From 1f5a1368adbad9d3bbfb0b5ea39401e01d19e914 Mon Sep 17 00:00:00 2001
From: Gorakh Nath Yadav <gnyadav@openobserve.ai>
Date: Thu, 30 Jul 2026 16:58:19 +0530
Subject: [PATCH] fix(installers): correct exit status on macOS bash 3.2
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

On macOS /bin/bash (3.2), the exit status of the last command in an EXIT
trap becomes the script's exit status. cleanup_on_exit ends with a
'[ -f ... ] && rm' test that legitimately fails when TEMP_FILES is empty,
so every installer exited 1 even on success — breaking any set -e wrapper
or CI that checks the curl|bash exit code. Linux bash 4/5 preserves the
original status, which is why this went unnoticed.

Capture $? on trap entry and re-exit with it, preserving both success (0)
and real failure codes. Verified on macOS bash 3.2 across the codex,
claude-code, and github-copilot installers.
---
 ai/lib/common.sh | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/ai/lib/common.sh b/ai/lib/common.sh
index 2464625..d8c0707 100644
--- a/ai/lib/common.sh
+++ b/ai/lib/common.sh
@@ -127,6 +127,11 @@ cleanup_on_error() {
 # but install_error_trap below replaces the EXIT trap — so we re-do the
 # bootstrap cleanup here. macOS mktemp writes to /var/folders/, Linux to /tmp/.
 cleanup_on_exit() {
+    # Capture the script's real exit status and re-exit with it below. On
+    # bash 3.2 (macOS /bin/bash) the status of the LAST command in an EXIT
+    # trap silently becomes the script's exit status — without this, the
+    # trailing `[ -f ... ]` test made every installer exit 1 even on success.
+    local exit_code=$?
     if [ -n "${COMMON_SH:-}" ] && [ -f "$COMMON_SH" ]; then
         case "$COMMON_SH" in
             /tmp/*|/var/folders/*) rm -f "$COMMON_SH" ;;
@@ -135,6 +140,7 @@ cleanup_on_exit() {
     for f in "${TEMP_FILES[@]:-}"; do
         [ -f "$f" ] && rm -f "$f"
     done
+    exit "$exit_code"
 }
 
 # Caller installs the trap via:  install_error_trap