Classify named AMQP recovery errors
This commit is contained in:
Dai Ha
+11
-11
@@ -116,11 +116,9 @@ check_log_path_matches_plist() {
|
||||
ok "log path check: script and plist agree ($resolved_out)"
|
||||
}
|
||||
|
||||
# Classify ERROR lines in one fresh log region. The logging layout abbreviates logger packages, so
|
||||
# match simple class names and the handler's actual ERROR messages. An error can be quiet only when
|
||||
# its own line names AmqpReplyInbox or LeadMailbox; the handler lines now seen in fleetd.out name
|
||||
# neither connection, so they deliberately stay unexplained. This avoids letting one connection's
|
||||
# recovery hide a failure in the other connection.
|
||||
# Classify ERROR lines in one fresh log region. AMQP failure messages now include the connection
|
||||
# name, so a recovery can clear only errors for its own connection. A candidate with neither name
|
||||
# remains unexplained: it must never be quieted by a recovery on the other connection.
|
||||
classify_amqp_connection_errors() {
|
||||
local log_file="$1" line pending_inbox=0 pending_lead_mailbox=0
|
||||
REDEPLOY_ERROR_COUNT=0
|
||||
@@ -132,12 +130,14 @@ classify_amqp_connection_errors() {
|
||||
*' ERROR '*|*' SEVERE '*)
|
||||
REDEPLOY_ERROR_COUNT=$((REDEPLOY_ERROR_COUNT + 1))
|
||||
case "$line" in
|
||||
*'ForgivingExceptionHandler'*'An unexpected connection driver error occurred'*|*'ForgivingExceptionHandler'*'Caught an exception during connection recovery!'*)
|
||||
case "$line" in
|
||||
*'AmqpReplyInbox'*) pending_inbox=$((pending_inbox + 1)) ;;
|
||||
*'LeadMailbox'*) pending_lead_mailbox=$((pending_lead_mailbox + 1)) ;;
|
||||
*) REDEPLOY_UNEXPLAINED_ERRORS=$((REDEPLOY_UNEXPLAINED_ERRORS + 1)) ;;
|
||||
esac
|
||||
*'AMQP connection fleetd-reply-inbox: An unexpected connection driver error occurred'*|*'AMQP connection fleetd-reply-inbox: Caught an exception during connection recovery!'*)
|
||||
pending_inbox=$((pending_inbox + 1))
|
||||
;;
|
||||
*'AMQP connection fleetd-lead-mailbox: An unexpected connection driver error occurred'*|*'AMQP connection fleetd-lead-mailbox: Caught an exception during connection recovery!'*)
|
||||
pending_lead_mailbox=$((pending_lead_mailbox + 1))
|
||||
;;
|
||||
*'AMQP connection'*'An unexpected connection driver error occurred'*|*'AMQP connection'*'Caught an exception during connection recovery!'*)
|
||||
REDEPLOY_UNEXPLAINED_ERRORS=$((REDEPLOY_UNEXPLAINED_ERRORS + 1))
|
||||
;;
|
||||
*) REDEPLOY_UNEXPLAINED_ERRORS=$((REDEPLOY_UNEXPLAINED_ERRORS + 1)) ;;
|
||||
esac
|
||||
|
||||
@@ -35,6 +35,8 @@ LOG
|
||||
}
|
||||
|
||||
test_recovery_patterns_match_source() {
|
||||
grep -F 'AMQP connection {}: {}' "$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' \
|
||||
"$ROOT/fleetd/src/main/java/dev/ltms/fleet/msg/AmqpReplyInbox.java" > /dev/null \
|
||||
|| fail "reply-inbox recovery pattern no longer matches source"
|
||||
@@ -45,8 +47,8 @@ test_recovery_patterns_match_source() {
|
||||
|
||||
test_attributed_recovered_connection_error() {
|
||||
cat > "$TMP/attributed-recovered.log" <<'LOG'
|
||||
2026-09-05 12:00:00 ERROR AmqpReplyInbox ForgivingExceptionHandler - An unexpected connection driver error occurred
|
||||
2026-09-05 12:00:01 INFO AmqpReplyInbox - AMQP connection recovered; cleared held replies for fresh redelivery
|
||||
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 "$REDEPLOY_ERROR_COUNT" "attributed-recovered total"
|
||||
@@ -54,31 +56,34 @@ LOG
|
||||
assert_equals 0 "$REDEPLOY_UNEXPLAINED_ERRORS" "attributed-recovered unexplained"
|
||||
}
|
||||
|
||||
test_real_error_shape_is_loud_when_unattributable() {
|
||||
# The real handler ERROR lines name neither inbox nor lead mailbox. Even though both connections
|
||||
# later recover, this region must stay loud because the recovery cannot be assigned safely.
|
||||
cat > "$TMP/real-error-shape.log" <<'LOG'
|
||||
17:37:53.537 ERROR [AMQP Connection 10.10.20.13:5672] c.r.c.i.ForgivingExceptionHandler - An unexpected connection driver error occurred
|
||||
18:20:33.027 ERROR [AMQP Connection 10.10.20.13:5672] c.r.c.i.ForgivingExceptionHandler - Caught an exception during connection recovery!
|
||||
18:30:00.000 ERROR [AMQP Connection 10.10.20.13:5672] c.r.c.i.ForgivingExceptionHandler - An unexpected connection driver error occurred
|
||||
19:00:00.000 ERROR [AMQP Connection 10.10.20.13:5672] c.r.c.i.ForgivingExceptionHandler - Caught an exception during connection recovery!
|
||||
20:00:00.000 ERROR [AMQP Connection 10.10.20.13:5672] c.r.c.i.ForgivingExceptionHandler - An unexpected connection driver error occurred
|
||||
21:00:00.000 ERROR [AMQP Connection 10.10.20.13:5672] c.r.c.i.ForgivingExceptionHandler - Caught an exception during connection recovery!
|
||||
21:41:39.329 INFO [AMQP Connection 10.10.20.13:5672] d.ltms.fleet.msg.LeadMailbox - AMQP lead mailbox connection recovered; cleared held messages for fresh redelivery
|
||||
21:41:48.693 INFO [AMQP Connection 10.10.20.13:5672] d.l.fleet.msg.AmqpReplyInbox - AMQP connection recovered; cleared held replies for fresh redelivery
|
||||
test_source_derived_error_shapes_recover_by_connection() {
|
||||
# These ERROR shapes come from AmqpConnectionFailureLogger on main. They need a live-log check
|
||||
# after redeploy because the new code has not yet written a production line.
|
||||
cat > "$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 real-error-shape.log
|
||||
assert_equals 6 "$REDEPLOY_ERROR_COUNT" "real-error-shape total"
|
||||
assert_equals 0 "$REDEPLOY_RECOVERED_AMQP_ERRORS" "real-error-shape recovered"
|
||||
assert_equals 6 "$REDEPLOY_UNEXPLAINED_ERRORS" "real-error-shape unexplained"
|
||||
classify_fixture source-derived.log
|
||||
assert_equals 6 "$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 is the reported unsafe shape. Both handler lines lack a connection kind, so LeadMailbox
|
||||
# recoveries must not consume either one.
|
||||
# This candidate has neither stable connection name, so LeadMailbox recovery must not consume it.
|
||||
cat > "$TMP/cross-unattributable.log" <<'LOG'
|
||||
2026-09-05 12:00:00 ERROR [AMQP Connection 10.10.20.13:5672] c.r.c.i.ForgivingExceptionHandler - An unexpected connection driver error occurred
|
||||
2026-09-05 12:00:01 ERROR [AMQP Connection 10.10.20.13:5672] c.r.c.i.ForgivingExceptionHandler - An unexpected connection driver error occurred
|
||||
2026-09-05 12:00:00 ERROR [AMQP Connection broker:5672] unknown - AMQP connection: An unexpected connection driver error occurred
|
||||
2026-09-05 12:00:01 ERROR [AMQP Connection broker:5672] unknown - AMQP connection: An unexpected connection driver error occurred
|
||||
2026-09-05 12:00:02 INFO [AMQP Connection 10.10.20.13:5672] d.ltms.fleet.msg.LeadMailbox - AMQP lead mailbox connection recovered; cleared held messages for fresh redelivery
|
||||
2026-09-05 12:00:03 INFO [AMQP Connection 10.10.20.13:5672] d.ltms.fleet.msg.LeadMailbox - AMQP lead mailbox connection recovered; cleared held messages for fresh redelivery
|
||||
LOG
|
||||
@@ -89,11 +94,10 @@ LOG
|
||||
}
|
||||
|
||||
test_attributed_cross_connection_errors_stay_loud() {
|
||||
# This fixture exercises the per-kind pending state should a future layout include a simple class
|
||||
# name on the handler line. LeadMailbox recovery cannot heal AmqpReplyInbox errors.
|
||||
# LeadMailbox recovery cannot heal AmqpReplyInbox errors.
|
||||
cat > "$TMP/cross-attributed.log" <<'LOG'
|
||||
2026-09-05 12:00:00 ERROR AmqpReplyInbox ForgivingExceptionHandler - An unexpected connection driver error occurred
|
||||
2026-09-05 12:00:01 ERROR AmqpReplyInbox ForgivingExceptionHandler - An unexpected connection driver error occurred
|
||||
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
|
||||
@@ -105,7 +109,7 @@ LOG
|
||||
|
||||
test_attributed_unrecovered_connection_error() {
|
||||
cat > "$TMP/unrecovered.log" <<'LOG'
|
||||
2026-09-05 12:00:00 ERROR AmqpReplyInbox ForgivingExceptionHandler - An unexpected connection driver error occurred
|
||||
2026-09-05 12:00:00 ERROR AmqpReplyInbox - AMQP connection fleetd-reply-inbox: An unexpected connection driver error occurred
|
||||
LOG
|
||||
classify_fixture unrecovered.log
|
||||
assert_equals 1 "$REDEPLOY_ERROR_COUNT" "unrecovered total"
|
||||
@@ -134,7 +138,7 @@ test_recovery_requirement_mutation_is_caught() {
|
||||
*' ERROR '*|*' SEVERE '*)
|
||||
REDEPLOY_ERROR_COUNT=$((REDEPLOY_ERROR_COUNT + 1))
|
||||
case "$line" in
|
||||
*'ForgivingExceptionHandler'*'An unexpected connection driver error occurred'*)
|
||||
*'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)) ;;
|
||||
@@ -163,9 +167,9 @@ test_shared_counter_mutation_is_caught() {
|
||||
*' ERROR '*|*' SEVERE '*)
|
||||
REDEPLOY_ERROR_COUNT=$((REDEPLOY_ERROR_COUNT + 1))
|
||||
case "$line" in
|
||||
*'ForgivingExceptionHandler'*'An unexpected connection driver error occurred'*|*'ForgivingExceptionHandler'*'Caught an exception during connection recovery!'*)
|
||||
*'AMQP connection'*'An unexpected connection driver error occurred'*|*'AMQP connection'*'Caught an exception during connection recovery!'*)
|
||||
case "$line" in
|
||||
*'AmqpReplyInbox'*|*'LeadMailbox'*) pending=$((pending + 1)) ;;
|
||||
*'fleetd-reply-inbox'*|*'fleetd-lead-mailbox'*) pending=$((pending + 1)) ;;
|
||||
*) REDEPLOY_UNEXPLAINED_ERRORS=$((REDEPLOY_UNEXPLAINED_ERRORS + 1)) ;;
|
||||
esac
|
||||
;;
|
||||
@@ -191,14 +195,44 @@ test_shared_counter_mutation_is_caught() {
|
||||
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_no_errors
|
||||
test_recovery_patterns_match_source
|
||||
test_attributed_recovered_connection_error
|
||||
test_real_error_shape_is_loud_when_unattributable
|
||||
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'
|
||||
|
||||
Reference in New Issue
Block a user