDEPLOY_UNEXPLAINED_ERRORS=$((REDEPLOY_UNEXPLAINED_ERRORS + 1)) ;;
esac
;;
*'AMQP connection recovered; cleared held replies for fresh redelivery'*|*'AMQP lead mailbox connection recovered; cleared held messages for fresh redelivery'*)
if [ "$pending" -gt 0 ]; then
pending=$((pending - 1))
REDEPLOY_RECOVERED_AMQP_ERRORS=$((REDEPLOY_RECOVERED_AMQP_ERRORS + 1))
fi
;;
esac
done < "$log_file"
REDEPLOY_UNEXPLAINED_ERRORS=$((REDEPLOY_UNEXPLAINED_ERRORS + pending))
}
if test_attributed_cross_connection_errors_stay_loud > "$TMP/shared-mutation-output" 2>&1; then
fail "shared counter mutation accepted cross-connection recovery"
fi
grep -F 'FAIL: cross-attributed recovered: expected 0, got 2' "$TMP/shared-mutation-output" > /dev/null \
|| fail "shared counter mutation failed without the expected assertion"
printf 'Shared-counter mutation: FAIL: cross-attributed recovered: expected 0, got 2\n'
}
test_unattributable_quiet_mutation_is_caught() {
classify_amqp_connection_errors() {
local log_file="$1" line
REDEPLOY_ERROR_COUNT=0
REDEPLOY_RECOVERED_AMQP_ERRORS=0
REDEPLOY_UNEXPLAINED_ERRORS=0
while IFS= read -r line || [ -n "$line" ]; do
case "$line" in
*' ERROR '*|*' SEVERE '*)
REDEPLOY_ERROR_COUNT=$((REDEPLOY_ERROR_COUNT + 1))
case "$line" in
*'AMQP connection'*'An unexpected connection driver error occurred'*|*'AMQP connection'*'Caught an exception during connection recovery!'*)
REDEPLOY_RECOVERED_AMQP_ERRORS=$((REDEPLOY_RECOVERED_AMQP_ERRORS + 1))
;;
*) REDEPLOY_UNEXPLAINED_ERRORS=$((REDEPLOY_UNEXPLAINED_ERRORS + 1)) ;;
esac
;;
esac
done < "$log_file"
}
if test_cross_connection_unattributable_errors_stay_loud > "$TMP/unattributable-mutation-output" 2>&1; then
fail "unattributable mutation accepted an unknown connection"
fi
grep -F 'FAIL: cross-unattributable recovered: expected 0, got 2' "$TMP/unattributable-mutation-output" > /dev/null \
|| fail "unattributable mutation failed without the expected assertion"
printf 'Unattributable mutation: FAIL: cross-unattributable recovered: expected 0, got 2\n'
}
test_detect_supervisor_launchd_only
test_detect_supervisor_systemd_only
test_detect_supervisor_none
test_detect_supervisor_systemd_installed_not_loaded_is_unclear
test_detect_supervisor_launchd_installed_not_loaded_is_unclear
test_detect_supervisor_systemd_probe_error_is_unclear
test_require_drivable_supervisor_refuses_ambiguous
test_require_drivable_supervisor_refuses_unclear
test_require_drivable_supervisor_accepts_known_kinds
test_count_daemon_pids
test_assert_single_daemon_accepts_one_pid
test_assert_single_daemon_rejects_two_pids
test_no_errors
test_recovery_patterns_match_source
test_attributed_recovered_connection_error
test_source_derived_error_shapes_recover_by_connection
test_cross_connection_unattributable_errors_stay_loud
test_attributed_cross_connection_errors_stay_loud
test_attributed_unrecovered_connection_error
test_other_error_is_unexplained
test_recovery_requirement_mutation_is_caught
test_shared_counter_mutation_is_caught
test_unattributable_quiet_mutation_is_caught
printf 'PASS: redeploy log classifier\n'