confd: Use system timeout value when loading startup/failure

Now that we have a unified timeout value that all front-ends should
honor, make sure to do so when loading startup-config and
failure-config during boot.

This fixes an issue where the system would fallback to failure-config
even though a legitimate startup-config was available, but took more
than 10s to apply.

doc/ChangeLog.md

@@ -14,7 +14,9 @@ All notable changes to the project are documented in this file.
   authorized users, like `admin`, to not being able to query status
   of OSPF or BFD.  Workaround by using the UNIX shell `sudo vtysh`.
   Regression introduced in v24.08.0
-
+- The timeout before giving up on loading the `startup-config` at boot
+  is now 1 minute, just like operations via other front-ends (NETCONF
+  and RESTCONF). This was previously (incorrectly) set to 10 seconds.
 
 [v24.08.0][] - 2024-08-30
 -------------------------

package/confd/confd.conf

@@ -12,13 +12,13 @@ service name:confd log:prio:daemon.err <run/bootstrap/success> \
 	-- Configuration daemon
 
 # Bootstrap system with startup-config
-run name:startup log:prio:user.notice norestart <pid/confd> \
-	[S] /usr/libexec/confd/load startup-config \
+run name:startup log:prio:user.notice norestart <pid/confd> env:/etc/default/confd \
+	[S] /usr/libexec/confd/load -t $CONFD_TIMEOUT startup-config \
 	-- Loading startup-config
 
 # Run if loading startup-config fails for some reason
-run name:failure log:prio:user.critical norestart <pid/confd> if:<run/startup/failure> \
-	[S] /usr/libexec/confd/load failure-config \
+run name:failure log:prio:user.critical norestart <pid/confd> env:/etc/default/confd if:<run/startup/failure> \
+	[S] /usr/libexec/confd/load -t $CONFD_TIMEOUT failure-config \
 	-- Loading failure-config
 
 run name:error :2 log:console norestart if:<run/failure/failure> \