On 02/05/2019 23:06, Magnus Holmgren via Exim-dev wrote: > That code is from SA-Exim, which implements a local_scan() function. What I > mean is: is there something wrong with this code, or has something been > broken > in Exim so that the permanent pool doesn't work?
It looks right per the exim docs. However, the local_scan facilities are not something the exim testsuite exercises; quite possibly something has been broken. They're also not something I've ever explored. I doubt the permanent-pool is broken as a general thing; all sorts of odd things would be visible in general operation. There's a couple a memory-related debug facilities it might be possible for you to try: - a main-section option "debug_store", boolean. Set true to add checks on whether any memory being used by an active exim $variable is covered by a blob being freed. - run under the test-harness; freed memory is overwritten by a distinctive pattern. -- Cheers, Jeremy -- ## List details at https://lists.exim.org/mailman/listinfo/exim-dev Exim details at http://www.exim.org/ ##
