fix(watchdog): eliminate SIGPIPE+pipefail phantom failures in health checks
echo "$list" | grep -q under set -o pipefail exits 141 whenever grep matches before echo finishes writing, so healthy browsers were reported 'CDP up but no page target' and killed every 2 min fleet-wide (load 19+). Use [[ == *glob* ]] (no pipe, no race) for the page/muse.ai stages, and grep -c (reads to EOF, never early-exits) for the netns check.
This commit is contained in:
@@ -64,9 +64,9 @@ healthy() {
|
||||
local list
|
||||
list="$(cdp_list)" \
|
||||
|| { HEALTH_FAIL_REASON="CDP unreachable on :$CDP_PORT"; return 1; }
|
||||
echo "$list" | grep -q '"type": "page"' \
|
||||
[[ "$list" == *'"type": "page"'* ]] \
|
||||
|| { HEALTH_FAIL_REASON="CDP up but no page target in list"; return 1; }
|
||||
echo "$list" | grep -E -q '"url": "https://muse\.ai' \
|
||||
[[ "$list" == *'"url": "https://muse.ai'* ]] \
|
||||
|| { HEALTH_FAIL_REASON="CDP up but not on muse.ai"; return 1; }
|
||||
warp_egress_healthy || return 1
|
||||
return 0
|
||||
|
||||
+1
-1
@@ -9,5 +9,5 @@ NODE="${1:?usage: netvm-exec.sh <node> -- <cmd> [args...]}"; shift
|
||||
[ "${1:-}" = "--" ] && shift
|
||||
[ $# -gt 0 ] || { echo "usage: netvm-exec.sh <node> -- <cmd> [args...]"; exit 1; }
|
||||
[ -f "/etc/netvm/${NODE}.conf" ] || { echo "no warp identity for '$NODE' (human: netvm-new-identity.sh $NODE)"; exit 1; }
|
||||
ip netns list 2>/dev/null | grep -q "^warp-${NODE}" || { echo "node '$NODE' is not up (netvm-node-up.sh $NODE)"; exit 1; }
|
||||
[ "$(ip netns list 2>/dev/null | grep -c "^warp-${NODE}")" -ge 1 ] || { echo "node '$NODE' is not up (netvm-node-up.sh $NODE)"; exit 1; }
|
||||
exec sudo -n "$NETVM_BIN/netvm-enter.sh" "$NODE" "$(id -u)" "$(id -g)" "$HOME" -- "$@"
|
||||
|
||||
Reference in New Issue
Block a user