I need to be able to collect core dumps, but since Asterisk's CWD is / (so claims procfs), it'll never be able to write them. I've confirmed my suspicions by allowing world write to / and SIGABRT the process, lo and behold, I had a core.
I can obviously use core_pattern to override system-wide, but I rather like the default behavior of dumping core to the cwd.
I am using Asterisk 220.127.116.11 and a MySQL database connected via ODBC to store CDRs. When my MySQL database isn't available when Asterisk starts or has an outage while Asterisk is running, I would expect Asterisk to retry to connect to the database, but this doesn't happen!