fix: ensure gateway restarts on install failure via ERR trap

[Zhe-SH-CN]

Description

When set -euo pipefail causes the install script to exit early (e.g. npm install fails, tarball extraction fails), the OpenClaw gateway was already stopped but never restarted, leaving the service down.

This PR adds an ERR trap after gateway stop in both install scripts (memos-local-plugin/install.sh and memos-local-openclaw/install.sh) that restarts the gateway on failure. The trap is cleared once the normal gateway start step succeeds.

Related Issue: Fixes #1391

Type of change

• Bug fix (non-breaking change which fixes an issue)

How Has This Been Tested

• bash -n syntax check passes for both scripts
• Manual code review confirming trap lifecycle (set after stop, cleared after start)

Checklist

• I have performed a self-review of my own code
• I have commented my code in hard-to-understand areas
• I have linked the issue to this PR

[Memtensor-AI]

🤖 Open Code Review

Target: PR #2152
Task: aaea6be5fdae958c
Base: dev-v2.0.24
Head: fix/install-trap-gateway-restart

🔍 OpenCodeReview found 2 issue(s) in this PR.

⚠️ 1 warning(s) occurred during review.

---

1. apps/memos-local-openclaw/install.sh (L323)

The ERR trap is set after gateway stop, but line 341 sets a separate EXIT trap (rm -rf "${TMP_PACK_DIR}") later. In bash, setting a new trap on a signal replaces any previously set trap for that signal — but more critically, trap - ERR at the end only clears the ERR trap; the EXIT trap is unaffected, which is correct. However, the ERR trap itself has a subtle issue: if the script is run with set -e (errexit), a failing command triggers ERR and then the shell exits. In that scenario the ERR trap fires gateway start, but the EXIT trap also fires afterward — potentially in an inconsistent order depending on shell behavior. More importantly, if gateway start itself fails inside the ERR trap (non-zero exit), the || true suppresses it, but the ERR trap re-firing is possible in some shells. Consider using a boolean flag pattern instead for more predictable behavior:

_gateway_needs_restart=1
_restart_gateway_on_exit() {
  [[ "$_gateway_needs_restart" == 1 ]] && "${OPENCLAW_BIN}" gateway start >/dev/null 2>&1 || true
}
trap '_restart_gateway_on_exit' EXIT
# ... at the normal success path:
_gateway_needs_restart=0

This avoids ERR-trap re-entrancy issues and integrates cleanly with the existing EXIT trap.

---

2. apps/memos-local-openclaw/install.sh (L412-L414)

gateway start is not guarded — if it fails (non-zero exit), the ERR trap fires and attempts another gateway start, then the script likely exits. The trap - ERR on the next line never executes in that case, leaving the trap technically irrelevant for that failure path. The start command should be checked explicitly before clearing the trap:

if ! "${OPENCLAW_BIN}" gateway start 2>&1; then
  error "Failed to start OpenClaw Gateway"
  exit 1
fi
trap - ERR

Generated by cloud-assistant via Open Code Review.

[Memtensor-AI]

⚠️ Automated Test Results: INCONCLUSIVE

Automated tests inconclusive (auto-generated test defect); treated as non-blocking. Manual review recommended. Details: Test failures are unrelated to the diff, which only modifies install.sh scripts. The failures are pre-existing test defects: one has an incorrect status assertion, and another references a non-existent 'role' column in the traces table.

Branch: fix/install-trap-gateway-restart