Prevent mesh firmware checks from interrupting node updates

Fix a new bug identified and introduced by commit: bae3640
And PR: #539

This bug appears to be a concurrent firmware management collision, essentially a race while flashing a node while checking it for updates from the primary at the same time.

I accidently ran into this, the seems to be that the node had prepared for the update, authenticated to its own WebUI, and was just about to hand the beta1 image to /upgrade.cgi

When at that exact time, I started an upgrade from the primary, which logged into the node to check for updates, triggered start_webs_update, and resulted in putting the node’s ASUS firmware update system into a competing update/check state and ultimately MerlinAU rebooted the node without flashing any firmware.

This may also be related to a report from JimbobJay here:
https://www.snbforums.com/threads/merlinau-v1-6-8-the-ultimate-firmware-auto-updater.96306/post-999444

Author: ExtremeFiretop

MerlinAU.sh

@@ -19,11 +19,11 @@
 set -u
 
 ## Set version for each Production Release ##
-readonly SCRIPT_VERSION=1.6.8
-readonly SCRIPT_VERSTAG="26090718"
+readonly SCRIPT_VERSION=1.6.9
+readonly SCRIPT_VERSTAG="26091601"
 readonly SCRIPT_NAME="MerlinAU"
 ## Set to "master" for Production Releases ##
-SCRIPT_BRANCH="master"
+SCRIPT_BRANCH="dev"
 
 ##----------------------------------------##
 ## Modified by Martinski W. [2024-Jul-03] ##
@@ -4335,7 +4335,7 @@ _ReEnableAsusTrendMicroProcesses_()
 }
 
 ##------------------------------------------##
-## Modified by ExtremeFiretop [2024-Jan-26] ##
+## Modified by ExtremeFiretop [2025-Sep-16] ##
 ##------------------------------------------##
 _DoCleanUp_()
 {
@@ -4353,6 +4353,10 @@ _DoCleanUp_()
    [ $# -gt 1 ] && [ "$2" -eq 1 ] && keepZIPfile=true
    [ $# -gt 2 ] && [ "$3" -eq 1 ] && keepWfile=true
 
+   # Clear the volatile F/W-update guard used by AiMesh primaries. #
+   # This value is intentionally never committed to NVRAM. #
+   nvram unset merlinau_fw_update 2>/dev/null
+
    # Stop the LEDs blinking #
    _Reset_LEDs_ 1
 
@@ -5429,9 +5433,9 @@ _DoMeshNodeLogin_()
     return "$?"
 }
 
-##----------------------------------------##
-## Modified by Martinski W. [2026-Jan-01] ##
-##----------------------------------------##
+##------------------------------------------##
+## Modified by ExtremeFiretop [2026-Sep-16] ##
+##------------------------------------------##
 # Trigger the node "Check for updates" (no waiting here)
 _MeshNodeTriggerFWCheck_()
 {
@@ -5443,6 +5447,7 @@ _MeshNodeTriggerFWCheck_()
     local safeID="$(_MeshSafeID_ "$nodeIPv4addr")"
     local nodeURL="$(_GetNodeURL_ "$nodeIPv4addr")"
     local cookieFile="/tmp/${runID}.${safeID}.cookie"
+    local nodeBusy nodeBusyRC
 
     # Check for Login Credentials #
     credsENC="$(Get_Custom_Setting credentials_base64)"
@@ -5463,6 +5468,30 @@ _MeshNodeTriggerFWCheck_()
         return 1
     fi
 
+    # Check if the AiMesh node is already performing a MerlinAU F/W update 
+    # before triggering the built-in firmware update check.
+    nodeBusy="$(curl -s -k "${nodeURL}/appGet.cgi?hook=nvram_get(merlinau_fw_update)" \
+    -H 'User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:109.0) Gecko/20100101 Firefox/115.0' \
+    -H 'Accept: application/json,text/plain,*/*' \
+    -H 'Accept-Language: en-US,en;q=0.5' \
+    -H 'Connection: keep-alive' \
+    -H "Referer: ${nodeURL}/index.asp" \
+    --cookie "$cookieFile" \
+    --max-time 2 2>/dev/null)"
+    nodeBusyRC="$?"
+
+    if [ "$nodeBusyRC" -eq 0 ] && echo "$nodeBusy" | grep -Eq '"merlinau_fw_update"[[:space:]]*:[[:space:]]*"1"'
+    then
+        Say "AiMesh Node [$nodeIPv4addr] entered an active MerlinAU F/W update before start_webs_update. Skipping firmware check."
+
+        # Best-effort logout of this primary-router session, then remove its cookie.
+        curl -s -k "${nodeURL}/Logout.asp" \
+        --cookie "$cookieFile" \
+        --max-time 2 >/dev/null 2>&1
+        rm -f "$cookieFile"
+        return 0
+    fi
+
     # Trigger firmware check (mimic WebUI "Check" button) #
     curl -s -k "${nodeURL}/start_apply.htm" \
     --referer "${nodeURL}/Advanced_FirmwareUpgrade_Content.asp" \
@@ -9755,11 +9784,15 @@ _Unmount_Eject_USB_Drives_()
     "$ejectUSB_OK" && return 0 || return 1
 }
 
-##----------------------------------------##
-## Modified by Martinski W. [2026-Jan-01] ##
-##----------------------------------------##
+##------------------------------------------##
+## Modified by ExtremeFiretop [2026-Sep-16] ##
+##------------------------------------------##
 _RunFirmwareUpdateNow_()
 {
+    local fwUploadResponseFile="/tmp/upload_response.txt"
+    local fwUploadDiagFile="${SETTINGS_DIR}/last_fw_upload_response.txt"
+    local curlRC=0  uploadHTTPcode=""
+
     # Double-check the directory exists before using it #
     [ ! -d "$FW_LOG_DIR" ] && mkdir -p -m 755 "$FW_LOG_DIR"
 
@@ -10219,6 +10252,15 @@ Please manually update to version ${GRNct}${MinSupportedFirmwareVers}${NOct} or
         fi
     fi
 
+    #------------------------------------------------------------------------#
+    # A volatile guard before restarting/logging into the WebGUI.
+    # Primary routers running MerlinAU can query this nvram value with appGet.cgi 
+    # and avoid triggering start_webs_update on this router mid-flash.
+    # Do not commit this value since a reboot should clear it automatically.
+    #------------------------------------------------------------------------#
+    nvram set merlinau_fw_update=1
+    rm -f "$fwUploadResponseFile" "$fwUploadDiagFile"
+
     #------------------------------------------------------------#
     # Restart the WebGUI to make sure nobody else is logged in
     # so that the F/W Update can start without interruptions.
@@ -10311,7 +10353,8 @@ Please manually update to version ${GRNct}${MinSupportedFirmwareVers}${NOct} or
         Say "Flashing ${GRNct}${firmware_file}${NOct}...\n${REDct}Please wait for reboot in about 4 minutes or less.${NOct}"
         echo
 
-        # *WARNING*: NO MORE logging at this point & beyond #
+        # Avoid persistent logging from this point during the normal flash path. #
+        # Failure diagnostics are written only if the router does not reboot. #
         sync ; sleep 2 ; echo 3 > /proc/sys/vm/drop_caches ; sleep 3
 
         ##-------------------------------------##
@@ -10327,7 +10370,7 @@ Please manually update to version ${GRNct}${MinSupportedFirmwareVers}${NOct} or
         # the following 'Curl' command MUST always be the last step in this block.
         # Do NOT insert any commands after it! (unless you understand the implications).
         #----------------------------------------------------------------------------------#
-        nohup curl -k "${routerURL}/upgrade.cgi" \
+        nohup curl -sS -k "${routerURL}/upgrade.cgi" \
         --referer "${routerURL}/Advanced_FirmwareUpgrade_Content.asp" \
         --user-agent 'Mozilla/5.0 (X11; Linux x86_64; rv:109.0) Gecko/20100101 Firefox/115.0' \
         -H 'Accept-Language: en-US,en;q=0.5' \
@@ -10340,7 +10383,9 @@ Please manually update to version ${GRNct}${MinSupportedFirmwareVers}${NOct} or
         -F 'preferred_lang=EN' \
         -F "firmver=${dottedVersion}" \
         -F "file=@${firmware_file}" \
-        --cookie "$cookieFile" > /tmp/upload_response.txt 2>&1 &
+        --cookie "$cookieFile" \
+        --write-out '\nMERLINAU_HTTP_CODE:%{http_code}\n' \
+        > "$fwUploadResponseFile" 2>&1 &
         curlPID=$!
 
         #----------------------------------------------------------#
@@ -10356,16 +10401,40 @@ Please manually update to version ${GRNct}${MinSupportedFirmwareVers}${NOct} or
            sleep 180
            if [ "$curlPID" -gt 0 ]
            then
-               kill -EXIT $curlPID 2>/dev/null || return
-               kill -TERM $curlPID 2>/dev/null
+               kill -EXIT "$curlPID" 2>/dev/null || return
+               kill -TERM "$curlPID" 2>/dev/null
            fi
         ) &
-        wait $curlPID ; curlPID=0
+
+        # Preserve Curl's actual result instead of discarding it. #
+        wait "$curlPID"
+        curlRC=$?
+        curlPID=0
+        uploadHTTPcode="$(sed -n 's/^MERLINAU_HTTP_CODE://p' "$fwUploadResponseFile" 2>/dev/null | tail -n 1)"
+
         #----------------------------------------------------------#
         # Let's wait for 3 minutes here. If the router does not
-        # reboot by itself after the process returns, do it now.
+        # reboot by itself after the process returns, 
+        # preserve any diagnostics then reboot.
+        # A successful flash reboots before this step.
         #----------------------------------------------------------#
         sleep 180
+
+        {
+            echo "MerlinAU v$SCRIPT_VERSION firmware upload diagnostics"
+            echo "Timestamp: $(date '+%Y-%m-%d %H:%M:%S %Z')"
+            echo "Router: $MODEL_ID"
+            echo "Firmware image: $firmware_file"
+            echo "Curl exit code: $curlRC"
+            echo "HTTP status: ${uploadHTTPcode:-UNKNOWN}"
+            echo "------------------------------------------------------------"
+            [ -s "$fwUploadResponseFile" ] && cat "$fwUploadResponseFile"
+        } > "$fwUploadDiagFile" 2>/dev/null
+        chmod 600 "$fwUploadDiagFile" 2>/dev/null
+
+        _MsgToSysLog_ "F/W upload did not cause the router to reboot within 180 seconds. Curl exit code [$curlRC], HTTP status [${uploadHTTPcode:-UNKNOWN}]."
+        _MsgToSysLog_ "F/W upload diagnostics saved to [$fwUploadDiagFile]."
+
         _ReleaseLock_
         /sbin/service reboot
     else